Note: Descriptions are shown in the official language in which they were submitted.
Artificially Intelligent Renewable Energy Planning Using Geographic
Information System
(GIS) Data
FIELD OF THE INVENTION
The present invention relates generally to geographic information systems
(GIS), and more
particularly to a computer implemented planning tool that interfaces with one
or more
geographic information systems to access geographic and environmental data
therefrom, and
apply Artificial Intelligence (Al) methods thereto to make planning
recommendations for
renewable energy projects.
BACKGROUND OF THE INVENTION
In recent years, due to increased awareness, most individuals consider climate
change as an
emergency. As the result of public opinion impacts on the issue of climate
change, many
governments at federal, state and local levels implemented policies that
address climate change.
One of the largest sources of greenhouse gas emissions from human activities
is from burning
fossil fuels for electricity. As a result, there is strong support for
promoting renewable sources
such as solar power and wind power. Nearly two-thirds of all new power
generation capacity
added in 2018 was from renewables and a third of global power capacity is now
based on
renewable energy.
In addition to climate change mitigation, renewable energy recourses can
provide economic
benefits and energy security if are implemented intelligently. There is an
increased interest from
private sector which their interests evolved from a strictly environmental
concern into a "strategic
concern driven by market forces". There are also many laws and energy policies
to encourage
public-private partnerships to leverage private capital and expertise to
support the development
of renewable energy projects.
There are massive flows of capital directed toward development of renewable
generation assets
and the energy sector is transforming very rapidly. One of the main challenges
that individual,
communities, corporations and developers face is selection of the best
combination of technology
types and sizes of the renewable assets to maximize the economic value and
increase energy
security.
Nowadays, consultation with an expert is necessary in order to identify
suitable locations for
establishing renewable energy power plants, and once an expert guided
selection of a suitable
location is made, then various other data is inputted to software applications
such as RETScreen
and COMFAR, which are useful in preparation of feasibility studies.
Applicant has realized that there exists a wealth of available data resources
that can be exploited
to massively simplify planning of renewable energy projects and ongoing
management of the
planned facilities once establish, using uniquely and inventively automated
systems and
techniques.
1
Date Recue/Date Received 2023-03-08
SUMMARY OF THE INVENTION
According to a first aspect of the invention, there is provided a computer-
implemented method
for at least partially automated planning of a renewable energy power plant
project, said method
comprising the following steps executed by one or more computer processors
embodied in one
or more computers operably connected to a communications network, said method
comprising:
(a) storing in memory, for each of plurality of different power plant types, a
different respective
criteria-weighting scheme for use in automated evaluation and selection of an
optimal
geographic location for said renewable energy power plant project;
(b) collecting user input data from a user;
(c) based at least partially on said user input, identifying one or more
planning constraints for
said planning of said renewable energy power plant project;
(d) accessing one or more geographic and environmental databases, at least one
of which is
embodied in one or more geographic information systems, and in which there is
stored both
geographic information and environmental information, of which said
environmental
information includes environmental data relevant to at least two types of
renewal energy
production that respectively correspond to two of said plurality of different
predetermined
power plant types, and for each of a plurality of potential candidate
geographic locations for
said renewable energy power plant project, retrieving from said geographic
information and
said environmental information a respective dataset containing a plurality of
performance
values for a plurality of geographic and environmental parameters, including
at least one
environmental parameter whose performance value is based on said environmental
data;
(e) for at least one of said plurality of different power plant types,
selecting the respective
criteria-weighting scheme corresponding thereto, and applying said selected
respective
criteria-weighting scheme against the respective datasets of the candidate
geographic
locations in a computer-executed multi-criteria decision-making (MCDM) process
in which at
least a subset of said plurality of parameters are used as criteria of said
MCDM process;
(f) based on results of said MCDM process, identifying the optimal geographic
location for the
renewable energy power generation plant from among said candidate geographic
locations;
and
(g) Communicating identification of said optimal geographic location to the
user.
Preferably, said predetermined power plant types include any two or more of
solar, wind,
hydroelectric and biomass power plants.
In one instance, step (b) includes receiving user-identification of an
intended power plant type
from among different user-selectable power plant options presented to the
user, each
corresponding to a different one of said predetermined power plant types, and
said at least one
of said plurality of different power plant types in step (e) consists of said
intended power plant
type.
In an alternative instance:
2
Date Recue/Date Received 2023-03-08
step (e) comprises applying the respective criteria-weighting schemes of the
plurality of
different power plant types in said computer-executed MCDM process, and
thereby identifying
respective best candidate locations for said plurality of different power
plant types;
step (f) comprises comparing evaluation results of those respective best
candidate locations
against one another, and selecting a best scoring one of said best candidate
locations as the
optimal geographic location; and step (g) comprises also communicating, from
among said
plurality of different power plant types, identification of a recommended
power plant type to
which said best scoring one of the best candidate locations corresponds.
Preferably said computer-executed selection of the intended power plant type
is based at least
partly on said one or more other planning constraints, and said one or more
other planning
constraints comprise a budgetary constraint designated in said user input
data.
Preferably the executed steps further include an equipment assessment step
comprising:
Using the one or more constraints, identifying equipment requirements for the
power plant
project; and searching one or more equipment supplier databases, and
identifying therefrom
candidate equipment options fulfilling said equipment requirements.
Preferably the equipment assessment step further comprises:
Assessing said candidate equipment options against one another to identify
optimal equipment
options; and determining a total cost of the optimal equipment options.
According to a second aspect of the invention, there are provided one or more
non-transitory
computer readable media having stored thereon executable statements and
instructions for
execution by one or more processors, said statements and instructions being
configured to, when
executed, perform a method according to the first aspect of the invention.
According to a third aspect of the invention, there is provided a system for
at least partially
automated planning of a renewable energy power plant project, said system
comprising one or
more computer processors embodied in one or more computers operably connected
to a
communications network by which said one or more computers are communicable
with one or
more geographical and environmental databases, at least one of which is
embodied in one or
more geographic information systems, and one or more non-transitory computer
readable media
according to the second embodiment of the invention, embodied in or connected
to said one or
more computers for execution of the statement and instructions on said one or
more non-
transitory computer readable media by said one or more processors of said one
or more
computers.
According to a fourth aspect of the invention, there is provided a computer-
implemented method
for finding optimal time periods for saving produced power in storage
batteries, selling said
produced power to a power network or consuming said power by using statistical
data and a
modified value iteration algorithm, said method comprising:
a. based on input data, retrieve identification of a type of renewable energy
concerned and an
associated capacity, power demand, produced power, and type and specifications
of one or
more storage batteries to be used for power storage;
3
Date Recue/Date Received 2023-03-08
b. retrieve statistical data from a database, and solve an optimal dispatch
problem using a
reinforcement learning approach and a dynamic programming algorithm; and
c. in solving said optimal dispatch problem, using a value iteration
algorithm to find an optimum
answer by repeating possible answers that converges the problem to an optimum
solution.
In one embodiment, the method further comprises further comprising issuing
command signals
to one or more control devices of a power plant to switch said control device,
according to the
optimal time periods, between:
a power storage state dispatching the produced power to the storage batteries;
a power consumption state dispatching the produced power to electrical loads;
and
a power selling state dispatching the produced power to a power network for
financial
compensation.
In competitive power markets, selection of the time for selling energy is very
significant. To have
the highest revenue from produce electricity, the operating condition of the
power plant can be
optimized. In the case of an electricity power system, the total load on the
system will generally
be higher during the daytime and lower during the late evening, when most
population is asleep.
An effective power management tool is disclosed to help users find the best
way to consume or
sell energy.
According to a fifth aspect of the invention, there are provided one or more
non-transitory
computer readable media having stored thereon executable statements and
instructions for
execution by one or more processors, said statements and instructions being
configured to, when
executed, perform the method according to the fourth aspect of the invention.
According to a sixth aspect of the invention, there is provided a system for
finding optimal time
periods for saving produced power in storage batteries, selling said produced
power to a power
network or consuming said power, said system comprising one or more computer
processors
embodied in one or more computers, and one or more non-transitory computer
readable media
according to the fifth aspect of the invention, embodied in or connected to
said one or more
computers for execution of the statement and instructions on said one or more
non-transitory
computer readable media by said one or more processors of said one or more
computers.
According to a seventh aspect of the invention, there is provided a computer-
implemented
method for finding an optimal geographic location for constructing an electric
vehicle charging
station that relies on renewable energy, said method comprising the following
steps executed by
one or more computer processors embodied in one or more computers operably
connected to a
communications network:
(a) accessing one or more geographic and environmental databases, at least one
of which is
embodied in one or more geographic information systems, and in which there is
stored both
geographic information and environmental information, of which said
environmental
information includes environmental data relevant to at least one type of
renewable energy
production, and for each of a plurality of potential candidate geographic
locations for said
electric vehicle charging station, retrieving from said geographic and
environmental
4
Date Recue/Date Received 2023-03-08
information a respective dataset containing a plurality of performance values
for a plurality
of geographic and environmental parameters, including at least one
environmental
parameter whose performance value is based on said environmental data;
(b) applying a criterion weighting scheme against the respective datasets of
the candidate
geographic locations in a computer-executed multi-criteria decision-making
(MCDM) process
in which at least a subset of said plurality of parameters are used as
criteria of said MCDM
process;
(c) based on results of said MCDM process, identifying the optimal geographic
location for the
electric vehicle charging station from among said candidate geographic
locations; and
(d) communicating identification of said optimal geographic location to a
user.
According to an eighth aspect of the invention, there are provided one or more
non-transitory
computer readable media having stored thereon executable statements and
instructions for
execution by one or more processors, said statements and instructions being
configured to, when
executed, perform a method according to the seventh aspect of the invention.
According to a ninth aspect of the invention, there is provided a system for
finding an optimal
geographic location for constructing an electric vehicle charging station,
said system comprising
one or more computer processors embodied in one or more computers, and one or
more non-
transitory computer readable media referenced in the immediately preceding
paragraph,
embodied in or connected to said one or more computers for execution of the
statement and
instructions on said one or more non-transitory computer readable media by
said one or more
processors of said one or more computers.
According to a tenth aspect of the invention, there is provided a computer-
implemented method
for evaluating conversion of produced power to cryptocurrency and calculating
associated costs
of said conversion, said method comprising the following steps executed by one
or more
computer processors embodied in one or more computers operably connected to a
communications network:
a. receiving user input on how much of the produced power should be converted
to
cryptocurrency;
b. calculating an amount of cryptocurrency that can be generated based on
an updated price of
the cryptocurrency received via said communications network;
c. identifying equipment requirements necessary to mine the cryptocurrency
using produced
power; and
d. searching one or more equipment supplier databases for equipment
fulfilling said equipment
requirements;
e. tallying a cost of located equipment in the one or more equipment supplier
databases that
fulfill said equipment requirements.
According to an eleventh aspect of the invention, there are provided one or
more non-transitory
computer readable media having stored thereon executable statements and
instructions for
execution by one or more processors, said statements and instructions being
configured to, when
executed, perform a method according to the tenth aspect of the invention.
Date Recue/Date Received 2023-03-08
According to a twelfth aspect of the invention, there is provided a system for
evaluating
conversion of produced power to cryptocurrency and calculating associated
costs of said
conversion, said system comprising one or more computer processors embodied in
one or more
computers, and one or more non-transitory computer readable media referenced
in the
immediately preceding paragraph, embodied in or connected to said one or more
computers for
execution of the statement and instructions on said one or more non-transitory
computer
readable media by said one or more processors of said one or more computers.
Among the foregoing embodiments, and other embodiments disclosed in more
detail herein
further below, are several useful, novel systems, methods and machines for
planning and
establishing renewable energy power plants and electric vehicle charging
stations, as well as
strategic planning and management of power output from such renewable energy
power plants.
