互联电网自动发电控制功率分配的改进逼近于理想解的排序--Q多目标优化算法

Stochastic optimal generation command dispatch of interconnected power grids based on improved multi-objective technique for order preference similar to an ideal solution–Q algorithm

中文关键词  改进TOPSIS--Q  强化学习  多目标优化  AGC功率分配

英文关键词  improved TOPSIS–Q  reinforcement learning  multi-objective optimization  AGC generation command dispatch

中文摘要

      本文提出了一种多目标决策与强化学习相结合的改进的逼近于理想解的排序(technique for order preference similar to an ideal solution, TOPSIS)--Q算法, 有效解决 了自动发电控制(automatic generation control, AGC)总功率 指令分配到风电、水电、火电等各类AGC机组的动态随机多目标优化问题. 算法采用3个不同的目标$Q$值矩阵进行迭代更新, 然后 利用改进TOPSIS方法对$Q$值矩阵进行多目标决策处理, 客观地给出各目标的动态最优权重系数, 从而得到各状态--动作对的综合评价判据. IEEE标准两区域模型仿真研究验证了改进TOPSIS--Q算法在AGC机组功率多目标动态优化分配过程应用的可行性和有效性, 在复杂 随机扰动的环境中提高系统CPS性能的同时, 有效降低了AGC机组调节成本和碳排放.

英文摘要

      This paper proposes an improved technique for order preference similar to an ideal solution (TOPSIS)–Q learning approach to solve the dynamic optimization of generation command dispatch (GCD) for automatic generation control (AGC) in a multi-energy power system. Three optimization objectives are simultaneously optimized by three different Q-value matrixes. Then dynamic optimal weight of each objective is calculated by improved TOPSIS method such that the evaluation criterion of each state-action is obtained. Case studies are carried out to evaluate the optimization performance of the proposed algorithm in the two-area load frequency control (LFC) power system model. Simulation results indicate that the proposed method is feasible and effective for dynamic optimization of GCD problem, and can reduce the regulating cost and carbon emissions while improving the control performance of AGC systems in the complex stochastic power system.