基于混合编码的配电网络故障恢复遗传算法优化

Genetic Algorithm Optimization of Power Distribution Grid Fault Recovery Based on Hybrid Encoding

染色体的编码和适应函数的构造是遗传算法实现配电网故障恢复优化求解的关键问题,直接决定着算法效率。提出了对网络拓扑结构进行节点深度编码,对非故障失电区的切负荷控制进行0-1二进制编码的遗传算法染色体的混合编码;对故障恢复建模采用多目标优化算法,减少权重系数主观性对最优解的影响,实现多优化目标协同进化的新思路,并进行了案例仿真分析。仿真验证结果表明,建立的混合编码方法能显著减少不可行解的数量,提高了算法的收敛速度,满足配电网故障恢复的及时性需求。

The code of chromosome and the structure of fit-ness function are the key problems to actualize optimal so-lution to the fault recovery of power distribution grid by u-sing genetic algorithm. A hybrid encoding scheme is pres-ented in this paper, including in-depth node encoding on topological structure of the network, and 0 - 1 binary enco-ding on load switch control of outage region caused by non-fault. The multi-objective optimization algorithm on fault recovery model could reduce the influence of weight coeffi- cient subjectivity, thus a new idea, the co-evolution of multi-objective optimization is achieved, and the simulation case is analyzed, which shows that the hybrid coding meth-od can significantly reduce the number of infeasible solu-tions and improve the algorithm＇s convergence speed, so the proposed method could meet the demands of real-time fault recovery of power distribution grid.