Among these various embodiments, numerous advantages and benefits can be seen,
including:
- a simple interface that allows a user to work easily for finding the
best location and resources
for establishing renewable energy power plant;
- efficient processes that can be run in a short time, such that the user
attains final output
quickly and easily;
- user cost savings resulting from a resource efficient solution that is
affordable for all type of
customers;
- financial loss prevention by avoiding establishment of a renewable power
plant with no
return, as the best and the most appropriate resources can be located based on
machine
learning with LI (Location Intelligence) and Data Mining;
- efficient use of existing online data resources (e.g. GIS database and
NASA database) for
computing all required parameter;
- distinction between different optimal time periods for saving produced
renewable energy or
selling to a power network, because the selling price to the network varies at
different times;
- finding locations with renewable energy resources and frequently
commuting traffic on those
paths, denoting ideal locations for vehicle charging stations for electric and
plug-in hybrid
vehicles;
- digital map preparations of the calculated optimal locations using
previous data and some
data mining solutions, for display of the best locations for any type of
renewable energy
sources directly on the map;
BRIEF DESCRIPTION OF THE DRAWING FIGURES
The above and other features of this invention are described in the following
Detailed Description
and shown in the following drawings:
FIGS.1A-1C are flowcharts illustrating respective sequential stages of an
artificially intelligent
computer-implemented process for planning of a renewable energy power plant
using a
combination of geographic information system resources and user-inputted
project constraints;
6
Date Recue/Date Received 2023-03-08
FIGS.2A-2B are flowcharts illustrating step 110 of FIG.1A in more detail,
where a computer-
implemented determination is made of a best power plant type for a user-
specified location;
FIG.3 is a flowchart illustrating step 129 of FIG.1C in more detail, where a
computer-implemented
determination is made of the most appropriate equipment for the power plant
project;
FIG,4 schematically illustrates contents of a geographic information system
(GIS) database from
which useful data is gathered for use in step 127 of FIG.1C, where a computer-
implemented
evaluation and identification of a best location and power plant type for the
power plant project
is made in instances where a user-desired location was not specified at step
106 of IFG.1A;
F1G.5 is a hierarchical flowchart of a computer-implemented multi-criteria
decision making
(MCDM) process used to evaluate and identify the best location for the power
plant project in
step 127 of FIG.1C;
FIG.6 is a flowchart illustrating an initial computer-implemented pre-
evaluation of candidate
power plant types for minimum performance requirements in a geographic search
area from
within which the best location is to be found step 127 of FIG.1C;
FIG.7 is a flowchart illustrating a subsequent computer-implementation of the
MCDM process of
FIG. 5, broken down into a separate sub-process performed for each candidate
power plant type
that was found to fulfill the minimum performance requirements in the pre-
evaluation stage of
FIG.6;
FIG,8 is a flowchart illustrating the solar energy MCDM sub-process of FIG.7;
FIG.9 is a hierarchical chart showing effective parameters used in the solar v
MCDM sub-process
of FIG. 8;
FIG.10 is a flowchart illustrating the wind v MCDM sub-process of FIG.7;
FIG.11 is a hierarchical chart showing effective parameters used in the wind
energy MCDM sub-
process of FIG. 10;
FIG.12 is a flowchart illustrating the hydroelectric energy MCDM sub-process
of FIG.7;
FIG.13 is a hierarchical chart showing effective parameters used in the
hydroelectric energy
MCDM sub-process of FIG. 12;
FIG.14 is a flowchart illustrating the biomass energy MCDM sub-process of
FIG.7;
FIG.15 is a hierarchical chart showing effective parameters used in the
biomass energy MCDM
sub-process of FIG. 14.
FIG.16 is a flowchart illustrating step 124 of FIG.18 in more detail, where a
computer-
implemented evaluation is carried out regarding mining of cryptocurrency using
produced power
from the planned power plant project;
FIG.17 is a flowchart illustrating a computer-implemented power management
tool for evaluating
optimal time windows in which to save, consume and sell produced power;
FIG,18 is a flowchart of a novel computer-implemented MCDM process for
identifying the best
location for establishment of an electric/hybrid vehicle charging station, in
a manner similar to
the power plant MCDM sub-processes of FIGS.5-15;
FIG. 19 is a hierarchical chart showing effective parameters used in the
charging station MCDM
process of FIG. 18.
7
Date Recue/Date Received 2023-03-08
FIG,20 is a flowchart illustrating presentation of final results of the
computer implemented
processes of the preceding figures to a user, which may include, or
collectively form all or part of,
a comprehensive feasibility study;
FIG.21 schematically illustrates computer-executed corn pilation of said
feasibility study;
FIG,22 is a schematic block diagram of a computer system for implementing the
processes
illustrated in the preceding figures.
DETAILED DESCRIPTION OF THE INVENTION
The following description of exemplary embodiments of the invention is not
intended to limit the
scope of the invention to these exemplary embodiments, but rather to enable
any person skilled
in the art to make and use the invention.
FIGS.1A & 1B illustrate an initial data collection stage of a computer-
implemented renewal energy
power plant planning process of the present invention. Executable statements
and instructions
stored in one or more non-transitory computer readable media are executed by
one or more
computer processors to carry out the described process, of which said computer
readable media
and computer processors may be embodied in one or more computers, which are
connected to a
communications network by which those one or more computers can interface with
one or more
geographic information system databases. For brevity, these one or more
computers may be
referred to herein as simply a "machine". With reference to FIG. 22, the
machine of the illustrated
embodiment is embodied, at least primarily, by a server 1003 and one or more
"internal"
databases hosted thereby or connected thereto, and in the illustrated example
including an OLDB
database 1004 and an OLAP database 1005. Reference to these databases as
"internal" is meant
as indication that they are hosted, owned or operated by, or on behalf of, the
same operating
entity as the machine specifically for the dedicated purpose of supporting and
enabling the
described operability thereof.
These internal databases 1004, 1005 are distinguished from other "external
databases" that are
also referenced herein, and from which the machine retrieves necessary
information to perform
the various tasks disclosed herein, yet which typically will not have been
dedicated or designed
for the dedicated purpose of supporting and enabling operation of the machine.
Instead, these
external databases are typically hosted, owned or operated by separate outside
entities other
than the operating entity of the machine, and are typically also used for one
or more other non-
dedicated purposes, whether on a limited access basis to only a selection of
authorized parties,
or on a wide scale publicly accessible basis. The illustrated example includes
at least three
external databases, including a Geographic Information System (GIS) database
1006, a separate
NASA database 1007, and one or more equipment supplier databases 1008. Users
1000 interact
with the server 1003 over the internet or other wide area network, for example
via a web
application 1001 by which a graphical user interface (GUI) is displayed to the
users 1000 on local
client devices 1002 (workstation, desktop, laptop or tablet computers; smart
phones, etc.),
whereby the machine, at the various illustrated and described steps of the
data collection stage,
8
Date Recue/Date Received 2023-03-08
can query a user for input and collect the user's responses via one or more
inputs of the client
device (mouse, touchscreen, touchpad, keyboard, voice command, etc.).
In step 100 of FIG.1A, the machine asks the user to identify themselves as one
of a selectable
number of predetermined "customer types", categorized for example as:
Residential, Industrial,
Asset Developer and Governmental. Next, the user is queried at step 101
whether they wish to
specify a maximum budget for the project. At decision node 102, if the user
did not specify a
maximum budget, then the process continues on to step 103, where the user must
specify a
capacity requirement of the power plant, i.e. a quantity of power production
(Watts) the plant
will need to able to produce, before proceeding to step 104. On the other
hand, if the user
specified a maximum budget at step 101, decision node 102 bypasses step 103,
and goes directly
to step 104. Here, the user is asked if they want to specify a desired
location for establishing
power plant. At resulting decision node 105, if the user knows the desired
location, the user is
prompted to specify LAT (Latitude) and LONG (Longitude) for the desired
location in step 106,
otherwise decision node 105 leads to step 107, where instead of a specific
desired location, the
user must specify a geographic search region, for example by marking of same
on a digital map
shown in the GUI, from within which an optimal location for the power plant is
to be intelligently
determined by the machine.
Next, at step 108, the user is asked whether they wish to specify a power
plant type for the power
plant project at hand, for example being presented with a quantity of
predetermined and
selectable power plant types to choose from, which may include any two or more
of Photovoltaic
Energy, Wind Energy, Hydroelectric Energy, Geothermal Energy, Biomass and
Biogas Energy,
Ocean Energy. At decision node 109/109A, if the user specified a desired
location at step 106, but
did not specify a power plant type at step 109, then the process continues on
to step 110, where
the machine will follow the decision-making logic shown in FIGS. 2A & 2B,
described further
below, in order to determine an optimal power plant type based on the budget
and capacity
constraints inputted at steps 101 and 103, before proceeding to step 111. On
the other hand, if
the user specified a power plant type at step 108, and/or specified a
geographic search region at
step 107 instead of a specific desired location at step 106, then decision
node 109 leads directly
to step 111. Here, the user is prompted to specify consumption usage type,
where the user must
specify whether they want to consume the produced power or want to sell it to
the network.
Turning to FIG. 1B, which is continuation of the FIG. 1A flowchart, at step
112, the user is
prompted for a breakdown between a quantity of power that they want to sell to
the network
(step 113) and a quantity of power that the power plant entity wants available
for direct
consumption (step 114). Next, at decision node 115, based on the customer type
specified at step
100, the machine determines which further input data is required from the
user. If user was self-
categorized as a residential user, then at step 116, the user must specify
power consumption and
number of electrical appliances (based on number of appliances, the machine
can calculate how
much energy is needed for consumption). If user was self-categorized as an
industrial user, then
in step 117, the user must specify power consumption and number of electrical
equipment. If user
9
Date Recue/Date Received 2023-03-08
was self-categorized as an asset developer, then in step 119, the user must
specify number of
houses that will be made and power consumption. If user was a governmental
user, in step 118,
user must specify power production capacity (if capacity wasn't already
specified in step 103) that
it has been required.
From each of the alternatively executable steps 116-119, next is step 120,
where the user is
prompted to specify whether energy storage capability is required. At decision
node 121, if the
user did specify a need for energy storage, the user must then specify how
much energy they
want to store at step 122. On the other hand, if energy storage capability was
not required, the
process bypasses step 122 and goes directly to step 123. Here, the user is
asked whether they
wish to dedicate any power capacity of the power plant to the task of mining
cryptocurrency. If
yes, then at step 124, the machine will execute the cryptocurrency related
steps shown in FIG.16,
and described further below.
Steps 100-124 of FIGS. 1A-1B collectively embody the initial data collection
stage, where the user-
inputted data at each data collection step described above denote design
constraints on the
power plant project being planned. After receiving all this user-inputted
data, the machine, at
step 125, starts to search the one more geographic information system (GIS)
databases for
environmental, geographical, social, urban and geological parameters. In some
embodiments,
including the detailed embodiment set forth below with reference to the
accompanying figures,
multiple databases may be queried to collectively obtain sufficient parameter
data, for example
communicating with the NASA POWER Project database for environmental
parameters (e.g. solar
irradiance data, and climate data such as wind data and air pressure data, as
is relevant to
planning of solar and wind power plants), and using a separate GIS database
for other parameters
(spatial, land cover, geomorphology, economic, etc.). In this example, the
NASA environmental
database also includes GIS functionality, and so both databases are referred
to herein as GIS
databases, though there could be embodiments in which the environmental
parameters are
retrieved from a non-GIS database. The machine searches the GIS databases for
data associated
with either the specific desired location identified at step 106, if
specified, or the geographic
search region identified at step 107.
Turning to FIG.1C, if a geographic search region was specified at step 107,
then the process skips
step 126 and jumps straight to step 127, where an intelligent search for an
optimal power plant
location is conducted, detailed explanation of which is given herein further
below, with additional
reference to FIGS.4-15.0n the other hand, if a specific desired location was
specified at step 106,
then the machine needs to obtain GIS parameter data for the particular
Latitude and Longitude
of that desired location, but the GIS database(s) may not contain parameter
data points for that
specific geographic coordinate. So, the machine, in step 126 of FIG.1C, uses a
2D interpolation
method to accurately estimate the GIS parameter data of the specific desired
location.
