基于改进细菌觅食算法的无功优化  

Reactive Power Optimization Based on Modified Bacterial Foraging Optimization Algorithm

作　　者：邓集瀚 杨俊杰[1] 魏春娟[1] 楼志斌[2] 

机构地区：[1]上海电力学院电子与信息工程学院,上海200090 [2]上海科学院,上海201203

出　　处：《仪表技术》2016年第5期37-40,共4页Instrumentation Technology

摘　　要：为改善BFO算法的寻优能力,提出一种改进的MBFO算法。首先在趋化操作中采用步长缩进的控制策略,同时位置较差的细菌能获得相对较长的步长,使算法能更好的在解空间进行搜索,并提升了最优点附近的挖掘能力;然后在聚群操作中引入近朱者赤近墨者黑的思想,让最佳者发挥吸引的作用,让最差者发挥排斥的作用,简化了群体效应影响值的计算;最后在趋化循环中引入赌轮盘选择,降低精英个体被选中的概率,防止解的劣化。将MBFO用于求解电力系统无功优化问题,通过IEEE-30节点的算例分析表明,MBFO比其他几种优化方法具有更快地收敛速度和更好地优化效果,故该算法在解决无功优化问题上可行且有效。In order to improve the searching ability of BFO algorithm,a Modified Bacterial Foraging Optimization Algorithm was proposed. In the chemotaxis loop,a control mode of decreasing step size was introduced. The bacteria in poor location had a relatively longer step,so that the algorithm could search the solution space more fully and enhance the search ability near global sides. In the swarming loop,the thought was introduced,that is,let the best one play the role of attraction,the worst one play repellant,to simplify the calculation of group effect value. In elimination and dispersal loop,roulette wheel selection was introduced to protect the elites and suppress the degeneracy phenomenon. Using MBFO to make reactive power optimization calculation,the example of IEEE-30 node shows that MBFO has faster convergence speed and better optimization effect than other optimization methods,so the algorithm is feasible and effective in solving reactive power optimization problem.

关 键 词：电力系统 无功优化 细菌觅食算法 步长 轮盘赌选择 

分 类 号：TM714.3[电气工程—电力系统及自动化]