CN115759475A - Data-driven multi-region power system economic environment scheduling method and system - Google Patents
Data-driven multi-region power system economic environment scheduling method and system Download PDFInfo
- Publication number
- CN115759475A CN115759475A CN202211579274.1A CN202211579274A CN115759475A CN 115759475 A CN115759475 A CN 115759475A CN 202211579274 A CN202211579274 A CN 202211579274A CN 115759475 A CN115759475 A CN 115759475A
- Authority
- CN
- China
- Prior art keywords
- power system
- target
- economic environment
- region power
- optimization
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000005457 optimization Methods 0.000 claims abstract description 80
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 25
- 241001206881 Myrmeleon inconspicuus Species 0.000 claims abstract description 23
- 239000003795 chemical substances by application Substances 0.000 claims abstract description 23
- 239000002803 fossil fuel Substances 0.000 claims abstract description 17
- 239000003344 environmental pollutant Substances 0.000 claims abstract description 10
- 231100000719 pollutant Toxicity 0.000 claims abstract description 10
- 238000003860 storage Methods 0.000 claims abstract description 6
- 230000006870 function Effects 0.000 claims description 30
- 230000005540 biological transmission Effects 0.000 claims description 12
- 230000014759 maintenance of location Effects 0.000 claims description 11
- 238000010248 power generation Methods 0.000 claims description 11
- 239000002245 particle Substances 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 7
- 238000012546 transfer Methods 0.000 claims description 7
- 241000282320 Panthera leo Species 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 6
- 238000004088 simulation Methods 0.000 description 6
- 238000004590 computer program Methods 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000003912 environmental pollution Methods 0.000 description 4
- 239000000446 fuel Substances 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 238000013526 transfer learning Methods 0.000 description 4
- 238000007477 logistic regression Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000012850 discrimination method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000009827 uniform distribution Methods 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a data-driven multi-region power system economic environment scheduling method, a data-driven multi-region power system economic environment scheduling system, electronic equipment and a computer-readable storage medium, and belongs to the technical field of power systems. The method comprises the steps of constructing a multi-region power system economic environment scheduling multi-objective optimization model considering multiple constraint conditions by taking the minimization of pollutant gas emission and the cost of fossil fuel as targets; according to the target and the multiple constraint conditions, a data-driven agent model is constructed to convert the multi-region power system economic environment scheduling multi-target optimization model, and the converted multi-region power system economic environment scheduling multi-target optimization model is solved through a multi-target ant lion algorithm; and obtaining a global optimal solution according to the solving result. Meanwhile, the convergence rate and accuracy of the optimization algorithm are improved, the operation time is reduced as much as possible, and the problem that the dimensionality of the economic emission scheduling problem of the multi-region power system is increased, so that a large amount of computing resources are occupied and a large amount of computing time is consumed is solved.
Description
Technical Field
The application relates to the technical field of power systems, in particular to a data-driven economic environment scheduling method and system for a multi-region power system.
Background
The statements in this section merely provide background information related to the present disclosure and may not constitute prior art.
With the expansion of power systems and the distribution of load centers, it is important to establish a multi-region power system with a plurality of load centers interconnected for safe and stable operation of the power system. Many countries give the power industry more autonomy to operations, and this situation effectively assists in economic operations and emissions reduction. Based on the above two factors, the multi-region power system economic dispatch can provide higher stability and economy for the power system. As an electric power industry with the most emission of polluted gas, the method is a significant technical problem of great research on how to consider environmental and economic benefits on the premise of ensuring that the power supply quality is not affected, and is also a key content for promoting energy conservation and emission reduction of electric power enterprises. The problem is preliminarily researched at present, and the main research thought is as follows: the method comprises the following steps of (1) simultaneously considering the power generation cost and the pollution gas emission problem of a thermal power generating unit, (2) considering different constraint conditions in a multi-region power generation and scheduling process, and (3) selecting a proper algorithm to analyze and solve on the basis of meeting the scheduling time requirement.
However, due to the expansion of the grid scale and the decentralization of the load center, the dimensionality of the economic emission scheduling problem of the multi-region power system is increased, so that a large amount of computing resources are occupied, and a large amount of computing time is consumed.
Therefore, how to make the optimal scheduling strategy more conform to the economic emission scheduling problem of the multi-region power system, that is, to improve the convergence speed and accuracy of the optimization algorithm and reduce the operation time as much as possible, is a technical problem that needs to be solved urgently by those skilled in the art at present.
Disclosure of Invention
In order to solve the defects of the prior art, the application provides a data-driven multi-region power system economic environment scheduling method, a data-driven multi-region power system economic environment scheduling system, an electronic device and a computer-readable storage medium, aiming at any multi-region power grid topology, after determining structural parameters, taking the aim of minimizing the power generation cost and minimizing the environmental pollution at the same time, considering the actual power grid operation flow, improving the convergence speed and accuracy of an optimization algorithm and reducing the operation time as far as possible.