Here, the machine uses bilinear interpolation, because there are two variables
(Latitude and
Longitude). Bilinear interpolation is performed using linear interpolation
first in one direction, and
Date Recue/Date Received 2023-03-08
then again in the other direction. Although each step is linear in the sampled
values and in the
position, the interpolation as a whole is not linear but rather quadratic in
the sample location. The
machine needs to determine interpolated parameter data point values (for solar
irradiance,
temperature, wind, speed, air pressure, spatial, land cover, etc.) for
coordinate point (x, y) using
the database's available parameter data point values for coordinate 0
pointsQii 0 0
,,12, ,21, ,22
where:
Qii 311)
412 = (X1fY2)
Q21 = (X2) Y1)
Q22 = (X20312)
Therefore, the machine computes the following equation:
fY = +
X2 - X X - X1
f f MO,
- -
X2 - X X - X1 r
f Y2) = f (Q12) + f (Q22) =
x2 -x1 x2 -x1
Then, the machine computes the parameter data point value of given location:
Y2 Y YYi
f (x,y) = f (x, + f(x, Y2)
YYi Y2 ¨
Y2 Y x2 ¨ x ¨
f(Qii) + i(Q21)
Y kx2 ¨ X2-X1
YYi ( ¨
(G.2) + _________________________________ ¨
f(Q22)
Y2 Y1 µx2 X2 -X1
1
= (x2 - xi) (Y2 - Yi) (f (Q11) (X2 - X)(Y2 y) + f (Q21)(x - xi) (y2 -
+ f (Q12) (x2 - x)(y - Yi) + f (Q22)(x - xi)(y- yi))
1
______________________ [X2 - X X - Xi] [f V211) f(Q12)1 v2 - I
(X2 - X1) (Y2 - Yi) f (Q21) f (Q22)J
Continuing with the scenario in which a specific desired location was
specified, and having now
attained a dataset composed of interpolated parameter data point values for
the specific desired
location, step 127 is bypassed. Next, at step 128, the machine calculates all
needed equipment
based on budget, efficiency and energy type, and connects to one or more
supplier databases via
a suitable WEB Service to find that needed equipment, as outlined in more
detail below in relation
to FIG.3. Finally, at step 129, the machine calculates key performance
indicators (KPI) to compile
11
Date Recue/Date Received 2023-03-08
into a feasibility study. After all such calculation, the machine, at step 130
saves all computed
results and generated feasibility study reports, for immediate or later
display, or electronic
transmission, to the user, thus completing the overall process.
FIGS. 2A & 2B illustrate step 110 of Fla1A in greater detail. In step 200
budget status will be
determined: if budget was not specified and capacity was instead specified,
the process proceeds
to step 201 and retrieves the previously specified budget; otherwise, the
process proceeds to step
202 to check if capacity was already specified, and if not, then collects user-
specified budget at
step 201 before continuing on to step 203, where the machine computes annual
solar energy
production. In step 204 if the budget is less than budget threshold for wind
energy (because for a
given amount of power capacity, it is known how much budget is needed based on
data that
stored in the local database(s) for wind energy), then the machine proceeds to
compute hydro
energy annual production in step 207 based on required power capacity. If the
budget exceeds
the wind energy threshold, then the machine computes wind energy annual
production in step
205 based on required power capacity.
After that, in step 206,the machine checks if the budget is less than budget
threshold for hydro
energy (because for a given amount of power capacity, it is known how much
budget is needed
based on data stored in the local database(s) for hydro energy), or if the
specified location is
beyond a threshold distance from the nearest river. If yes, then the process
proceeds to step 209
for computing biomass energy annual production; if no, then the process
proceeds to step 207
for computing hydro energy annual production. After that in step 208, the
machine checks if the
budget is less than a budget threshold for biomass energy (because for a given
amount of power
capacity, it is known how much budget is needed based on data stored in the
local database(s)
for biomass energy), or if the location is beyond a distance threshold from
the nearest city
(because close proximity to a city is most suitable for biomass energy due to
availability of
municipal waste as biofuel). If yes, then the process proceeds to step 219,
where the machine
displays solar capacity and selects a photovoltaic power plant as the final
plant type decision
(based on the logic that establishing a wind farm, hydro power plant or
biomass power plant
usually costs more and requires a certain minimum cost, and this minimum cost
may vary at
different times). Otherwise, the process proceeds to step 209 for computing
biomass energy
annual production.
Next, in step 210 of FIG.2B, the machine compares the capacity of solar and
wind energy annual
production, and if the wind capacity is less than solar capacity, the machine,
in step 212, compares
solar capacity with hydro capacity; otherwise the process proceeds to step 211
for comparing
wind capacity and hydro capacity.
In step 211, the machine compares wind capacity with hydro capacity, and if
solar capacity is
greater than hydro capacity, then in step 218 the machine compares solar
capacity with biomass
capacity, and if solar capacity is greater than biomass capacity. then in step
219 the machine
selects a photovoltaic solar power plant as the final plant type decision.
In step 211, if wind capacity is less than hydro capacity, then the process
proceeds to step 214 for
comparing hydro capacity and biomass capacity; otherwise the process proceeds
to step 213 for
comparing wind capacity and biomass capacity. If wind capacity is greater than
biomass capacity,
then in step 215, the machine selects wind energy as the final plant type
decision; otherwise, the
12
Date Recue/Date Received 2023-03-08
process proceeds to step 217, where the machine selects biomass energy type as
the final plant
type decision.
In step 212, if solar capacity is greater than hydro capacity, then the
process jumps to step 218,
where the machine compares solar capacity with biomass capacity, and if solar
capacity is greater
than biomass capacity, then in step 219 the machine selects a photovoltaic
solar power plant is
the final plant type decision; otherwise, the process proceeds to step 217,
where biomass energy
is selected as the final plant type decision.
In step 212, if solar capacity is less than hydro capacity, then the process
proceeds to step 214 for
comparing hydro capacity and biomass capacity, and if hydro capacity is
greater than biomass
capacity, then in step 216, the machine selects hydroelectricity as the final
plant type decision;
otherwise, in step 217, the machine selects biomass energy as the final plant
type decision.
The algorithm of main idea of choosing the best source is shown in below. Note
that, as
mentioned above, there are many limitations for establishing solar, wind,
hydro and biomass
power plant. Constructing a wind farm requires a certain minimum cost x1 and
it is not possible
to establish a wind farm for a capacity lower than a certain number y2. And
also, it is not possible
to establish a solar power plant for a capacity greater than a certain number
Yi. Constructing a
hydro power plant requires a certain minimum cost x2and it is not possible to
establish a hydro
power plant for a capacity lower than a certain number y3. Constructing a
biomass power plant
requires a certain minimum cost x3and it is not possible to establish a
biomass power plant for a
capacity lower than a certain number y4.
INITIALIZE;
IF Budget is given AND Capacity is given THEN
IF Budget <x1 or Capacity < y2 THEN
Wind will be removed from the list
ELSE
COMPUTE annual wind energy production
END IF
IF Budget <x2 or Capacity <y3 THEN
Hydro will be removed from the list
ELSE
COMPUTE annual hydro energy production
END IF
IF Budget <x3 or Capacity < y4 THEN
13
Date Recue/Date Received 2023-03-08
Biomass will be removed from the list
ELSE
COMPUTE annual biomass energy production
END IF
IF Capacity < yi THEN
COMPUTE annual solar energy production
END IF
END IF
IF Budget is given AND Capacity is not given;
IF Budget <x1 THEN
Wind will be removed from the list
ELSE
COMPUTE annual wind energy production
END IF
IF Budget < x2 THEN
Hydro will be removed from the list
ELSE
COMPUTE annual hydro energy production
END IF
IF Budget < x3 THEN
Biomass will be removed from the list
ELSE
COMPUTE annual biomass energy production
END IF
COMPUTE annual solar energy production
END IF
14
Date Recue/Date Received 2023-03-08
IF Budget is not given AND Capacity is given;
IF Capacity > yi THEN
Solar will be removed from the list
ELSE
COMPUTE annual solar energy production
END IF
IF Capacity <Y2 THEN
Wind will be removed from the list
ELSE
COMPUTE annual wind energy production
END IF
IF Capacity <y3 THEN
Hydro will be removed from the list
ELSE
COMPUTE annual hydro energy production
END IF
IF Capacity <y4 THEN
Biomass will be removed from the list
ELSE
COMPUTE annual Biomass energy production
END IF
END IF
FIG.3 is a more detailed illustration of step 129 from FIG.1C, where the
machine searches and
chooses optimal equipment for the renewable power plant being planned. For
establishing a
renewable power plant and calculating its capacity, daily and yearly power
production, system
efficiency, initial cost, maintenance cost and preparing feasibility study
reports, it is useful to find
all the required equipment in detail, where after this equipment data can be
presented to the
user and/or used for various computations. Finding appropriate equipment is a
complex problem
and can be done by searching in one or more equipment supplier databases and
computing its
efficiency and final price for establishing a power plant. Obviously, this
process can't be done by
individuals, and needs a calculation loop for doing query and calculating
required data. Referring
back to steps 101-103 of FIG.1A, there are two different perspectives from
which the equipment
Date Recue/Date Received 2023-03-08
problem can be approached: from a specified budget perspective, or from a
capacity requirement
perspective. In the scenario where a budget has been specified at step 101,
the machine
computes and allocates a budget share of the overall specified budget to each
piece of required
equipment for the particular type of power plant concerned (as previously
specified at step 108,
or determined at step 110 or 127). The necessary equipment breakdown criteria
for each
predetermined power plant type is known, and stored in a manner accessible to
the machine, for
example in a reference database hosted thereby or connected therewith, to
enables these
searches on the equipment supplier database(s) based on these criteria. This
equipment search
process is accordingly based on each renewable power plant type cost breakdown
and calculates
an estimate of how much money should be spent for each equipment need. In the
second
scenario, where power plant capacity has been specified at step 103, the
machine computes the
required quantity and quality (i.e. production capacity) of each piece of
required equipment to
achieve the specified capacity requirement of the power plant, then searches
out the price of
each piece of required equipment for the next computation. All of above
computation will be
done by searching in supplier's database for an optimal solution.
So, with reference to FIG. 3, the machine first checks at decision node 300
whether budget or
capacity was specified earlier on at step 101 or 103. If budget was specified,
the process continues
to step 301, or if capacity requirement was specified, the process bypasses
step 301 and proceeds
straight to step section 302. In step 301, an attainable capacity achievable
within the specified
budget is calculated, and then the machine calculates an estimated budget
share price of each
equipment according to the stored cost breakdown criteria of the power plant
type concerned.
These estimates are used to subsequently guide an assessment of accurate true
costs by searching
in equipment supplier database(s) in section 303.
The calculations at step 301 can be represented by the following formulas:
Pi
Pni
A = BP
Bpi -
A=
Bp _
Or we can consider:
= Bpi , a2 = Bp2
Where: pn is the percentage budget share of equipment n, B is the specified
budget and an is the
estimated price of equipment n.
In step 302, the machine computes required equipment based on power capacity
that the
machine will compute based on the budget to achieve the ability of producing
demand power
that has been specified by the user. The calculation of this section can be
represented by the
following formulas:
16
Date Recue/Date Received 2023-03-08
Pn = M n
Where: P, is nominal power capacity (subscript n in 1), represents nominal
power), M is power
production capacity of main equipment to produce power such as: solar panel,
turbine, etc. and
n is quantity of main equipment.
In section 303, machine searches on the equipment supplier database(s) to find
equipment based
on and the results of step 302. For the scenario in which budget was
specified, searching in
database is based on the estimated budget share price of the required
equipment, determined at
earlier step 301, and so the machine searches for equipment of the required
type whose supplier-
listed cost is less than the estimated budget share price. In the scenario
where capacity was
specified instead of budget, searching in database is based on required
equipment to achieve the
specified capacity requirement. In section 304, having identified candidate
equipment options in
search step 303, the machine evaluates each candidate equipment option based
on a combination
and price and efficiency, and compares the different candidate equipment
options against one
another to determine an optimal candidate equipment option with the best
price/ efficiency
result.
