中文摘要：

针对电力网络无功优化问题,增加考虑了电压稳定指标,采用改进的差分演化算法同时对网损和电压稳定指数进行优化.该算法基于一般差分演化算法,在进化的不同阶段对算法参数缩放因子F和杂交概率CR进行自适应调整与控制,增加种群多样性的同时,能够调整搜索方向快速找到最优化方案.应用IEEE 30节点和IEEE 57节点系统进行了测试,结果表明所用算法相比遗传算法和一般差分演化算法,不仅能够有效减少有功网损,还能让PQ负荷节点远离电压崩溃点,并具有计算速度快、鲁棒性好等优点.

英文摘要：

An improved difference evolution algorithm JADE is proposed to optimize the net loss and volt- age stability index in the reactive power optimization problem. Based on the general difference evolution al- gorithm, the algorithm change the algorithm parameter scaling factor F and crossover rate CR at different stages of evolution to adapt adjustment and control, which increase the diversity of the population, while the search direction can be adjusted quickly to find the best solution. By testing the IEEE 30-bus and IEEE 57-bus system , it can be obtained that, compared with genetic algorithms and general difference evolution algorithm, the algorithm could effectively reduce the active power loss, but also let PQ nodes away from voltage collapse point, which has a fast convergence rate and the advantages of robustness.