In a first aspect, the application provides a data-driven multi-region power system economic environment scheduling method;
the data-driven multi-region power system economic environment scheduling method comprises the following steps:
constructing a multi-region power system economic environment scheduling multi-objective optimization model considering multiple constraint conditions by taking the minimum pollutant gas emission and the fossil fuel cost as targets;
according to the target and the multiple constraint conditions, a data-driven agent model is constructed to convert the multi-region power system economic environment scheduling multi-target optimization model, and the converted multi-region power system economic environment scheduling multi-target optimization model is solved through a multi-target ant lion algorithm;
and according to the solving result, obtaining a global optimal solution as a decision basis for optimizing and scheduling each region of the power system in the current period.
Further, the multi-objective optimization model for economic environment scheduling of the multi-zone power system is expressed as follows:
Minimize [F(P),E(P)],
Subject to:g i (P)＝0,i＝1,…,M 1 ，
h j (P)≤0,j＝1,…,M 2 ,
wherein F (P) is an optimization objective function of fossil fuel cost, E (P) is an optimization objective function of pollutant gas emission, g i (P) is the inequality constraint involved, h j (P) is the equation constraint involved.
Further, the multiple constraint conditions comprise generator active upper and lower limit constraints, power balance constraints, node voltage amplitude constraints, line power flow constraints, transmission line safety constraints, regional hot standby transfer constraints and line loss constraints.
Preferably, the active power upper and lower limit constraints of the generator are expressed as:
the power balance constraint is expressed as:
the node voltage magnitude constraint is expressed as:
the line flow constraint is expressed as:
the transmission line safety constraints are expressed as:
the regional hot standby transfer constraint is expressed as:
the line loss constraint is expressed as:
further, the specific steps of constructing the data-driven agent model to convert the multi-region power system economic environment scheduling multi-objective optimization model comprise:
solving a plurality of feasible solutions meeting the constraint conditions according to the multiple constraint conditions; calculating an optimization target value corresponding to the feasible solution according to the feasible solution and the multi-objective optimization model for the economic environment scheduling of the multi-zone power system;
and according to the feasible solution and the corresponding optimization target value, constructing a data-driven agent model to replace the multi-objective optimization model for the economic environment scheduling of the multi-region power system.
Further, the concrete steps of solving the transformed multi-region power system economic environment scheduling multi-target optimization model through the multi-target ant lion algorithm comprise:
initializing ant populations and ant lion populations, and randomly initializing the power generation capacity of each generator in each area;
assigning the ant population and the ant lion population as the corresponding generated energy of each generator in each area which is initialized randomly, and solving the non-dominated solution of each optimization target;
traversing all individuals in the population, and calculating a non-dominant solution according to an individual optimization result; when all individuals in the population are traversed, executing a single-dimension retention strategy operation to update the ant population; the non-dominated solutions are sorted according to the initialization value.
Further, the specific step of executing the one-dimensional retention policy operation to update the ant population includes:
dividing the ant population into four parts according to the fitness value of the optimization target;
finding the minimum fitness value of the optimization target in the population, and respectively executing position moving operation on the four parts of ant populations to obtain intermediate position variables;
acquiring a new population position according to the intermediate position variable;
and executing a single-dimension optimal retention strategy, and judging whether to update the dimension of the ion or not according to each dimension of each particle.
Further, the specific step of obtaining the global optimal solution according to the result of the solution includes:
aiming at an optimization objective function of the pollution gas emission and the fossil fuel cost, calculating a membership function value corresponding to a non-dominated solution of the optimization objective function through a membership function;
and carrying out regularization processing on the membership function value and solving to obtain a global optimal solution.
In a second aspect, the application provides a data-driven economic environment scheduling system for a multi-region power system;
data drive-based multi-region power system economic environment scheduling system includes:
a model building module configured to: constructing a multi-region power system economic environment scheduling multi-objective optimization model considering multiple constraint conditions by taking the minimum pollution gas emission and fossil fuel cost as targets;
a solving module configured to: according to the target and the constraint condition, a data-driven agent model is constructed to convert the multi-region power system economic environment scheduling multi-target optimization model, and the converted multi-region power system economic environment scheduling multi-target optimization model is solved through a multi-target ant lion algorithm;
a decision module configured to: and according to the solving result, obtaining a global optimal solution as a decision basis for optimizing and scheduling each region of the power system in the current period.
In a third aspect, the present application provides an electronic device;
an electronic device comprises a memory, a processor and computer instructions stored on the memory and run on the processor, wherein the computer instructions are run by the processor to complete the steps of the data-driven-based multi-region power system economic environment scheduling method.