One possible algorithm for executing step 304 is given below, where if the
budget, but not the
capacity requirement, was specified, then the attainable capacity (that can be
produced based on
the specified budget and the power plant type concerned) is first calculated.
Then based on the
specified or calculated attainable capacity, the machine searches the
reference database to find
all required equipment types for power plan type concerned, then searches on
equipment
supplier data base(s) (via implemented web service(s)) and finds all candidate
equipment options
for each required equipment type, and then calculates price/efficiency for
every candidate
equipment option. Since the capacity of every candidate equipment option is
known, the machine
will find the optimal candidate equipment option that has the best
price/efficiency and the best
capacity capability relative to the user's required capacity, and selects this
as the recommended
equipment option for the given equipment type.
INITIALIZE;
IF Budget is specified THEN
Calculate the capacity based on energy type and budget
END IF
FOR i in n /*finding all required equipment based on energy type and
capacity*/
FOR j in k /*finding an equipment in all supplier's database*/
CREATE a list of all equipment for ei,j
CALCULATE price/ efficiency of each equipment ei,j
END LOOP
END LOOP
FOR i inn
17
Date Recue/Date Received 2023-03-08
FOR j in k
COMPARE price/ efficiency of each equipment
COMPARE Capacity of each equipment
IF price/ efficiency ei > e+1, AND Capacity eu > ei+Li
RETURN ei,j
ELSE IF price/ efficiency ej,1 > e+1, AND Capacity eid <
RETURN eij
ELSE
RETURN ei+i,j
END IF
END LOOP
END LOOP
And finally in section 305, and after full searching of the equipment supplier
database(s), the
machine prepares a list of the recommended equipment option for each required
equipment type
(for example, using the above algorithm). Once identified, these recommended
equipment
options may be used in other future machine executed computations, for example
in generation
of one or more various reports, and are shown to the user, and/or stored in
computer-readable
memory of the machine for future retrieval and display. Examples of the
required equipment
types for the different power plant types may include, for instance, solar
panels and inverters for
photovoltaic solar power plants; blades, gearboxes and generators for wind
turbine power plants;
hydraulic turbines and hydroelectric generators for hydroelectric power
plants; biomass furnaces,
boilers and steam turbines for biomass power plants. Note that, for
establishment of a power
plant under a certain circumstance, required main equipment may be the same as
previously
calculated, so the stored data and their technical and economic results can be
advantageous.
FIGS.4-15 elaborate on the details of step 127 for determining the best
location for establishing
the power plant when a specifically desired location was not specified by the
user. The novel
methodology employed for these purposes denotes a significantly valuable
aspect of the present
invention. In the following detailed embodiment of this methodology, all
effective parameters
for planning a renewable energy power plant are considered, in an artificially
intelligent
evaluation and decision-making process that cannot be implemented as a mental
or pen-and-
paper exercise, at least in part owing to its use of such an extensive breadth
of parameters. There
are a vast variety of significant parameters falling into such categories as:
Spatial considerations,
Land Cover classification, Geomorphology and Economic considerations, which
individually and
collectively have considerable impact on power production and other aspects of
a renewable
energy power plant. In the following methodology, these parameters and their
effects on the best
location for establishing a renewal energy power plant are considered.
Renewable energy sources
18
Date Recue/Date Received 2023-03-08
are inextricably tied to their location and the geographical parameters
associated therewith, and
so producing electricity is directly connected with geographical location.
The process of finding the best locations for the setting up of renewable
power plants requires
the gathering of data with regards to relevant factors of influence. The use
of Geographical
Information System (GIS) has gained a lot of popularity as a site suitability
analysis tool involving
the assimilation of spatially referenced data in a problem-solving
environment.
In this problem, a great deal of factors plays a significant role in the
assessment of suitable
locations, and as mentioned above, can be categorized into such categories as
spatial
considerations, land cover classification, geomorphology, economic
considerations, etc. By
employing one or more Geographical Information System (GIS) databases as a
data resource, and
applying a Multi-Criteria Decision-Making (MCDM) based approach for analyzing
said data, an
effective computer-implemented method for selecting optimal locations for
renewable power
plants is derived. The MCDM analytical approach provides a substantial means
to handle the
shortages of GIS in analyzing cases involving complex criteria and objectives.
The GIS serves as a
valuable tool in the MCDM problem addressed herein, in which geo-referenced
information plays
a crucial role. The employed GIS approach has the capabilities of data
storage, data management,
calculations, analysis, and visualization of the raw georeferenced data in a
meaningful manner.
The MCDM approach is particularly well suited for solutions to complex
problems where multiple
factors affect a single goal. The MCDM approach provides a suitable option
through the evaluation
and comparison of the characteristic properties of the alternatives. Thus, by
combining GIS
resources with an MCDM analytical approach, a unique and cohesive framework is
possible that
can handle the complex spatial planning problem addressed by the present
invention.
GIS can be generally defined as tools for consulting, analyzing and editing
data, maps and spatial
information in general. GIS are systems that embody geographic information
databases in
association with a digital map, from which the geographic coordinates of each
point within the
mapped geographical area can be obtained. This means that it is possible to
search in both
directions, obtaining information on the map, or performing the search
directly from the
database. One of the most significant features of GIS is it can represent an
area by using many
layers, of which each layer represents a different factor. In the presently
faced problem of
selecting the best location for renewable power plant, there are plenty of
relevant and
parameters, and as a result, each parameter can be represented by a layer
(FIG.4). These
parameters may have different effects on different types of power plants, and
so the weight of
these parameters should be varied among the different types of renewable
energy power plants
being considered.
MCDM is a computer-implemented procedure that consists in finding the best
alternative among
a set of feasible alternatives. The main hierarchical flow chart of Multiple-
criteria decision-
making (MCDM) is given in the FIG.5. In the context of the present invention,
the alternatives are
different candidate locations for the power plant project being planned, and
from among which
the best candidate location is to be identified.
An MCDM problem with m alternatives and n criteria can be expressed in matrix
format as:
19
Date Recue/Date Received 2023-03-08
Ca]
z = F Z11 === Zln
. .
Zmi Zmn
W [W1 === Wit]
M= AxZ
Cl Cn.
A111- zn
M=
Ami[zmi Zmni
=== .. Wn
Where Ai; A2; Am are feasible alternatives; C1; C2; ...; Cn are evaluation
criteria; Zu is the
performance value of alternative Ai under criterion Ci; and wi is the weight
of criterion C. The
decision matrix can be represented by the following table:
Objective Criterion 1 (C1) Criterion 2 (C2) .
Criterion n (Ca)
Alternative 1 (A1) zn Z12 = = = Zln
Alternative 2 (A2) Z21 Z22 = = = Z2n
Alternative m (Am) Zmi Zm2 . . . Zmn
Weight (w) w1 wz = wn
To properly determine the weight of each criterion or factor involved in the
final outcome of the
resulting layers, the Analytic Hierarchy Process AHP method, proposed by
Saaty, is used within
the MCDM. It is a pair-wise comparison procedure of the criteria that is based
on a square matrix
in which the number of rows and columns is defined by the number of criteria
to weigh. This
method bases its operation in the calculation of the distances to the ideal
point and the anti-ideal
point. This methodology was chosen because it does not require an assessment
by a
knowledgeable expert for each of the alternatives; they can be evaluated
directly from the
database provided by the GIS assessment of each criterion for each
alternative. Basically, AHP has
three underlying concepts: structuring the complex decision problem as a
hierarchy of goal,
criteria and alternatives, pair-wise comparison of elements at each level of
the hierarchy with
respect to each criterion on the preceding level, and finally vertically
synthesizing the judgements
over the different levels of the hierarchy.
Date Recue/Date Received 2023-03-08
Here, it is assumed that there are n different and independent alternatives Ai
(i = 1, n) and
that they have the weights wi = 1,
n), respectively. Also, it is assumed that the quantified
judgements on pairs of alternatives (Ai, 111) are represented in an nxn matrix
as follows:
.==
A1 an
:
A = [aiii =
nxn i =
an1 .== ann
Where: alternatives are Ai (i= 1, n), while the weight ratio of pairs is given
in each element au
of matrix A, which can be expressed by the following expression:
Wi
au =
Wi
Where: wi and w are local weights of elements i and j in relation to another
element. The values
assigned to au according to the Saaty scale are usually in the interval of 1-9
or their reciprocals.
The following table presents Saaty's scale of preferences in the pair-wise
comparison process.
Definition Intensity of Importance
Ai is equally important to A1 1
Ai is slightly more important than 111 3
Ai is strongly more important than 111 5
Ai is very strongly more important than Ai 7
Ai is extremely more important than Aj 9
Intermediate values 2, 4, 6, 8
The process consists of decomposing a complex problem into a hierarchy,
keeping the goal at the
highest position of the hierarchy, criteria and sub- criteria at intermediate
levels and sub-levels of
the hierarchy, and decision alternatives at the lowest level of the hierarchy.
A pairwise
comparison of the elements at a given hierarchy against the elements in the
next higher level is
conducted in order to evaluate their relative preference with respect to each
other. A rating scale
of 1-9 is used to grade the preference between two elements. In this scale the
lowest value 1
implies less importance and the highest value 9 implies strong importance. The
other values (2-
8) imply intensities of importance graded between 1 and 9. The scale is used
to attribute weight
coefficients for the quantifiable and no quantifiable elements which are
obtained in the form of
a weight coefficients vector, reflecting the intensity of importance of each
option relative to the
goal stated at the highest position of the hierarchy. On this basis, pairwise
comparison matrices
are constructed considering the following principles: "a given element of the
matrix is equivalent
21
Date Recue/Date Received 2023-03-08
to itself, i.e., equal to 1 and the value of element a with respect to element
b is the reciprocal of
the value of element b with respect to element a".
The following equation gives the values of the normalized entries (aii) of
matrix M:
a! = _________ , for i = 1, 2, ..., n
E7--1 ail
and the following equation is used to compute the local weights of criteria
wI ¨ _________ , for i = 1, 2, ... , n
Before using Multi-Criteria Decision-Making (MCDM) to find the optimal
location for establishing
renewable energy power plants, for best results, the points on the GIS map
selected as candidate
locations for the power plant should meet a set of minimum requirements for
establishing a
power plant. By this, it is meant that it is not economical to set up power
plants in geographic
locations that do not meet the minimum requirements, and there is no need to
include such
unsuitable locations in the MCDM analysis. These unsuitable locations are thus
pre-emptively
excluded from the alternatives list. To pre-emptively filter out such
unsuitable locations from the
pool of potential candidate locations, it is necessary to define a minimum
number of super
effective parameters and check whether predefined thresholds for those super
effective
parameters are met for each of the potential alternatives.
These super effective parameters for each potential candidate location are
checked and
controlled by checking the relevant GIS layers, and if they do not meet the
minimum
requirements, the GIS data points for that potential candidate location stored
in a new exclusion
GIS layer as unusable points and are not checked as an alternative in the MCDM
process, having
been pre-emptively excluded therefrom.
FIG.6 illustrates this pre-emptive filtering out of unsuitable locations from
the alternatives list so
that only suitable candidate locations populate the final set of alternatives
for the MCDM analysis.
