基于改进布谷鸟算法的微电网优化调度  

Optimal Scheduling of Microgrid Based on Improved Cuckoo Search Algorithm

作　　者：李飞 魏小城 陈勇弟 郭广森 张建华 LI Fei;WEI Xiaocheng;CHEN Yongdi;GUO Guangsen;ZHANG Jianhua(School of Electrical Engineering and Automation,Jiangsu Normal University,Xuzhou 221116,China)

机构地区：[1]江苏师范大学电气工程及自动化学院,江苏徐州221116

出　　处：《控制工程》2024年第11期1963-1971,共9页Control Engineering of China

基　　金：江苏省高等学校自然科学基金资助项目(22KJA580003)。

摘　　要：为提高微电网日运行经济效益及环保效益,同时解决传统布谷鸟算法寻优精度差、易陷入局部最优的问题,提出了一种基于复合混沌映射的维间反向扰动学习布谷鸟算法。首先,引入统计参数映射、动态步长因子策略及指数递减惯性权重偏好游走策略以提高寻优精度和收敛速度;其次,采用动态自适应发现概率来控制全局搜索和局部搜索的切换频率。将所提算法通过10个典型测试函数进行仿真,结果表明改进的布谷鸟算法相比灰狼等元启发式算法具有更好的寻优精度、求解稳定性、收敛速度;最后,再将改进的布谷鸟算法用于求解微电网并网模式下的多目标优化调度问题。仿真结果表明,改进的布谷鸟算法能有效提高微电网日运行总效益及系统运行稳定性。In order to improve the daily operation economy and environmental protection benefits of microgrid,as well as to solve the problem that the traditional Cuckoo search algorithm has poor optimization accuracy and is easily prone to local optimization,an inter-dimensional reverse perturbation learning Cuckoo search algorithm based on composite chaotic mapping is proposed.Firstly,to increase optimization accuracy and convergence speed,statistical parameter mapping,dynamic step factor strategy,and exponentially decreasing inertia weight preference wandering strategy are introduced.Secondly,dynamic adaptive discovery probability is employed to manage the switching frequency between global and local search.The proposed algorithm is simulated through ten typical test functions,and the results demonstrate that the improved Cuckoo search algorithm outperforms the grey wolf and other metaheuristic algorithms in terms of optimization accuracy,solution stability,and convergence speed.Finally,in the grid-connected mode of the microgrid,the improved Cuckoo search algorithm is applied to solve the multi-objective optimal operation problem.The simulation results show that the improved Cuckoo search algorithm significantly increase the overall daily operation benefit of the microgrid and system operation stability.

关 键 词：微电网 优化调度 布谷鸟算法 指数递减惯性权重 反向扰动学习 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]