In a fourth aspect, the present application provides a computer-readable storage medium;
a computer readable storage medium for storing computer instructions, which when executed by a processor, perform the steps of the above-mentioned data-driven multi-region power system economic environment scheduling method.
Compared with the prior art, the beneficial effects of this application are:
1. the application provides a data-driven multi-region power system economic environment scheduling method based on data driving, transfer learning and ant lion algorithm, which minimizes the power generation cost of a generator set on the premise of meeting the power load, minimizes the emission of pollution gas, reduces the use of fossil fuel to the maximum extent, greatly reduces the operation time and can provide decision basis for power enterprises in time;
2. according to the method, the line power flow constraint and the maximum transmission line constraint are introduced into the constraint conditions, and the running condition of the actual power grid is combined, so that the optimized scheduling strategy has practical guiding significance;
3. the method is based on a deep belief network model, a target function is reconstructed by using a data-driven agent auxiliary method, and the original expensive target function is replaced by a fast-calculated deep belief network agent regression model based on data driving; based on a pre-training-fine-tuning transfer learning method, pre-training and fine-tuning a network part of a deep belief network, and quickly establishing agent models of other regions by using network information of the agent model of one region; on the basis of a multi-objective ant lion algorithm, the defects brought by the traditional weight and method are overcome, meanwhile, a single-dimensional optimal retention strategy is introduced for adapting to a multi-objective optimization method, the convergence, uniformity and ductility of the solution at the front edge of pareto in the iteration period are improved, the membership value of all non-dominated solutions is obtained, and the non-dominated solution with the maximum membership value is the decision basis of the optimal scheduling in the period.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the application and, together with the description, serve to explain the application and are not intended to limit the application.
Fig. 1 is a schematic flowchart of an economic environment scheduling system of a multi-zone power system based on data driving according to an embodiment of the present disclosure;
fig. 2 is a schematic flow chart of a multi-target ant lion algorithm provided in the embodiment of the present application;
fig. 3 is a schematic structural diagram of a system for testing forty generators in four areas of a simulation object according to an embodiment of the present application;
fig. 4 is a schematic diagram of pareto optimal leading edges generated under all constraints considered for a four-region forty-generator test system (with a load of 10500 MW) provided by an embodiment of the present application.
Detailed Description
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present application. As used herein, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise, and it should be understood that the terms "comprises" and "comprising", and any variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
The embodiments and features of the embodiments of the present invention may be combined with each other without conflict.
Interpretation of terms:
fitness value: calculating a value of an optimization objective function;
non-dominated solution: and (4) unit output decision, namely the active power output by each generator in each area.
Example one
In the prior art, due to expansion of the scale of a power grid and decentralization of a load center, the dimensionality of the economic emission scheduling problem of a multi-region power system is increased, a large amount of computing resources are occupied, and a large amount of computing time is consumed; therefore, the application provides a data-driven multi-region power system economic environment scheduling method.
The data-driven-based economic environment scheduling method for the multi-region power system has the following ideas: aiming at any multi-region power grid topology, after determining structural parameters, taking the minimization of power generation cost and the minimization of environmental pollution as targets, and considering the actual power grid operation trend, firstly, replacing an original multi-region power generation cost model and an original multi-region environmental pollution model by using a data-driven deep belief network regression model, secondly, using a transfer learning method based on pre-training and fine-tuning so as to realize the quick establishment of a proxy model between different regions, and finally, applying an improved multi-target ant lion algorithm to carry out iterative computation to obtain a group of non-dominated solutions of each region, forming pareto optimal leading edges of two optimization targets of power generation cost-environmental pollution of different regions, and then obtaining an optimal scheduling decision basis of the period on a group of single regions by using a membership function method.
Next, the method for scheduling economic environment of multi-region power system based on data driving disclosed in this embodiment will be described in detail with reference to fig. 1 to 4. The data-driven multi-region power system economic environment scheduling method comprises the following steps:
s1, determining an optimization target and constraint conditions, and constructing a multi-region power system economic environment scheduling multi-target optimization model; specifically, aiming at simultaneously minimizing pollutant gas emission and fossil fuel cost, generator active upper and lower limit constraints, power balance constraints, node voltage amplitude constraints, line power flow constraints, transmission line safety constraints, regional hot standby transfer constraints and line loss constraints are selected as constraint conditions, and a multi-region power system economic environment scheduling multi-objective optimization model is constructed.
Illustratively, the optimization objective 1 is the minimum cost of fossil fuel, and the optimization objective function of the cost of fossil fuel is:
wherein, a ij 、b ij 、c ij 、d ij 、e ij For the jth generator fuel cost coefficient in the ith zone, N and N g Indicating the number of zones participating in the scheduling and the number of generators, P ij The active power output by the jth generator of the ith area.