At step 400, the machine identifies a geographic search region, for example
based on user
selection thereof on a digital map thereof in the GUI, or retrieval from
memory of a user-selected
search region already specified at step 107. In step 401, the machine
retrieves defined minimum
requirement values of the super effective parameters for the predetermined
power plant types,
which as mentioned previously may include some or all of the following non-
exhaustive examples:
= Solar Energy
= Wind Energy
= Hydropower Energy
= Biomass Energy
for which the super effective parameters are, respectively:
= Solar Irradiance (GHQ
22
Date Recue/Date Received 2023-03-08
= Average Wind Velocity
= Rivers average flow rate
= Distance to Urban Area
The minimum required values of above parameters are predefined values stored
in memory for
retrieval by the machine, and exemplary examples of which are given in the
following table. In
step 402, based on the machine's identification of the super effective
parameters (or restriction
factors), the machine prepares a related GIS layer for each super effective
parameter by using the
geographic search area defined in step 401. Non-limiting examples of possible
minimum
requirement values for the different power plant types are given in the
following table:
Solar Wind Hydroelectric
Biomass
Solar Irradiance (GNI) >4.5
kWh/m2/day
Average Wind Velocity > 5 m/s
Rivers Average Flow Rate > 2.07 m3/s
Distance to Urban Area <10
km
At step 403, the machine is now ready to start checking the minimum
requirement for each energy
type, for example shown as four separate steps 404, 405, 406 and 407 for four
different power
plant types: solar, wind, hydroelectric and biomass. In step 404, machine uses
the GIS layer of
solar Global Horizontal Irradiance Gill (kWh/m2/day) in raster mode, then
machine reads GHI data
and compares it with the associated threshold of that restriction factor, and
if any point does not
meet the threshold, machine add this raster point to a new exclusion GIS
layer. The following
algorithm shows the method of this computation:
INITIALIZE;
FOR i in all raster points
IF raster GHli < 4.5 kWh/m2/clay:
ADD this raster point to exclusion layer
END IF
END LOOP
In step 405, in similar fashion to step 404, the machine prepares a GIS layer
of Average Wind
Velocity (m/s) in raster mode, then the machine reads the Average Wind
Velocity data and
23
Date Recue/Date Received 2023-03-08
compares it with the associated threshold of that restriction factor, and if
any point does not meet
the threshold, the machine adds this raster point to another new exclusion GIS
layer. The
following algorithm shows the method of this computation:
INITIALIZE;
FOR i in all raster points
IF raster Average Wind Velocityt < 5 m/s:
ADD this raster point to exclusion layer
END IF
END LOOP
In step 406, in similar fashion to step 405, the machine prepares a GIS layer
of Rivers Average Flow
Rate (m3/s) in raster mode, then the machine reads the Rivers Average Flow
Rate data and
compares it with the associated threshold of that restriction factor and if
any point does not meet
the restriction factor, machine add this raster point to another new exclusion
GIS layer. The
following algorithm shows the method of this computation:
INITIALIZE;
FOR i in all raster points
IF raster Rivers Average Flow Ratet <2.07 m3/s:
ADD this raster point to exclusion layer
END IF
END LOOP
And finally in step 407, in similar fashion to steps 404-406, the machine
prepares a GIS layer of
Distance to Urban Area (km) in raster mode, then the machine reads the
Distance to Urban Area
data and compares it with the associated threshold of that restriction factor,
and if any point does
not meet the threshold, the machine adds this raster point to another new
exclusion GIS layer.
The following algorithm shows the method of this computation:
INITIALIZE;
FOR i in all raster points
IF raster Distance to Urban Areat > 10 km:
ADD this raster point to exclusion layer
24
Date Recue/Date Received 2023-03-08
END IF
END LOOP
In step 408, the machine gathers and combines all of the forgoing exclusion
layers into a new
Combined Exclusion Layer in a raster map. The points specified in this
combined exclusion layer
are not suitable for use in the MCDM analysis, and are thus prohibited points
specifically excluded
therefrom. At step 409, pre-emptive exclusion of unsuitable locations for the
power plant is
complete, whereby the scope of candidate locations within the geographic
search region has been
effectively narrowed for more efficient processing in the MCDM analysis that
follows in FIG.7.
FIG.7 presents the MCDM analysis, broken down into four sub-processes each
dedicated to a
respective one of the predetermined power plant types, and which are listed
below:
ò Solar Energy Multi-Criteria Decision-Making sub process in step 410
ò Wind Energy Multi-Criteria Decision-Making sub process in step 411
ò Hydropower Energy Multi-Criteria Decision-Making sub process in step 412
ò Biomass Energy Multi-Criteria Decision-Making sub process in step 413
The four decision-making sub-processes are performed separately using the
relevant GIS layers
and effective parameters for the given power plant type, and the best
respective location to
establish a solar power plant, wind power plant, hydroelectric power plant and
biomass power
plant is determined separately, and the corresponding score will be calculated
for each for
evaluation of the different location/plant-type combinations against one
another. In step 414, the
overall results are reviewed and merged, and in step 415, the best
location/plant-type in the
geographic search area will be selected and displayed to the user, and/or
stored in memory for
later retrieval and display. Steps 414 and 415 will be explained in more
detail further below, after
first elaborating on the details of the plant-specific MCDM sub processes 410,
411, 412 and 413.
Firstly, the solar energy MCDM sub process is shown in FIG.8, wherein in step
416, the
alternatives, which are the raster pixels in the GIS layer of solar irradiance
(GHQ, are determined
and prepared in raster mode for the computations. In step 417, criteria and
sub-criteria of
effective parameters for establishing photovoltaic power plant, as shown in
FIG.9, are identified
by the machine, based on stored instructions in memory, for use in subsequent
steps and
calculations.
In this detailed but non-limiting example, there are five main criteria and
nineteen sub-criteria for
the solar energy power plant type. In step 418, the weights of the criteria
and sub-criteria related
to the solar energy criteria and sub-criteria identified in step 417, having
been predetermined,
and calculated in accordance with the Saaty method mentioned earlier, either
by, or with the aid
or guidance of a knowledgeable expert, and then stored in computer-readable
memory of the
machine as normalized Saaty pairwise comparison matrices, are now retrieved
from memory for
use of this predetermined solar-specific weighting scheme in the solar MCDM
sub process. The
sub-criteria comparison with calculated local weight for each type of
parameters are shown
below.
Date Recue/Date Received 2023-03-08
Spatial sub-criteria pairwise comparison matrix with local weights
Sub-Criteria Road Railway Power Grid Settlements Weight
Road an au a13 a14 wsi
Railway a21 a22 a23 a24 W52
Power Grid a31 a32 a33 a34 w53
Settlements a41 a42 a43 a44 ws4
Land cover sub-criteria pairwise comparison matrix with local weights
Sub- Barren Grass- Bush- Emerging Farm
Pastures Forest Wetland Weight
Criteria Lands land land Forests Land
Barren "1/.
Lands 1
an au a" a14 als a16 a17 a"
Grassland a21 a22 a23 a24 a25 a26 a7 a28 WL, 2
Bushland a31 a32 a33 a34 a35 a36 a37 a38 w/.3
Pastures a41 a42 a43 a44 a45 a46 a47 a48 WL4
Emerging
Forests /251 (252 a53 a54 ass a56 a57 4258 w/..
5
Farm Land a61 a62 a63 a64 a65 a66 a67 a68 WL 6
Forest a71 a72 a73 a74 (275 a76 a77 am WL7
Wetland a81 a82 a83 a84 (285 a86 a87 a88 WL 8
Geomorphology sub-criteria pairwise comparison matrix with local weights
Sub-Criteria Slope Orientation Elevation Weight
Slope an an a13 wGi
Orientation a21 a22 a23 wG2
Elevation a31 a32 a33 WG3
26
Date Recue/Date Received 2023-03-08
Economic sub-criteria pairwise comparison matrix with local weights
Sub-Criteria Population Tourist Night Power
Unemployment Weight
Consumption
Population a11 a12 a13 a14 WEi
Tourist Night a21 a22 a23 a24 WE 2
Power Consumption a31 a32 a33 a34 WE 3
Unemployment a41 a42 a43 a44 WE 4
The predetermined expert-derived weights may be periodically updated to
account for changing
circumstances. In the weights of the sub-criteria factors, the alphabetic
subscript in the weight w
is shorthand for the the sub-criteria name. So, in the term w51 the "S"
represents "Spatial" and
the "1" represents the first factor in the matrix. Likewise, "L" in wLi
represents "Land Cover", "G"
in wGi represents "Geomorphology", "E" in wEi represents "Economic".
Main criteria pairwise comparison matrix with local weights
Main-Criteria GHI Spatial Land
Cover Geomorphology Economic Weight
GHI an an a13 a14 a15 1411
Spatial a21 a22 a23 a24 a25
Land Cover a31 a32 a33 a34 a35 IN3
Geomorphology a41 a42 a43 a44 a45 w4
Economic a41 a52 C153 a54 a55 w5
It should be noted that in this case that there is a problem with a
hierarchical structure including
some criteria and sub-criteria (due to the range and variety of effective
parameters, it is better to
classify them hierarchically), and so the machine calculates the average
weights of criteria and
sub-criteria first, as per the formula given above for w, and saves these as
local weights. Later
on, in a ranking function detailed herein further below, global weighting of
the sub-criteria will be
applied by multiplying the averaged local weights of the sub-criteria by the
average weights of
their upper-level criterion.
Referring again to Figure 8, in step 419, the machine uses the GIS layer for
completing a
performance matrix of each alternative. Every pixel in the raster map is an
alternative (a candidate
location for a solar power plant) and its performance value for a given
criteria must be collected
from GIS by using the related layer. The following equation presents
performance value of each
alternative for related criteria:
27
Date Recue/Date Received 2023-03-08
Z11 === Zln
=
Zin1 = == Znin
Objective Criterion 1 (C1) Criterion 2 (C2) Criterion
n (Ca)
Alternative 1 (A1) Zll Z12 = = = Zin
Alternative 2(A2) Z21 Z22 = = = Z2n
" . . . . . . .
Alternative m (Am) Zmi Zm2 Zmn
Weight (wi) w1 14/2 = = = wn
In Step 420, the machine computes the normalized weight of the performance
matrix (which
may alternatively be referred to as a "decision matrix"). The machine first
applies vector
normalization to normalize the performance matrix values:
Tj zii
= ______________ , for i = 1, 2, ..., n and j = 1,2, ... , m
Z1i
Where ?If is normalized value and zu is the performance value of the
alternative i for the
criterion j.
Though not necessarily calculated by the machine at this stage, but in
supportive illustration of
the calculation of ranking score functions below, a weighted normalized
decision matrix can be
computed by multiplying the weight of each decision criterion (wi) by each
already normalized
performance value of the decision matrix for that criterion, as presented in
the following
equation:
Vii = wi x for i = 1, 2, ... , n and j =
1, 2, , m
Vin
V =[
Vm1 Vinn
So, in step 421, the machine calculates a ranking score for each alternative
(candidate location)
using a predetermined solar-specific ranking function. The calculated ranking
scores of the
candidate locations for a solar power plant are then compared to identify the
best candidate
location for a solar power plant. The equation of the solar-specific ranking
function in this specific
example is given below:
F = w1R1 +
w2(ws1R2 + w52R3 + w53R4 + Ws 4R5)
28
Date Recue/Date Received 2023-03-08
W3(WL1R6 WL2R7 WL3R8 WL,4R9 WL5R10 WL6Rii WL7R12 WL8R13)
W4(WG1R14 WG2R15 WG3R16)
W5(WE1R17 WE2Ri8 WE3Ri9 wE4R20)
Where w1, w2, w3, w4 and ws are the local weights of main criteria:
: weight of GNI criterion
w2 : weight of Spatial criterion
w3 : weight of Land Cover criterion
w4 : weight of Geomorphology criterion
: weight of Economic criterion
wE2, wE3 and ws4 are the local weights of Spatial sub-criteria:
wE1 : weight of Road criterion
wE2 : weight of Railway criterion
wE3 : weight of Power Grid criterion
wE4 : weight of Settlements criterion
wL2, wL3, wc4, wc5, wc6, wL7 and wL8 are the local weights of Land Cover sub-
criteria:
wLi : weight of Barren Lands criterion
wL2 : weight of Grassland criterion
: weight of Bushland criterion
wL4 : weight of Pastures criterion
: weight of Emerging Forests criterion
wL,6 : weight of Farm Land criterion
wL7 : weight of Forest criterion
w/.8 : weight of Wetland criterion
wGi, wG2 and wG3 are the local weights of Geomorphology sub-criteria:
wGi : weight of Slope criterion
wG2 : weight of Orientation criterion
wG3 : weight of Elevation criterion
wEi, wE2, wE3 and wE4 are the local weights of Economic sub-criteria:
wEi : weight of Population criterion
wE2 : weight of Tourist Night criterion
wE3 : weight of Power Consumption criterion
wE4 : weight of Unemployment criterion
And also, term "Re" represents normalized performance value of each
alternative:
29
Date Recue/Date Received 2023-03-08
: performance value of GNI raster
R2 : performance value of Road raster
R3 : performance value of Railway raster
R4 : performance value of Power Grid raster
Rs : performance value of Settlements raster
R6 : performance value of Barren Lands raster
R7 : performance value of Grassland raster
R8 : performance value of Bushland raster
R9 : performance value of Pastures raster
Rlo : performance value of Emerging Forests raster
Rn : performance value of Farm Land raster
R12 : performance value of Forest raster
R13 : performance value of Wetland raster
R14 : performance value of Slope raster
: performance value of Orientation raster
R16 : performance value of Elevation raster
R17 : performance value of Population raster
R18 : performance value of Tourist Night raster
R19 : performance value of Power Consumption raster
R20 : performance value of Unemployment raster
Using this solar-specific ranking function, the machine calculates a
respective solar-specific
ranking score for each of the alternatives by using the relevant GIS layers in
raster mode.
