电工技术学报2024，Vol.39Issue(7)：2017-2032,16.DOI:10.19595/j.cnki.1000-6753.tces.230186
基于集合经验模态分解和多目标遗传算法的火-多储系统调频功率双层优化
Two-Layer Optimization of Frequency Modulated Power of Thermal Generation and Multi-Storage System Based on Ensemble Empirical Mode Decomposition and Multi-Objective Genetic Algorithm
摘要
Abstract
Aiming at the power scheduling problem of frequency modulation involving multiple energy storage stations distributed in different network nodes of regional power grid,a two-layer optimization strategy for frequency modulated power of thermal generation and multi-storage system based on ensemble empirical mode decomposition(EEMD)and multi-objective genetic algorithm(MOGA)is proposed.This strategy includes a thermal power-energy storage frequency modulation power optimization layer and a multi energy storage power station frequency modulation power optimization layer:The upper layer counts the respective advantages of thermal power-energy storage allocation resources as well as the residual frequency modulation capacity of thermal power units and energy storage,constructs the optimization distribution model of thermal power-energy storage frequency modulation power,and completes the distribution of thermal power-energy storage frequency modulation power.The lower layer introduces the adaptive weight coefficient about frequency modulation cost and state of charge(SOC)comprehensive state,takes frequency modulation cost and SOC comprehensive state as the optimization objective,and completes the distribution of frequency modulation power among multiple energy storage systems. In order to verify the effectiveness of the above control strategies,this paper takes an actual AGC command from a certain location in China for simulation analysis.The specific conclusions are as follows: Firstly,the thermal power-energy storage frequency modulation power allocation model constructed in this article can decompose the original AGC instructions into the AGC instructions that the thermal power unit as a whole and the energy storage power station as a whole need to bear.The decomposed signal has a good tracking effect on the original signal.The frequency modulation signal borne by the thermal power unit under this strategy follows the original frequency modulation signal,which is 64.29% higher than the traditional filtering strategy. Secondly,the dual level optimization model for frequency regulation power proposed in this article can improve the overall frequency regulation economy of the regional power grid.Compared with the capacity proportion allocation strategy,the frequency regulation cost of thermal power units is reduced by 3.00% ,and the unit wear is also slightly reduced;Compared to the proportional allocation strategy,the overall unit energy consumption cost of energy storage plants has been reduced by 3.6% . Finally,the frequency modulation power allocation model proposed in this article can coordinate the frequency modulation cost and SOC recovery of each energy storage station.On the basis of reducing the overall frequency regulation cost,the SOC maintenance effect of each energy storage station is 34.78% better than the proportional allocation strategy;The overall SOC balance effect of multiple energy storage power stations is 52.69% better than the proportional allocation strategy.关键词
多火电储能系统/二次调频/双层优化控制/多目标遗传算法(MOGA)自适应权重系数Key words
Multi thermal power-energy storage system/secondary frequency modulation/two-layer optimal control/multi-objective genetic algorithm(MOGA)/adaptive weight coefficient分类
信息技术与安全科学引用本文复制引用
李翠萍,司文博,李军徽,严干贵,贾晨..基于集合经验模态分解和多目标遗传算法的火-多储系统调频功率双层优化[J].电工技术学报,2024,39(7):2017-2032,16.基金项目
国家电网有限公司科技项目资助(5108-202299257A-1-0-ZB). （5108-202299257A-1-0-ZB）