The optimization objective 2 is that the emission amount of the polluted gas is minimum, and the optimization objective function of the emission of the polluted gas is as follows:
wherein alpha is ij 、β ij 、γ ij 、ε ij 、λ ij To the discharge coefficient of the polluted gas of the jth power generator in the ith area, P ij Active power output by jth generator for ith area, N and N g Indicating the number of zones participating in the dispatch and the number of generators.
The active upper and lower line constraints of the generator are expressed as:
wherein S is ti Represents the power on the ith line and,
representing the maximum transmission power, N, allowed on the ith line line Representing the number of lines.
The power balance constraint is expressed as:
wherein, P di Denotes the i-th zone load, P lossi Represents the i-th area line loss, T ip Representing the transferred power of the tie between the ith and pth zones.
The node voltage magnitude constraint is expressed as:
wherein,
and
respectively representing minimum and maximum allowed node voltagesVoltage, V i Is the voltage magnitude at node i.
The line flow constraint is expressed as:
wherein S is ti Represents the power on the ith line and,
represents the maximum transmission power allowed on the ith line and Nline represents the number of lines. The transmission line safety constraints are expressed as:
wherein,
representing the maximum power that can be safely transmitted between the ith and pth zones.
The regional hot standby transfer constraint is expressed as:
wherein S is pj Is the hot standby power reserved by the jth generator in the pth area; s p,req Is the hot standby power, RC, required for the p-th zone ip Is the transferred hot standby power for the ith zone and the qth zone.
All the above constraint conditions are inequality constraints, and in this embodiment, equality constraints are also considered, where the equality constraints are power balance constraints:
the line loss is obtained by solving the following tidal current equation:
further, the line loss
Wherein G is ij Is the conductance of the branch connecting nodes i and j, θ ij Is the voltage phase difference of nodes i and j, V j Is the voltage amplitude of node j, P loss For line loss, P i Active power, Q, for the output of the ith generator i And the reactive power output by the ith generator.
In summary, the multi-objective optimization model for economic environment scheduling of the multi-zone power system in the embodiment may be expressed as follows:
Minimize [F(P),E(P)],
Subject to:g i (P)＝0,i＝1,…,M 1 ，
h j (P)≤0,j＝1,…,M 2 ,
wherein F (P) and E (P) respectively represent optimization goal 1 and optimization goals 2,g (P) and h (P) are respectively related equality and inequality constraints, and M 1 、M 2 The number of equality and inequality constraints, respectively.
S2, according to the target and the multiple constraint conditions, a data-driven agent model is constructed to replace a multi-region power system economic environment scheduling multi-target optimization model; specifically, a data-driven proxy model based on a deep belief network is constructed to replace the original two objective functions by using a data-driven proxy auxiliary technology and a feasible solution of a known optimization function. The method comprises the following specific steps:
s201, solving a plurality of possible solutions meeting the constraint conditions according to the constraint conditions.
S202, constructing a data-driven agent model based on the deep belief network to replace an optimization target model by using the solved possible solutions and the calculated optimization target value, wherein the model is divided into a deep belief network part, a logistic regression part and a neural network part.
And S203, calculating a fitness value of data used for training the multi-region power system economic environment scheduling multi-objective optimization model and a fitness value calculated by using the data-driven agent model.
S204, calculating cosine similarity of the two groups of fitness values, if the cosine similarity is larger than 0.9, finishing building the data-driven proxy model of the first area, otherwise, returning to the step S202 to re-build the data-driven proxy model; specifically, the data-driven proxy model is a regression proxy model of a mixed deep belief network and a back propagation neural network.
S205, after the data-driven agent model of the first region is built, the agent models of other regions are quickly built by applying a pre-training-fine-tuning transfer learning technology.
Further, in some embodiments, the data-driven agent model of each sub-region is quickly constructed by using a pre-training-fine-tuning migration learning technique, and the specific steps include:
s2051, decomposing the multi-region power system into a plurality of scheduling sub-regions, and dividing the constructed data-driven agent model of the scheduling sub-regions into three parts, namely a deep belief network part, a logistic regression layer part and an error reverse propagation neural network;
s2052, copying the data of the first scheduling sub-area obtained in the step S204 to drive a deep belief network part of the proxy model, finely adjusting the deep belief network part according to actual data of the corresponding scheduling sub-area, and reconstructing a logistic regression part and a neural network part;
s2053, the operations in step S2052 are performed on each scheduling sub-area until the data-driven proxy models of all the areas are obtained.