Accordingly, each pixel of the specific geographic search region in the GIS
map, except for those
in the combined exclusion layer, has its own solar-specific ranking score. In
step 422, the
computed ranking score values of the pixels are compared, and at final step
423, the pixel with
the highest value is selected as the best solar-specific power plant location.
The Best Location for Solar PV Power Plant = Max (F1), for i = 1,2, ..., n
Like the solar energy shown in FIG.8, the procedure is the same for wind
energy in FIG.10. In step
424, the alternatives, which are the raster pixels in the GIS layer of wind
velocity, are determined
and prepared in raster mode for the computations. In step 425, criteria and
sub-criteria of
effective parameters for establishing wind energy power plant as shown in the
FIG.11 are listed
for future calculations:
There are five main criteria and 19 sub-criteria for wind energy power plant.
The whole process
of determining the best location to build wind energy is similar to solar
energy with different
parameters, weights and values. The sub-criteria are the same, but the main
criterion is different,
which is given below. Steps 426-431 are the equivalent of above-described
steps 418-423 of
FIG.8.
Main criteria pairwise comparison matrix with local weights
Land
Main-Criteria VelocityWind Spatial Geomorphology Economic Weight
Cover
Date Recue/Date Received 2023-03-08
Wind Velocity an au a13 a14 a15 14/1
Spatial a21 a22 a23 a24 a25 W2
Land Cover a31 a32 a33 am. a35 N/3
Geomorphology a41 a42 a43 a44 a45 W4
Economic a41 a52 a53 a54 a55 1475
Like the solar and wind energy shown in FIG.8 and FIG.10, the procedure is the
same for
hydroelectric energy in FIG.12. In step 432, the alternatives, which are the
raster pixels in the GIS
layer of rivers average flows, are determined and prepared in raster mode for
the computations.
In step 433, criteria and sub-criteria of effective parameters for
establishing hydroelectric energy
power plant as shown in the FIG.13 are listed for future calculations:
There are five main criteria and 19 sub-criteria for hydroelectric energy
power plant. The whole
process of determining the best location to build hydroelectric energy is
similar to solar and wind
energy with different parameters, weights and values. The sub-criteria are the
same, but the main
criterion is different, which is given below. Steps 434-439 are the equivalent
of above-described
steps 418-423 of FIG.8.
Main criteria pairwise comparison matrix with local weights
Rivers
Main-Criteria Flows Spatial Land Cover Geomorphology Economic Weight
Rivers Flows an a12 a13 a14 a15 14/1
Spatial a21 a22 a23 a24 a25 w2
Land Cover a31 a32 a33 a34 a35 41/3
Geomorphology a41 a42 a43 a44 a45 14/4
Economic a41 a52 a53 a54 a55 w5
Like the other energy types that mentioned in FIG.8, FIG.10 and FIG.12, the
procedure is the same
for biomass energy in FIG.14. In step 440. In step 441, criteria and sub-
criteria of effective
parameters for establishing biomass energy power plant as shown in the FIG.15
are listed for
future calculations.
There are four main criteria and 19 sub-criteria for biomass energy power
plant. The whole
process of determining the best location to build biomass energy is similar to
the energy types
examined before. With different parameters, weights and values. The sub-
criteria are the same,
but the main criterion is different, which is given below. Steps 442-447 are
the equivalent of
above-described steps 418-423 of FIG.8.
31
Date Recue/Date Received 2023-03-08
Main criteria pairwise comparison matrix with local weights
Main-Criteria Spatial Land Cover Geomorphology Economic
Weight
Spatial a11 alz a13 a14
Land Cover a21 a22 a23 a24 14/2
Geomorphology a31 a32 a33 a34 W3
Economic a41 a42 a4.3 a44 1414
Referring back to Figure 7, the power specific MCDM sub-processes 410-413 are
now complete,
from which the machine has separately determined a respective four best
geographic locations
for solar, wind, hydropower and biomass power plants. In step 414, the ranking
scores of these
four best geographic locations from the power specific MCDM sub-processes in
FIG.8, FIG.10,
FIG.12 and FIG.14, are compared. This comparison is possible because the sub-
criteria of the four
energy-specific sub-processes are the same as one another, and the values of
the main criteria
are normalized. Finally, in step 415 machine identifies which of these four
best energy-specific
locations has the best ranking score, and selects this location and plant type
as the optimal
location and power plant type to display to the user, and/or store in memory
for future retrieval
and display.
This can be expressed as follows:
Ls = Max (F1), for i = 1,2, ... , n
= Max (13, for i = 1, 2, ..., n
Lh = Max (F1), for = 1, 2, n
Lb = Max (F1), for i = 1, 2, ..., n
Lail = Max (Ls, Lw, Lh,Lb)
Where Ls is the best location for solar PV power plant, Lw is the best
location for wind power
plant, Lh is the best location for hydropower power plant, Lb is the best
location for biomass
power plant and Lath is the final optimal and energy-specific location.
Turning now from the best-location determination to another novel aspect of
the invention,
FIG.16 illustrates a series of computations executed by the machine at step
124 of FIG.1B if mining
of cryptocurrency by produced power was specified by the user. Mining of
cryptocurrency
requires electricity for to operate the mining equipment and cooling systems.
of the machine
surveys the user about potential mining cryptocurrency with electricity
generated from
renewable sources, and can also determine the required mining equipment,
cooling systems and
their costs. As with the resultant data from all other processes described
herein, the
computational results from this cryptocurrency mining evaluation be used to
populate a final
32
Date Recue/Date Received 2023-03-08
feasibility study. All of these steps are complex and required accurate
computation. To calculate
the amount of cryptocurrency that can be generated, the updated price of the
cryptocurrency is
received through an API. The user can decide how much electricity should be
converted to
cryptocurrency, and the machine performs the relevant calculations. In step
500, the user is
queried as to whether available power should be dedicated solely to
cryptocurrency mining, or to
a combination of cryptocurrency mining and other general consumption. This
amount of
available power may be the same user-specified power capacity inputted at step
103 of FIGIA or
step 201 of FIG. 2A, or the budget-constrained capacity determined at step 301
of FIG.3. If the
user decides to convert some of the available power to cryptocurrency and a
balance of the
available for general consumption, the process continues to step 501.
Alternatively, if the use
opts to convert all of the available power to cryptocurrency mining, the
process bypasses step
501 and proceeds directly to step 502. In step 501, the machine determines the
relative split of
available power between cryptocurrency mining and general consumption based on
user choice,
for example based on a percentage split or specified wattage. The equations of
this part are given
below:
Wer = Wn(Scr)
Sõ = 100 ¨ Sco
Where:
Wcr = Amount of power conversion to cryptocurrency (W)
Scr = Share of power conversion to cryptocurrency (%)
Sco Share of power for general consumption (%)
In step 502, the machine computes and determines cryptocurrency mining
equipment
requirements from a mining equipment database, which may or may not be
embodied by, or
found among, the one more equipment supplier databases used in FIG. 3 to
search for power
plant equipment. This section uses We,. as a key for finding required
equipment by using the
following algorithm:
INITIALIZE;
FOR i in n
CREATE a list of all mining equipment
END LOOP
FOR i in [mining equipment quantity]
COMPARE Watts of each mining equipment (Wei) with Wm. :
33
Date Recue/Date Received 2023-03-08
IF We, = Wei
RETURN Wei
ELSE IF We, < Wei
RETURN min(Wei)
ELSE IF We, > Wei
RETURN max(Wei)*Quantity Wei
END IF
END LOOP
In step 503, the machine calculates cryptocurrency mining profit based on
selected mining
equipment and an electricity price that is calculated by a function by using
an electricity price API,
that may also be used in the earlier processes described above. These
calculations make use of a
"total network hash rate" that is obtained by the API, and a "mining device
hash rate",
"block reward", and "block time" that are in the mining equipment database.
The equation of
cryptocurrency mining profit is given below:
Cryptocurrency Mining Profit = 3600 hPb card Wcardcpow er
i'nett block 1000 ( $ hour)
Where:
P = Price of cryptocurrency
(seconds)
b = Block reward
device as rate ( hashes \
hcard k
= Mining dice hash (seconds)
'hashes \
hnet = Total network hash rate
(seconds)
(second\
tb lock = Block time
k block )
(hour)
P i mining i = Power consumption of device
Wcard
(
= Price of produced power kw h)
Cpower
34
Date Recue/Date Received 2023-03-08
In step 504, the machine calculates the cost of all cryptocurrency mining
equipment based on the
selected items in the mining equipment database, and then machine computes
required cooling
system componentry for a mining system composed of that selected mining
equipment, and
calculates the added costs of that cooling system componentry. For calculating
required cooling
system, the following equation may be used:
( BTU ) = 1 3.4129 (Watts)
\hour(hour) BTU
____ = 1 1/12000 (Tonne)
0.2844Kard
Tõ = 1000 (Tonne)
Where:
Tõ = Tonne of generated heat by mining device (Tonne)
(hour)
Wcard Power consumption of mining device
k
Then machine calculates required cooling system costs by its capability for
removing generated
heat of the mining equipment. And finally, the machine calculates the total
cost of mining and
cooling equipment that was selected from the mining equipment database by
using the following
equation:
Costtotat = Costegaipment + ICostcooting
Where:
COStequipment Costs of each equipment
COSt
-equipment = Costs of each cooling devices
In step 505, the machine sends the calculated cryptocurrency financial data
for using in feasibility
study calculations. Such data preferably includes at least anticipated revenue
from mining
cryptocurrency, and the cost the proposed mining and cooling systems to be
assembled from the
selected mining equipment and cooling componentry.
Date Recue/Date Received 2023-03-08
FIG.17 illustrates a power management tool implemented via executable software
of the
machine. This feature helps users to determine the best time for saving
produced power, selling
the produced power to a power network, or consuming the produced power. For
this purpose,
the machine employs statistical data concerning the cost of electricity, and
by using an iteration
algorithm, the machine predicts the best times for storage/sale/consumption of
power, and
displays the determined scheduling to the user. The machine takes the type of
renewable energy
concerned and its capacity, power demand, produced power, type and
specifications of the
storage battery/batteries that are used for saving power for determined
storage periods, related
statistical data from one or more of the internal databases (e.g. OLAP
database 1005) and then
solves the optimal dispatch problem using a reinforcement learning approach
and a dynamic
programming algorithm. Experience has shown that the value iteration algorithm
finds optimum
answer by repeating possible answers that converges the problem to a good
optimum solution.
