Search

CN-122001009-A - Method and device for deciding power supply of outgoing base, electronic equipment and storage medium

CN122001009ACN 122001009 ACN122001009 ACN 122001009ACN-122001009-A

Abstract

The invention relates to the technical field of power system planning and discloses a method, a device, electronic equipment and a storage medium for deciding power supply of an outgoing base, wherein the method comprises the steps of collecting data of a plurality of reference indexes of the outgoing base; the method comprises the steps of establishing a multi-constraint power supply decision model based on a business scene of an outgoing base, generating an initial solution set by adopting a genetic algorithm, carrying out model solving based on data of each reference index and the initial solution set to obtain an accurate solution corresponding to each initial solution in the initial solution set, screening an effective solution from the accurate solutions, optimizing the genetic algorithm based on the effective solution, returning to the step of generating the initial solution set by adopting the genetic algorithm until a preset stopping condition is reached, determining the accurate solution solved last time as a target solution, and obtaining a power supply decision scheme based on the data of each reference index and all target solutions. The invention realizes the global optimal decision of the multi-power collaborative optimization of the outgoing base through the deep collaborative fusion of the genetic algorithm and the mixed integer programming.

Inventors

  • Xiao Lingjuan

Assignees

  • 国能经济技术研究院有限责任公司

Dates

Publication Date
20260508
Application Date
20251231

Claims (10)

  1. 1. A method of outgoing base power decision making, the method comprising: collecting data of a plurality of reference indexes of an outgoing base; based on the business scene of the outgoing base, a multi-constraint power supply decision model is established; Generating an initial solution set by adopting a genetic algorithm; solving the multi-constraint power supply decision model based on the data of each reference index and the initial solution set to obtain an accurate solution corresponding to each initial solution in the initial solution set; screening at least one valid solution from among all the exact solutions, optimizing the genetic algorithm based on the at least one valid solution; Returning to the step of generating an initial solution set by adopting a genetic algorithm until a preset stopping condition is reached, determining the accurate solution solved for the last time as a target solution, and obtaining a power supply decision scheme based on the data of each reference index and all target solutions.
  2. 2. The method of claim 1, wherein the establishing a multi-constraint power decision model based on the traffic scenario of the outgoing base comprises: setting a plurality of variables to be solved, a plurality of constraint conditions and an objective function; fine tuning the plurality of variables to be solved, the plurality of constraint conditions and/or the objective function based on the business scenario of the outgoing base; and establishing the multi-constraint power supply decision model based on the finely adjusted variables to be solved, the constraint conditions and the objective function.
  3. 3. The method of claim 1, wherein the generating an initial solution set using a genetic algorithm comprises: Generating a plurality of original combinations based on data boundaries corresponding to a plurality of target indexes, wherein the target indexes are the installed capacity of each equipment type of the delivery base, and each original combination comprises data corresponding to all target indexes; When the number of the plurality of original combinations reaches a preset number, calculating the fitness of each original combination by adopting a fitness function of the genetic algorithm; Performing iterative optimization on all original combinations by adopting a plurality of genetic operations of the genetic algorithm to obtain a plurality of optimized original combinations, wherein the number of the optimized original combinations is consistent with the number of the original combinations before iterative optimization; Repeating the process of calculating the fitness and iterative optimization based on the optimized original combinations until reaching a preset iterative stopping condition to obtain the optimized original combinations obtained by the final iterative optimization as target original combinations; and calculating the fitness of each target original combination, sorting all target original combinations according to the fitness, and determining the initial solution set from the sorted target original combinations.
  4. 4. The method of claim 2, wherein solving the multi-constraint power decision model based on the data of each reference index and the initial solution set to obtain an exact solution corresponding to each initial solution in the initial solution set comprises: For each initial solution in the initial solution set, inputting the initial solution into the multi-constraint power supply decision model, substituting the initial solution into a corresponding variable to be solved, and solving the multi-constraint power supply decision model by adopting a solver to obtain a solution of each variable to be solved; Integrating solutions of each variable to be solved according to the time dimension and the equipment type to generate a power generation scheduling scheme; based on the solution of each variable to be solved and the data of each reference index, carrying out economic evaluation and environmental evaluation to obtain an evaluation result; And taking the initial solution as a set of installed capacities, and taking the set of installed capacities, the power generation scheduling scheme and the evaluation result corresponding to the initial solution as an accurate solution corresponding to the initial solution.
  5. 5. A method according to claim 3, wherein each exact solution comprises a power generation scheduling scheme; The screening at least one valid solution from among all the exact solutions includes: for each accurate solution, calculating the average power transmission channel utilization rate corresponding to the accurate solution; Determining a first power transmission channel constraint based on a numerical interval in which the average power transmission channel utilization rate is located; calculating the single-period output fluctuation amplitude based on the accurate solution power generation scheduling scheme, and determining the period duty ratio of the single-period output fluctuation amplitude larger than a fluctuation threshold value; determining a first new energy fluctuation constraint based on the period duty cycle; and eliminating the first power transmission channel constraint into a loose constraint and/or the first new energy fluctuation constraint into a loose constraint accurate solution, and determining the eliminated accurate solution as an effective solution.
  6. 6. The method of claim 5, wherein said optimizing said genetic algorithm based on said at least one effective solution comprises: Calculating a first average value of the average power transmission channel utilization rate corresponding to all the effective solutions, and determining a second power transmission channel constraint based on a numerical interval in which the first average value is located; Calculating a second average value of time period duty ratios corresponding to all the effective solutions, and determining a second new energy fluctuation constraint based on the second average value; and adjusting the fitness function, the preset iteration stop condition and the variation rate of the genetic algorithm based on the second power transmission channel constraint and the second new energy fluctuation constraint.
  7. 7. The method of claim 1, wherein each target solution comprises a set of installed capacities, a power generation scheduling scheme, and an evaluation result, the evaluation result comprising a net present value; The power decision scheme is obtained based on the data of each reference index and all target solutions, and comprises the following steps: Carrying out consistency verification on all target solutions, and eliminating target solutions which do not pass the verification to obtain reserved solutions; Determining a fusion weight for each retention solution based on a net present value in an evaluation result of the retention solution; and respectively carrying out weighted average on the installed capacity set, the power generation scheduling scheme and the evaluation result of all the reserved solutions based on the fusion weight of each reserved solution to obtain a target installed capacity set, a target power generation scheduling scheme and a target evaluation result, thereby forming the power supply decision scheme.
  8. 8. An outgoing base power decision device, the device comprising: the acquisition module is used for acquiring data of a plurality of reference indexes of the outgoing base; The building module is used for building a multi-constraint power supply decision model based on the business scene of the outgoing base; the generation module is used for generating an initial solution set by adopting a genetic algorithm; The solving module is used for solving the multi-constraint power supply decision model based on the data of each reference index and the initial solution set to obtain an accurate solution corresponding to each initial solution in the initial solution set; an optimization module for screening at least one valid solution from among all the exact solutions, optimizing the genetic algorithm based on the at least one valid solution; The decision module is used for returning to the step of generating the initial solution set by adopting the genetic algorithm until the preset stop condition is reached, determining the accurate solution solved for the last time as a target solution, and obtaining a power supply decision scheme based on the data of each reference index and all the target solutions.
  9. 9. An electronic device, comprising: a memory and a processor communicatively coupled to each other, the memory having stored therein computer instructions that, when executed, perform the outgoing base power decision method of any one of claims 1 to 7.
  10. 10. A computer readable storage medium having stored thereon computer instructions for causing a computer to perform the outgoing base power decision method of any one of claims 1 to 7.