S3, solving the data-driven agent model through an improved multi-target ant lion algorithm; the multi-objective ant lion algorithm is improved by updating ant populations by using a single-dimension retention optimization strategy. The method specifically comprises the following steps:
s301, initializing relevant parameters of the multi-target ant lion algorithm, wherein the relevant parameters of the multi-target ant lion algorithm comprise a population sample and algorithm parameters, and the active upper and lower limit constraints and the power balance constraints of the generator are simultaneously met when the population sample is initialized; and simultaneously, randomly initializing the power generation amount of each generator in each area to represent a possible solution of an optimization target.
S302, calculating fitness values of all individuals according to the initial values, namely, aiming at different initial values, scheduling an optimization objective function in a multi-objective optimization model according to the economic environment of the multi-zone power system, and solving the fossil fuel cost and the pollutant gas emission of each generator in each zone;
s303, assigning the ant population and the ant lion population as the corresponding generated energy of each generator in each area which is initialized randomly, solving a non-dominated solution according to the fitness value, and storing the obtained non-dominated solution into an external archive; if the number of the non-dominant solutions exceeds a preset value of an external file, executing a congestion ordering rule, and deleting redundant non-dominant solutions;
s304, recalculating a virtual fitness value corresponding to the initialization value by applying the constructed data-driven agent model, solving a non-dominated solution according to the virtual fitness value, and storing the obtained non-dominated solution into an internal archive; if the number of the non-dominant solutions exceeds the preset value of the internal file, executing a congestion ordering rule and deleting redundant non-dominant solutions;
s305, iteration starting:
s3051, traversing all individuals in the population, specifically, obtaining the next generation ant position after executing the single-dimensional retention strategy operation, updating the current ant lion position, processing the generator and the area transfer constraint condition, updating the next generation ant position, obtaining the virtual fitness values corresponding to all the individuals, updating the internal files by utilizing the internal files and the next generation ant virtual fitness values, and ending the traversal.
Further, in some embodiments, the single-dimension retention policy specifically includes:
(1) The population is divided into four parts, namely: the population of which the fitness values of the first target and the second target are smaller than the average value of the fitness values of the corresponding first target and the second target of the current population is named as a 'win-win population'; only a population in which the fitness value of the object one is smaller than the average value of the fitness values of the corresponding objects one of the current population is named as a 'first victory population'; only the population with the fitness value of the second target smaller than the average value of the fitness values of the second targets corresponding to the current population is named as a 'second winning population'; the population of which the fitness values of the first target and the second target are both larger than the average value of the fitness values of the corresponding first target and the second target of the current population is named as a failure population;
(2) Finding out minimum fitness values of a target I and a target II in the population, wherein the minimum fitness values are named as 'first minimum' and 'second minimum', and corresponding individuals are named as 'first minimum decision' and 'second minimum decision';
(3) Different operations are performed on the four populations, which are respectively: the 'win-win population' moves to the direction of the vector sum of 'minimum decision number one' and 'minimum decision number two'; "first victory population" moves to "first minimum decision"; "winning population II" moves to "minimum decision II"; the 'failure population' moves to the direction of the vector sum of the 'first minimum decision' and the 'second minimum decision', and the position of the whole population is named as P1;
(4) Finding out the minimum fitness values of a first target and a second target in the population, namely 'maximum first' and 'maximum second', and corresponding individuals are named 'maximum first decision' and 'maximum second decision';
(5) The operation of 'driving towards and avoiding the harmful substance' is carried out on the four groups, which are respectively as follows: "win-win population" moves towards and away from the direction of the vector sum of "minimum decision number one" and "minimum decision number two"; the 'first winning population' moves towards the 'first minimum decision' and away from the 'first maximum decision' direction; "winning population two" moves towards "least decision two" and away from "most decision two"; the 'failure population' moves to the direction of the vector sum of the 'first minimum decision' and the 'second minimum decision' and is far away from the direction of the vector sum of the 'first maximum decision' and the 'second maximum decision', and the overall population position is named as P2;
(6) Finding out the particles with the smallest radius of the niche in the population, moving all the particles to the particles with the smallest radius of the niche, and naming the position of the whole population as P3;
(7) The new population position is the vector sum of P1, P2 and P3, wherein the coefficient sum corresponding to P1, P2 and P3 is 1;
(8) And executing a single-dimension optimal retention strategy, judging each dimension of a particle, firstly generating a random number which meets the uniform distribution between 0 and 1, if the random number is more than 0.2, not changing the dimension of the particle, and if the random number is less than or equal to 0.2, changing the dimension of the particle into the corresponding dimension of a new population position.
S3052, calculating non-dominated solutions according to the individual optimization results, storing the obtained non-dominated solutions into an internal archive, and if the number of the non-dominated solutions exceeds a preset value of the internal archive, executing a congestion ordering rule and deleting redundant non-dominated solutions;
s3053, calculating a corresponding fitness value according to the updated individual value of the internal file; finding out the individual with the minimum value corresponding to the two targets respectively, and storing the obtained fitness value of the individual with the corresponding minimum value into an external archive; if the number of the non-dominant solutions exceeds a preset value of an external file, executing a congestion ordering rule, and deleting redundant non-dominant solutions;
and S3054, finishing iteration and generating a pareto optimal front edge of each region.