The machine uses a modified value iteration algorithm to solve this optimal
dispatch problem and
makes the best possible decision for storing/selling/consuming produced
electricity.
In the following non-limiting example, starting at step 600 of FIG.17, a
photovoltaic solar power
plant is considered. At any given hour (h) the value of energy produced by the
power plant is
obtained by following equation:
C * (P ¨ F)
Where, C is output in MW, P is price of electricity in $/MWh and F is
photovoltaic power plant's
hourly operating costs. In step 601, the machine assumes that the power plant
unit does not have
any operating constraints, in which case the optimal dispatch problem
represented in the
following form:
(Selling where P > F
C =
Saving where P <F
If the price of electricity in $/MWh is greater than the photovoltaic solar
power plant's current
operating costs per hour, the machine proceeds to step 602, and records a
general profit-
motivated preference for sale of generated power, but if the price of
electricity in $/MWh is less
than the photovoltaic solar power plant's current operating costs per hour,
the machine instead
proceeds to step 603, and records a general profit-motivated preference for
sale of generated
power for storage of generated power.
From either of step 602 or 603, the process proceeds to step 604, where the
value of generated
power is computed using the following equation:
Unit Capacity *1 Max( P(h) ¨ F (h) , 0
36
Date Recue/Date Received 2023-03-08
However, in reality the power plant has many operating constraints which make
the generation
optimal dispatch very challenging. Here are some of the constraints:
There is variable operating management cost which we call vom.
There is a limitation for saving power in the batteries
So in step 605, final income can be found by the following equation:
R(h) = P(h) ¨ F(h)¨ Vom
Where, P(h) is power price at hour "h", F(h) is photovoltaic power plant
current costs per hour
and Vom is variable operating and management cost.
Then, the machine computes the decision d(h) per hour in the step 606 by the
following form:
1 if unit is in selling mode
d(h) =
0 if unit is in saving mode
Where, d(h) is decision of power unit condition. Therefore, number of
switching is:
Ild(h + 1) ¨d(h)I
If d(h) = 1 machine decides for selling and goes to step 608. If d(h) = 0
machine decides for
saving and goes to step 607.
Then in step 609 machine defines optimal dispatch problem as the following
equation:
Max Id(h)R(h)¨ Switching Cost *11d(h + 1) ¨ d(h)I
where:
IW(h + 1) ¨ d(h)I < Max Switching
and:
id(1)
d(2)
D = , d(h) E (0,1)
\d (N)J
Where feasible switching space is the space of all acceptable switching, and
min runtime and
downtime are not violated.
37
Date Recue/Date Received 2023-03-08
The term "Dynamic Programming" (DP) refers to a collection of algorithms that
can be used to
compute optimal policies given a perfect model of environment as a Markov
decision process
(MDP). Classical DP algorithm are of limited utility in reinforcement learning
both because of their
assumption of a perfect model and because of their great computational
expense, it is assumed
that the environment is a finite MDP, i.e. that its state and action sets, S
and A(s), for s E S. are
finite, and that its dynamics are given by a set of transition probabilities
as the following form:
Pa(s, s') = Pr(st+i = s' 1st = s, at = a)
Expected immediate rewards is obtained by the following form:
Ra(s, s') = Efrt+ilat = a, st = s, s1 = s'}, for all sE S,a E A(s)
The key idea of DP is the use of value function to organize and structure the
search for good
policies. Optimal policies can be easily attained once the optimal value
function ( V*) is found,
which satisfy the bellman optimality equation:
V* (s) = maxE(rt+a + yr (st+Dist = s, at = a}
a
= max Pa(s, s9[Ra(s, s') + yV*(s)] for all seS,ae
A(s)
a
s'
An approach to solve dynamic programming problem is value iteration algorithm.
Value iteration
is similar to backward dynamic programming for finite horizon problems.
The basic version of value iteration algorithm is given as follow:
initialize V(s) arbitrarily
loop until policy good enough
loop for S E S
loop for a E A
Q (s, a) := R(s, a) + y Es, Es T (s, a, s') V(e)
V (s) := max Q (s, a)
a
end loop
end loop
Value iteration algorithm is similar to the backward dynamic programming
algorithm. Rather than
using a sub script t, which is decremented from T back to 0, an iteration
counter n is used that
starts at 0 and increases until we satisfy a convergence criterion. The
decision for condition of the
generators is a decision between "turn on" and "turn off", and this decision
depends on the
decision we made prior to that hour. For example, the generator is turned on
at hour "X" with
"U" as minimum runtime, then no decision can be made at hour X+1, X+2,..., X+U-
1. Due to
minimum runtime constraint, the same is true when the generator is turned off:
assuming "0"
represents minimum downtime, no decision can be made in hour X+1, X+2,..., X+0-
1. Therefore,
the state of generator is defined as the following equation:
38
Date Recue/Date Received 2023-03-08
d(h))
S(h) = (DT
UT
where:
1 if generator is on at hour h
d(h) =
0 if generator is off at hour h
and, DT(down time) E (1,2, == , 0) means number of consecutive hours that unit
was off before
hours, and UT(up time) e (1)2, === U) means number of consecutive hours that
unit was on
before hours, therefore the value function for any given state is defined as
the following equation:
17(.5(h))
And number of states is equal to:
2xUx0xHxS
Where, 2 represent decision (), U is uptime, 0 is downtime, H is number of
hours and S is
1
maximum number of switching. To approximate the value function as following
equation:
17(5(h)) = V (h, d (h), DT , UT)
Using ff (h) were:
d E [OM
(h) = M ax (V (h, d(h), DT, UT)), DT E (1,2, , 0)
UT e (1,2, U)
Now if V (h) is known for all hours, decision vector D can be calculated as:
\
D= d(.2)
\d(N)/
If d(h ¨1) = 1 then:
(h) = Max (i7 (h + 1) + R(h),V (h + 0))
and if d(h¨ 1) = 0 then:
39
Date Recue/Date Received 2023-03-08
u-i
V (h) = Max (v (h +1) ,1R(h + i)17- (h + U) The dynamic programming problem is
solved in environment of Markov decision process using
value iteration algorithm. Therefore, value iteration algorithm for optimal
dispatch problem can
be defined as follow:
Initialization
1 : Initialize V (h) for all h
2 : set d(t)=1 for tE{1,...,U}
And h 4- U+1
If Vi_i R(i) +17 (U + 1) > 17 (2)
Otherwise
Set d(t)=0
And h 2
If d(h-1)=1 then
Set d(t)=0 for tE{h,h+1,...,h+0-1}
If V (h + 0) > V (h + 1) + R(h)
And set h 4- h+0
Otherwise set d(h)=1
And h h+1
If d(h-1)=0 then
Set d(t)=1 for tE{h,h+1,...,h+U-1}
If 17 (h + U) +Erol R(h + 0 > 17 (h + 1)
And set h h+U
Otherwise set d(h)=0
And h 4- h+1
Update Value Function using V (h) = n_hD(i)R(i)
V given decision vector D=(61(1))
d(N)
Set d(0) = 0 an go to step (1)
If updated -17 is different from prior 17
By solving this problem, the machine can prepare a time table and schedule for
managing
produced power. This schedule contains time to consume, store or sell produced
power, this time
table can be used by one or more control devices of an existing and
operational power plant that
manages the dispatch of electricity to the storage battery/batteries, to the
power network or to
power consuming loads based on input commands from the machine.
In step 610, the machine prepares read only raw data into a human readable
schedule/time-table
for users, for their informed knowledge of the best time for using, saving or
selling electricity, and
may simultaneously prepare a command signal schedule for automated operation
of the control
device(s) 611 for practical usage of this scheduling data.
Date Recue/Date Received 2023-03-08
In step 612, using the generated schedule data, the control device(s) is/are
operated according to
the transmitted control signals at any given time, on the computed schedule
data.
If the decision at any given time is to saving produced power in the storage
batteries, in step 613
control device saves power, and again this power storage period continues
until a decision to
instead consume the produced power is made. If such decision to consume
produced power
occurs, then in step 614, the control device(s) dispatch (es) produced power
for consumption by
electrical loads. And finally, if the decision to sell produced power occurs,
then in step 615, the
control device(s) dispatch (es) produced power for selling to the network.
FIG.18 illustrates a computer-implemented method for finding an optimal
location for
establishing an electric/hybrid vehicle charging station, which can be
executed using similar
methodology to that disclosed above for finding the optimal location for a
renewable energy
power plant.
In view of increasing greenhouse gas emissions, use of electric/hybrid
vehicles has become more
popular as an appropriate way to reducing greenhouse gas emissions and air
pollution.
Nevertheless, a crucial consideration for the wide adoption of electric
vehicles is charging stations.
Finding the best location for establishing electric car charging station is a
tough problem and
depends on numerous parameters. This problem resembles the foregoing problem
of finding the
best location for establishing renewable energy power plants, as addressed
above with reference
to FIG.4 to FIG.15. The strategy employed herein for establishing electric car
charging stations is
based on renewable energy sources, particularly solar energy using
photovoltaic solutions. There
are plenty of important parameters such as: Spatial, Geomorphology and
Economic that have high
impact on final result, power production and customer base when considering
establishment of
electric vehicle charging stations. In the detailed embodiment that follows,
these parameters and
their effects on finding the best location for electric vehicle charging
stations are computed.
In step 700 of FIG. 18, the machine collects input from the data, which may
include at least a
subset of the same inputted data described above in relation to FIG.1, and in
the illustrated
example includes user-specification of a geographic search region within which
the user wants to
find an optimal geographic location for a charging station. in step 701, the
alternatives, which are
the raster pixels in the GIS layer of solar irradiance (GHI) of selected area,
are determined and
prepared in raster mode for the computations. then, criteria and sub-criteria
of effective
parameters for establishing charging station, these parameters can be divided
into the categories
of: Spatial considerations, Geomorphology and Economic considerations. these
parameters
contain many items that are given in the FIG.19 and identified by the machine,
based on stored
instructions in memory, for use in subsequent steps and calculations. In this
detailed but non-
limiting example, there are four main criteria and nine sub-criteria for the
solar energy electric
car charging station. In step 702, the weights of the criteria and sub-
criteria related to the solar
energy electric car charging station criteria and sub-criteria identified in
step 701, having been
predetermined, and calculated in accordance with the Saaty method mentioned
earlier related to
finding the best location for establishing renewable power plant, either by,
or with the aid or
guidance of a knowledgeable expert, and then stored in computer-readable
memory of the
machine as normalized Saaty pairwise comparison matrices, are now retrieved
from memory for
use of this predetermined solar energy charging station weighting scheme. It
should be noted that
since the construction of a solar charging station is very similar to a solar
power plant, the process
41
Date Recue/Date Received 2023-03-08
is similar, with the difference that the amounts of weights will be different.
The sub-criteria
comparison with calculated local weight for each type of parameters are shown
below:
Spatial sub-criteria pairwise comparison matrix with local weights
Sub-Criteria Existing Charging Car Public Weight
Station Parking Transport
Existing an a12 C113 Wsi
Charging Station
Car Parking a21 a22 a23 WS2
Public Transport a31 a32 a33 ws3
Geomorphology sub-criteria pairwise comparison matrix with local weights
Sub-Criteria Slope Orientation Elevation Weight
Slope a11 a12 a" wci
Orientation a21 a22 a23 wG2
Elevation a31 a32 a33 WG3
Economic sub-criteria pairwise comparison matrix with local weights
Sub-Criteria Production Attraction Population Weight
Production an a12 a13 WEi
Attraction a21 a22 a23 WE2
Population a31 a32 a33 WE3
The predetermined expert-derived weights may be periodically updated to
account for changing
circumstances. In the weights of the sub-criteria factors, the alphabetic
subscript in the weight w
is shorthand for the the sub-criteria name. So, in the term w51 the "S"
represents "Spatial" and
the "1" represents the first factor in the matrix. Likewise, "G" in
wGi represents
"Geomorphology", "E" in wEi represents "Economic".
