基于双层粒子群算法求解电力市场均衡  

Equilibrium Solution for Electricity Market Based on Bi-level Particle Swarm Optimization Algorithm

作　　者：谢畅[1] 王蓓蓓[1] 赵盛楠[1] 谈健 黄俊辉 谢珍建 XIE Chang1,WANG Beibei1, ZHAO Shengnan1, TAN Jian2, HUANG Junhui2, XIE Zhenjian2(1. School of Electrical Engineering, Southeast University, Nanjing 210096, Jiangsu Province, China; 2. State Grid Jiangsu Economic Research Institute, Nanjing 210008, Jiangsu Province, Chin)

机构地区：[1]东南大学电气工程学院,江苏省南京市210096 [2]国网江苏省电力有限公司电力经济技术研究院,江苏省南京市210008

出　　处：《电网技术》2018年第4期1170-1176,共7页Power System Technology

基　　金：国网江苏省电力公司科技项目“高受电比例受端电网辅助服务机制的研究”的资助

摘　　要：针对电力市场均衡策略的求解,现阶段国内外广泛采用非线性互补算法和循环迭代的方法。其中,非线性互补算法是通过Karush-Kuhn-Tucker条件将发电商最优反应函数转换成一组半光滑非线性互补方程组,松弛因子维度高且算法异常复杂。而循环迭代法采用重复求解各发电商的最优反应函数的思想,常常因为收敛性差而求不到均衡解。针对上述问题,提出了一种基于双层粒子群算法求解市场均衡的思路。其中,外层粒子群算法搜索可行域内的策略组合,以提出的纳什适应度函数为寻优标准;内层粒子群算法和内点法结合计算外层中的纳什适应度函数值。采用标准3节点和IEEE 30节点测试系统对所提算法进行了仿真,求得结果与采用非线性互补算法的结果完全一致,验证了所提出的双层粒子群算法的有效性。It is a complex and difficult problem to solve equilibrium strategy of electricity market, and nonlinear complementarity algorithm and cyclic iteration method are widely adopted in China and abroad to solve it. The nonlinear complementarity algorithm converts optimal response function of generator to a group of semi-smooth nonlinear complementary equations based on Karush-Kuhn-Tucker condition with high dimension of relaxation factor and very complex algorithm. The iterative method adopts the idea of repeatedly solving optimal reaction function of each generator, and the equilibrium solution is often not obtained because of poor convergence. In view of current algorithm problem, this paper proposes a bi-level particle swarm optimization（PSO） algorithm to solve the equilibrium strategy. In this algorithm, the outer layer PSO searches strategy combination within feasible domain with the proposed Nash fitness function as optimization criterion. The inner layer PSO and interior point method are used to calculate Nash fitness function value in the outer layer. Finally, standard 3-node and IEEE 30-node test system are used to simulate the proposed algorithm, with obtained results similar to the nonlinear complementarity algorithm. Simulation results verify effectiveness of the proposed bi-level PSO algorithm.

关 键 词：电力市场 纳什均衡 双层粒子群算法 纳什适应度函数 

分 类 号：TM73[电气工程—电力系统及自动化]