In the implementation process of the arithmetic example, initial parameters used for improving the multi-target ant lion algorithm can be initially assigned only according to experience, but the initial values are obviously not optimal parameters. In the running process of the algorithm, a certain parameter can be properly adjusted while other parameters are kept unchanged, then the quality of a simulation result is observed, and each parameter is continuously adjusted by the method, so that the optimal simulation effect is achieved.
S4, obtaining a global optimal solution as a decision basis for optimizing and scheduling each region of the power system at the current time period according to the solved result; in this embodiment, a discrimination method based on a membership function is provided, and a non-dominated solution having a maximum membership value is selected as a scheduling decision basis in the current time period. The method comprises the following specific steps:
s401, calculating a membership function value corresponding to a non-dominated solution of each optimized objective function:
wherein, F i,k For the kth solution of the ith optimization objective,
and
respectively, the minimum value and the maximum value of the ith optimization target.
S402, for each individual non-dominant solution, dividing mu i,k Regularization to give μ j ：
Wherein N is 1 Where =2 is the number of optimization targets, and M is the number of non-dominant solutions.
S403, solving mu j And obtaining a non-dominant solution corresponding to the maximum value as a scheduling decision basis of the current time interval.
In this embodiment, the obtained power grid operating parameters include: (1) the node parameters mainly comprise the voltage amplitude and the phase angle of an active load (3) node in a scheduling period of distribution (2) of the PQ, the PV and a reference node, the maximum voltage (4) born by the node, the active power output by a generator node and the maximum active power (5) born by the node, the maximum active power (6) allowed to be output by each generator, and branch parameters of the maximum active power (6) allowed to be output by each generator: the method comprises the following steps of (1) branch resistance, reactance, susceptance per unit value, capacity allowed by a long (short) distance power transmission branch, and maximum and minimum phase angles (7) allowed by the branch, and the regional hot standby requirement of the maximum allowed transmission capacity (8) of a transmission line.
The power grid parameters used by the calculation data adopted by the scheduling method can be obtained actual power grid operation parameters, and can also be any simulation object, such as four-area forty generator test systems. The simulation example selects four-area forty-generator test systems, and the structure diagram of the system is shown in fig. 3. Using fig. 3 as an example, the load of the current scheduling period is 10500MW, which relates to the extreme solutions of the two optimization objectives as shown in table 1 (fuel cost in units of @/h, pollutant gas emission in units of ton/h).
TABLE 1 Fuel cost extreme solution
In fig. 4, for the simulation example of fig. 3, the pareto optimal frontier of each region obtained by the present invention is composed of 10 non-dominant solutions, which is a set of trade-off solutions formed for two optimization objectives of environment and economy, and all the non-dominant solutions are the optimal solutions of the present scheduling period. The extreme value solution in fig. 4 is (827228.68,177386.3), and the size and the breadth of the extreme value solution directly determine the advancement of the optimization method.
In Table 1, the minimum fuel cost obtained by using the method of the invention is 827228.68 this th, which is lower than the other three methods 6411.46 th, 6977.18 th and 435.31 th respectively. The other extreme solution, namely the pollutant gas emission of 177386.3ton/h, is also the minimum of all methods, so the method is more advanced than other existing methods.
Compared with the conventional economic environment scheduling method, the method has the advantages that the performance is better than that of other scheduling methods on the premise of meeting the basic power supply requirement, the power generation cost is reduced, and meanwhile, the emission of pollution gas is effectively reduced. Taking the extreme points in table 1 as an example, if the solution is taken as the scheduling basis, and the load is assumed to be always stable at 10500MW, compared with the improved multi-objective ant lion algorithm and the NSOS algorithm, based on the method of the present invention, the fossil fuel cost can be saved by about 167452 yuan per day, and the fossil fuel cost can be saved by about 61120097 yuan per year.
Example two
The embodiment discloses a multizone electric power system economic environment dispatch system includes:
a model building module configured to: constructing a multi-region power system economic environment scheduling multi-objective optimization model considering multiple constraint conditions by taking the minimum pollutant gas emission and the fossil fuel cost as targets;
a solving module configured to: according to the target and the constraint condition, a data-driven agent model is constructed to convert the multi-region power system economic environment scheduling multi-target optimization model, and the converted multi-region power system economic environment scheduling multi-target optimization model is solved through a multi-target ant lion algorithm;
a decision module configured to: and according to the solving result, obtaining a global optimal solution as a decision basis for optimizing and scheduling each region of the power system in the current period.