42
Date Recue/Date Received 2023-03-08
Main criteria pairwise comparison matrix with local weights
Main-Criteria GHI Spatial Geomorphology Economic Weight
GHI a11 a12 a13 a14 w1
Spatial a21 a22 a23 a24 W2
Geomorphology a31 a32 a33 a34
Economic a41 a42 a43 a44 W4
As mentioned earlier, due to the similarity of the construction of the solar
charging station with
the solar power plant, other MCDM calculations are similar to the previous
ones, which are given
in the descriptions of FIG.8. Steps 419, 420, 421 and 422 of FIG.8 are
included in step 703, and
similar to the same steps, the calculation operation is performed according to
the weights of this
problem.
The calculated ranking scores of the candidate locations for a solar charging
station are then
compared to identify the best candidate location for a solar charging station.
The equation of the
solar charging station-specific ranking function is given below:
F = +
W2 (Ws R2 + W52R3 +W3R4)
W4 (WG Rs WG2R6 WG 3R7)
W5(WEiR8 W E 2R9 ws3R10)
Where w1, w2, w3 and w4 are the local weights of main criteria:
: weight of GHI criterion
w2 : weight of Spatial criterion
w3 : weight of Economic criterion
: weight of Geomorphology criterion
w51, w52 and ws3 are the local weights of Spatial sub-criteria:
wsi : weight of Existing Charging Station criterion
Ws2 : weight of Car Parking criterion
ws3 : weight of Public Transport criterion
wGi, wG 2 and wG3 are the local weights of Geomorphology sub-criteria:
wGi : weight of Slope criterion
wG2 : weight of Orientation criterion
43
Date Recue/Date Received 2023-03-08
wG3 : weight of Elevation criterion
wE2 and wE3 are the local weights of Economic sub-criteria:
wEi : weight of Production criterion
wE2 : weight of Attraction criterion
wE3 : weight of Population criterion
And also, term "RI" represents normalized performance value of each
alternative:
: performance value of GHI raster
R2 : performance value of Existing Charging Station raster
R3 : performance value of Car Parking raster
R4 : performance value of Public Transport raster
R5 : performance value of Slope raster
R6 : performance value of Orientation raster
R7 : performance value of Elevation raster
R8 : performance value of Production raster
R9 : performance value of Attraction raster
: performance value of Population raster
Using this solar charging station-specific ranking function, the machine
calculates a respective
solar charging station-specific ranking score for each of the alternatives by
using the relevant GIS
layers in raster mode. Accordingly, each pixel of the specific geographic
search region in the GIS
map, except for those in the combined exclusion layer, has its own solar
charging station-specific
ranking score. These weights can be different in various situations and the
above equation will
likely or definitely change over time, and thus require period update in the
software. And finally,
in step 704, the machine identifies the optimal location for establishing an
electric/hybrid vehicle
charging station using solar energy sources according to the candidate
location with the greatest
ranking score value, and displays same to the user, and/or stores same in
memory for later
retrieval and display.
The Best Location for Solar Charging Station = Max (Fe), for i = 1,2, ...,
n
FIG,20 illustrates computer-implemented formulation of a feasibility study,
before establishing a
power plant, so that the user can estimate the financial, technological,
working capital and cash
flow projections resources that will be needed to ensure the successful
launching of the power
plant.
First, at the step 800, system will check whether a desired location has for
the power plant has
been specified by user or not. If the location has been specified by user,
then the process bypasses
step 801 and jumps ahead to step 802.If no user-desired location was
specified, then in step 801
the machine will calculate the best location using Machine learning algorithms
with the Location
Intelligence described above in relation to FIG.4 to FIG.15. At step 802, the
user's specified
location or the machine's calculated best location will be shown to user in a
map. Next, at step
44
Date Recue/Date Received 2023-03-08
804 generated diagrams are displayed such as: cash Flow Diagram, Payback
Period, etc. have been
made.
In step 805 the system shows all numeric data such as: Net Percent Value
(NPV), Payback Period,
Internal Rate of Return (IRR), Benefit-Cost Ratio (BCR), Levelized Cost of
Energy (LCOE), etc.
In step 806 some useful charts are shown. These charts are based on
Feasibility Study's data. The
machine computes Net Percent Value (NPV) by the following equation:
Ci C2 Ct
NPV = ¨Co + 1+ r+ (1+ r)2 + + __________
(1 + r)t
Where: ¨00 represents Initial Investment, C represents cash flow, r represents
discount rate and
T is time.
PPBThen the machine calculates Payback Period by the following equation:
Cost of Project (Investment)
=
Annual Cash inflows
The machine computes Internal Rate of Return (IRR) by the following equation:
NPV = ________ Ct Co
Li (1 + r)t
t-i
Where: ct is net cash inflow during the period t, r is discount rate, t is
number of time periods and
Co is total initial investment cost.
The machine computes Benefit-Cost Ratio (BCR) by the following form:
vN ICFt [Benefits]i
BCRI9"[Benefits] I 44=0 (1 +
=
IPV [Cost]I ICPt [Cost] I
t=0 (1 + it)t
Then the machine computes Levelized Cost of Energy (LCOE) by the following
equation:
vN + mt + Ft
LCOEGt=1 (1 + r)t
= _____________________
Et
r)t
Where: It is investment expenditures in the year t, Mt is operations and
maintenance
expenditures in the year t, Ft is fuel expenditures in the year t, E t is
electricity generation in the
year t, r is discount rate and n represent economic life of the system. Then,
the machine computes
annual investment depreciation by the following equation:
A =P(1 ¨ r)t
Date Recue/Date Received 2023-03-08
Where: A is amount, P is original value, r is discount rate and t is time (in
year). The machine
calculates the working capital by this form:
Working Capital = Current Assets ¨ Current Liabilities
Then the machine calculates production costs by the following form:
Production Cost
= Direct Labor + Direct Material + Overhead Costs on Manufacturing
Then the machine calculates Profitability index by this form:
Percent Value of Future Cash Flows
Profitability index = _____________________________
Initial Investment
After that, the machine computes rate of internal return by the following
equation:
NPVa
IRR = t2+ (rb ¨ ra)
NPVa ¨ NPVb
Where: ra is lower discount rate and rb is higher discount rate.
Then the machine calculates other financial data by the following equations:
Total Debt
Financial Leverage =
Shareholder'sEquity
Net Sales
Fixed Invesment =
Average Fixed Assets
Working Capital = Current Assets ¨ Current Liabilities
Production Cost = Direct Labor + Direct Materials + Overhead costs
Net Sales
Fixed Invesment =
Average Fixed Assets
Cost Price = Direct Labor + Direct Materials + Overhead costs + Marketing
Costs
+ Tools
Discounted Cash Flow = ________ Ct
LI (1 r) t
t-1
Percent Value of Future Cash Flow
Profitability index = _____________________________
Initial Invesment
46
Date Recue/Date Received 2023-03-08
In step 807 the machine displays many figures to help for constructing
Renewable power plant
such as: Process Flow Diagram (PFD) of the power plant, etc. In step 808 some
practical reports
are displayed. These reports are based on Location Data, Geographical Data,
and Feasibility Study
Data, and so on.
F1G.21 schematically illustrates the machine's automated compilation of the
feasibility study's
detailed reports. This is the main core of the feasibility study computation
that will generate
useful data to show to the users. The many inputs may include direct user data
entry, and/or
retrieval of other computed data. Among these inputs, at Section 900 machine
gets Time
planning, in Section 901 machine gets Products (Electricity), in Section 902
machine gets Currency
unit, in Section 903 machine gets Inflation rate, in Section 904 machine gets
Participants, in
Section 905 machine gets Discount rate, in Section 906 machine gets Fixed and
Initial costs, in
Section 907 machine gets Pre-Operation costs, in Section 908 machine gets
Production costs, etc.
In Section 909 Machine searches and picks required data from the internal, GIS
and NASA
databases, and optionally other data sources. In Section 910 Machine takes
these inputs and
computes feasibility study output by some Numerical Calculation methods and
gives feasibility
study outputs. In Section 911 Machine gives Financial review, In Section 912
Machine gives
Estimation of fixed investment, In Section 913 Machine gives Estimation of
working capital, In
Section 914 Machine gives Estimation of production costs, In Section 915
Machine gives
Estimation of annual investment depreciation, In Section 916 Machine gives
Estimation of the
total capital required for the project, In Section 917 Machine gives
Estimation of the cost price by
costs, In Section 918 Machine gives Determining the sources of financing the
project and its
financial cost, In Section 919 Machine gives Analysis of project revenues and
costs, In Section 920
Machine gives Determine the profit and loss performance of the plan for the
entire investment
and the stock brought, In Section 921 Machine gives Economic studies, In
Section 922 Machine
gives Determine the net cash flow of the entire investment, In Section 923
Machine gives Analysis
of discounted cash flow, In Section 924 Machine gives Internal rate of return,
In Section 925
Machine gives Determination of the net present value, In Section 926 Machine
gives
Determination of the rate and period of internal return, In Section 927
Machine gives Determining
the payback period, In Section 928 Machine gives Profitability index, In
Section 929 Machine gives
Perform project sensitivity analysis, In Section 930 Machine gives Risk
analysis, In Section 931
Machine gives Analysis of financial ratios, In Section 932 Machine gives
Preparation of financial
statements, In Section 933 Machine produces Cash flow diagram, In Section 934
Machine
produces Payback period diagram, In Section 935 Machine produces Process flow
diagram and
finally In Section 936 Machine produces required equipment.
In at least some embodiments contemplated herein, users with any level of
knowledge can plan
a renewable energy power plant by entering minimal input data, such as:
latitude and longitude
of a desired location or geographical search area, budget or electricity
generation capacity.
The described methodologies generate valuable output data, for example the
best location for
establishing the power plant (if not user-specified), the best type of energy
in that area based on
the angle of radiation, temperature, wind speed and other influential
parameters. The wide scope
of parameters cannot be calculated mentally or manually, and at least some of
them are extracted
from NASA and GIS databases that without an effective hardware and software
combination,
47
Date Recue/Date Received 2023-03-08
cannot be extracted and used in a meaningful way to fulfill the same ends as
the disclosed
invention.
In particularly preferable embodiments, after calculating the effective
parameters for selection of
a type of energy and geographical location, the system begins to calculate the
required equipment
with the best rate based on the type of energy and the amount of energy
production needed, or
attainable within a user-allotted budget. Again, this step cannot be done
manually to any degree
of comparable scope and efficiency as the machine automated solution proposed
herein. Through
a unique web service implemented by the equipment suppliers, the equipment
data and
characteristics are made available to the system, and the best equipment can
be found based on
the customer's budget and equipment efficiency. This selection may be based on
machine
learning so that the system selects the best choice based on the previous
data.
In preferred embodiments, the machine also calculates the amount of energy
that can be
produced, which is done based on artificial intelligence mechanisms, and then
an economic
feasibility study can be performed, which includes calculations of IRR, NPV,
NCF and diagrams. It
is related to cash flow and profitability in the coming years.
Preferred embodiments also implement optimal power management of consumption
time, to
manage consumption of produced power to achieve the best economic benefits.
This feature
offers the best time for consuming power, saving power in storage batteries or
selling power to
the network, and also it can connect with a power management device and send
commands for
optimally managing produced power.
Preferred embodiments also include calculation of the benefits of converting
the produced power
into a cryptocurrency by considering the cryptocurrency equipment requirements
for such
capability.
In at least one embodiment, the best locations to build charging stations for
electric vehicles is
calculated on the basis of renewable energy, with consideration to other
parameters such as
urban details and cost-effective renewable energy based on geographical
features.
Since various modifications can be made in the invention as herein above
described, and many
apparently widely different embodiments of same made, it is intended that all
matter contained
in the foregoing specification shall be interpreted as illustrative only and
not in a limiting sense.
48
Date Recue/Date Received 2023-03-08