Description

Method and device for deciding power supply of outgoing base, electronic equipment and storage medium Technical Field The invention relates to the technical field of power system planning, in particular to an outgoing base power supply decision method, an outgoing base power supply decision device, electronic equipment and a storage medium. Background The outgoing base is an energy base for intensively developing new energy sources such as wind power, photovoltaic and the like in a large scale, and the electric power is mainly sent to a load center through extra-high voltage and other trans-regional power transmission channels. With the large-scale development and centralized grid connection of new energy sources, the base has the following problems that the output of the new energy sources is influenced by the weather conditions to present strong fluctuation and intermittence, the unbalance risk of power supply and demand is easy to cause, the ultra-high voltage transmission channel has strict capacity limitation, the output and the transmission capacity of the power sources are required to be accurately and comprehensively planned, and the base comprises multiple power source types such as wind power, photovoltaic, coal power, energy storage and the like, the technical characteristic difference of the various power sources is obvious, and the difficulty of multi-power source collaborative scheduling is greatly increased. In order to solve the problems, the prior art adopts manual experience to combine Excel table analysis and a simple linear programming tool (such as a Lingo basic edition) to realize the power supply decision of an outgoing base, and the process is that a dispatcher manually estimates the power generation requirement of the next day according to the load data of the previous day and distributes the power output of each power supply according to the fixed principle of' new energy priority Internet surfing and coal electric power covering. However, the power supply decision is based on manual experience to predict the load, and only the generated energy and the output energy are concerned, so that the decision is inaccurate, the time consumption of manual calculation is long, the real-time fluctuation of the new energy output cannot be adapted, and the refined scheduling requirement of a modern energy system is difficult to meet. Disclosure of Invention The invention provides an outgoing base power supply decision method, an outgoing base power supply decision device, electronic equipment and a storage medium, which are used for solving the problems of low efficiency, poor precision and weak adaptability of the existing outgoing base power supply decision method. In a first aspect, the present invention provides a method for outgoing base power decision making, the method comprising: collecting data of a plurality of reference indexes of an outgoing base; based on the business scene of the outgoing base, a multi-constraint power supply decision model is established; Generating an initial solution set by adopting a genetic algorithm; Solving the multi-constraint power supply decision model based on the data of each reference index and the initial solution set to obtain an accurate solution corresponding to each initial solution in the initial solution set; Screening at least one effective solution from all the accurate solutions, and optimizing a genetic algorithm based on the at least one effective solution; Returning to the step of generating an initial solution set by adopting a genetic algorithm until a preset stopping condition is reached, determining the accurate solution solved for the last time as a target solution, and obtaining a power supply decision scheme based on the data of each reference index and all target solutions. The invention can comprehensively and accurately reflect the actual condition of the power system by collecting the data of a plurality of reference indexes of the outgoing base, ensures that the subsequent power supply decision fits the actual scene of the outgoing base, and takes the feasibility, the economy and the compliance into consideration. And (3) establishing a multi-constraint power supply decision model by considering the actual service scene of the outgoing base, and ensuring that a power supply decision scheme meets the actual operation requirement. And generating an initial solution set by adopting a genetic algorithm, inputting the initial solution set into a model for solving to obtain an accurate solution, screening effective solutions from the accurate solutions for algorithm optimization, enabling the initial solution output by the genetic algorithm to be more suitable for the solving characteristics of the mixed integer programming model, and strengthening the synergistic optimization effect of the double algorithms. And repeating the process by using the optimized algorithm, and determining a power supply decision scheme based on the accur