It should be noted here that the model building module, the solving module and the deciding module correspond to the steps in the first embodiment, and the modules are the same as the corresponding steps in the implementation examples and application scenarios, but are not limited to the disclosure in the first embodiment. It should be noted that the modules described above as part of a system may be implemented in a computer system such as a set of computer-executable instructions.
EXAMPLE III
The third embodiment of the present invention provides an electronic device, which includes a memory, a processor, and a computer instruction stored in the memory and run on the processor, where the computer instruction is run by the processor to complete the steps of the data-driven multi-region power system economic environment scheduling method.
Example four
The fourth embodiment of the present invention provides a computer-readable storage medium, configured to store computer instructions, where the computer instructions, when executed by a processor, perform the steps of the above-mentioned data-driven multi-region power system economic environment scheduling method.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In the foregoing embodiments, the descriptions of the embodiments have different emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims (10)
1. The data-driven economic environment scheduling method of the multi-region power system is characterized by comprising the following steps:
constructing a multi-region power system economic environment scheduling multi-objective optimization model considering multiple constraint conditions by taking the minimum pollutant gas emission and the fossil fuel cost as targets;
according to the target and the multiple constraint conditions, a data-driven agent model is constructed to convert the multi-region power system economic environment scheduling multi-target optimization model, and the converted multi-region power system economic environment scheduling multi-target optimization model is solved through a multi-target ant lion algorithm;
and according to the solving result, obtaining a global optimal solution as a decision basis for optimizing and scheduling each region of the power system in the current period.
2. The data-driven-based multi-region power system economic environment scheduling method of claim 1, wherein the multi-region power system economic environment scheduling multi-objective optimization model is expressed as:
Minimize[(),E(P)],
Subject to: i (P)＝0,＝1,…, 1 ，
h j ()≤0,j＝1,…,M 2 ,
wherein F (P) is an optimization objective function of fossil fuel cost, E (P) is an optimization objective function of pollutant gas emission, g i (P) is the inequality constraint involved, h j () Are constrained by the equations involved.
3. The data-driven-based economic environment scheduling method for the multi-region power system as claimed in claim 1, wherein the multiple constraint conditions include generator active upper and lower limit constraints, power balance constraints, node voltage amplitude constraints, line tide constraints, transmission line safety constraints, region hot standby transfer constraints and line loss constraints;
further, the active upper and lower limit constraints of the generator are expressed as:
the power balance constraint is expressed as:
the node voltage magnitude constraint is expressed as:
V i min ≤V i ≤V i max ,＝1,…, bus
the line flow constraint is expressed as:
the transmission line safety constraints are expressed as:
the regional hot standby transfer constraint is expressed as:
the line loss constraint is expressed as:
4. the method according to claim 1, wherein the step of constructing the data-driven agent model to convert the multi-region power system economic environment scheduling multi-objective optimization model comprises:
solving a plurality of feasible solutions meeting the constraint conditions according to the multiple constraint conditions; calculating an optimization target value corresponding to the feasible solution according to the feasible solution and the multi-objective optimization model for the economic environment scheduling of the multi-zone power system;
and according to the feasible solution and the corresponding optimization target value, constructing a data-driven agent model to replace the multi-objective optimization model for the economic environment scheduling of the multi-region power system.
5. The data-driven multi-region power system economic environment scheduling method as claimed in claim 1, wherein the concrete steps of solving the transformed multi-region power system economic environment scheduling multi-target optimization model through the multi-target lion algorithm include:
initializing ant populations and ant lion populations, and randomly initializing the power generation capacity of each generator in each area;
assigning the ant population and the ant lion population as the corresponding generated energy of each generator in each area which is initialized randomly, and solving the non-dominated solution of each optimization target;
traversing all individuals in the population, and calculating a non-dominant solution according to an individual optimization result; when all individuals in the population are traversed, executing a single-dimension retention strategy operation to update the ant population; the non-dominant solutions are sorted according to the initialization value.
6. The method according to claim 5, wherein the step of performing the one-dimensional retention policy operation to update the ant population comprises:
dividing the ant population into four parts according to the fitness value of the optimization target;
finding the minimum fitness value of the optimization target in the population, respectively executing position moving operation on the four parts of ant populations, and obtaining intermediate position variables;
acquiring a new population position according to the intermediate position variable;
and executing a single-dimension optimal retention strategy, and judging whether to update the dimension of the ion or not according to each dimension of each particle.
7. The data-driven multi-region power system economic environment scheduling method according to claim 1, wherein the specific step of obtaining a global optimal solution according to the solution result comprises:
aiming at an optimization objective function of the pollution gas emission and the fossil fuel cost, calculating a membership function value corresponding to a non-dominated solution of the optimization objective function through a membership function;
and carrying out regularization processing on the membership function value and solving to obtain a global optimal solution.
8. A multi-region power system economic environment scheduling system is characterized by comprising:
a model building module configured to: constructing a multi-region power system economic environment scheduling multi-objective optimization model considering multiple constraint conditions by taking the minimum pollution gas emission and fossil fuel cost as targets;
a solving module configured to: according to the target and the constraint condition, a data-driven agent model is constructed to convert the multi-region power system economic environment scheduling multi-target optimization model, and the converted multi-region power system economic environment scheduling multi-target optimization model is solved through a multi-target ant lion algorithm;
a decision module configured to: and according to the solving result, acquiring a global optimal solution as a decision basis for optimal scheduling of each region of the power system at the current time period.
9. An electronic device comprising a memory and a processor and computer instructions stored on the memory and executed on the processor, the computer instructions when executed by the processor performing the steps of any of claims 1-7.
10. A computer-readable storage medium storing computer instructions which, when executed by a processor, perform the steps of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211579274.1A CN115759475A (en) | 2022-12-08 | 2022-12-08 | Data-driven multi-region power system economic environment scheduling method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211579274.1A CN115759475A (en) | 2022-12-08 | 2022-12-08 | Data-driven multi-region power system economic environment scheduling method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115759475A true CN115759475A (en) | 2023-03-07 |
Family
ID=85346633
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211579274.1A Pending CN115759475A (en) | 2022-12-08 | 2022-12-08 | Data-driven multi-region power system economic environment scheduling method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115759475A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116632808A (en) * | 2023-03-08 | 2023-08-22 | 南方电网科学研究院有限责任公司 | Power scheduling optimization method and device and nonvolatile storage medium |
-
2022
- 2022-12-08 CN CN202211579274.1A patent/CN115759475A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116632808A (en) * | 2023-03-08 | 2023-08-22 | 南方电网科学研究院有限责任公司 | Power scheduling optimization method and device and nonvolatile storage medium |
| CN116632808B (en) * | 2023-03-08 | 2024-05-28 | 南方电网科学研究院有限责任公司 | Power scheduling optimization method and device and nonvolatile storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107579518B (en) | Power system environment economic load dispatching method and apparatus based on MHBA | |
| Wang et al. | Demand side management of plug-in electric vehicles and coordinated unit commitment: A novel parallel competitive swarm optimization method | |
| Mahdad et al. | Optimal power flow for large-scale power system with shunt FACTS using efficient parallel GA | |
| Suresh et al. | Microgrid energy management system with embedded deep learning forecaster and combined optimizer | |
| CN105048479B (en) | A kind of idle packet adjusting method of photovoltaic plant | |
| CN112018762B (en) | Electric vehicle charging optimization scheduling method considering transmission and distribution cooperation with reactive voltage constraint | |
| Xi et al. | A virtual generation ecosystem control strategy for automatic generation control of interconnected microgrids | |
| CN115759475A (en) | Data-driven multi-region power system economic environment scheduling method and system | |
| CN114140022A (en) | Multi-virtual power plant distributed dynamic economic dispatching method and system | |
| CN111509784B (en) | Uncertainty-considered virtual power plant robust output feasible region identification method and device | |
| CN115439000A (en) | Power distribution network block division method considering wind-solar-load power uncertainty and correlation | |
| CN116706917A (en) | Intelligent park collaborative regulation and control method and system based on rapid alternating direction multiplier method | |
| CN114784831A (en) | Active power distribution network multi-objective reactive power optimization method based on mobile energy storage | |
| CN110380420A (en) | A kind of distributed multiple target optimization algorithm of quick self-adapted analytic expression safety | |
| Liu et al. | Multi-objective optimal allocation of TCSC for power systems with wind power considering load randomness | |
| Li | Multi-objective optimization design for battery pack of electric vehicle based on neural network of radial basis function (RBF) | |
| Zhang et al. | Two-timescale coordinated voltage regulation for high renewable-penetrated active distribution networks considering hybrid devices | |
| Zhang et al. | Many-objective optimal power flow problems based on distributed power flow calculations for hierarchical partition-managed power systems | |
| CN112036654B (en) | Photovoltaic power station and electric vehicle charging network planning method based on co-evolution | |
| Zeng et al. | Non-Convex optimal power flow implementation by distributed Meta-Heuristic optimization algorithm | |
| CN116306310A (en) | Knowledge distillation-based economic environment scheduling method and system for multi-region power system | |
| CN110729759A (en) | Method and device for determining distributed power supply configuration scheme in micro-grid | |
| CN108521131A (en) | Hybrid cross iteration method for solving reactive power optimization model | |
| Kim et al. | A new concept of power flow analysis | |
| Xie et al. | Investment Strategy of Reactive Power Compensation Scheme in Wind Turbine Distribution Network Based on Optimal Allocation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |