基于群体智能算法的电力台区数据分析技术
    点此下载全文
引用本文:张 欣1,庄 园1,宁学玲1,治 强2.基于群体智能算法的电力台区数据分析技术[J].计算技术与自动化,2022,(2):87-91
摘要点击次数: 80
全文下载次数: 0
作者单位
张 欣1,庄 园1,宁学玲1,治 强2 (1.国网吉林省电力有限公司培训中心吉林 长春 1300002.国网吉林综合能源服务有限公司吉林 长春 130000) 
中文摘要:电力系统经济负荷运行调节中,存在复杂的非凸、非线性问题,难以实现电力系统运行可靠性和经济性。传统群体智能算法只是单一模仿生物群体行为,不能很好地进行电力系统负荷经济调节。因此提出了一种基于群体智能算法统一框架的粒子群算法,算法根据不同的适应度隶属度函数赋予不同惯性权重,通过外点法构造辅助函数将非可行域的约束以罚函数形式写入目标函数,并建立变罚系数提升目标函数求解过程的搜索范围和收敛速度,同时由搜索策略和变异策略,提升全局搜索能力。通过仿真结果表明:改进PSO算法相较于传统遗传算法、粒子群算法的收敛速度更快,解的离散度更小,收敛精度更高。
中文关键词:粒子群算法  群体智能框架  罚系数  经济负荷分配  收敛精度
 
Analysis of Power Station Based on Group Intelligent Algorithm
Abstract:In the power system economic load operation regulation, there are complex nonconvex and nonlinear problems, which is difficult to realize the operation reliability and economy of the power system. The traditional group intelligent algorithm only imitates the behavior of biological groups and cannot perform the power system load economic regulation well.This paper proposes a particle group algorithm based on the unified framework of group intelligence algorithm, which gives different inertia weights according to different fitness membership function, constructs the auxiliary function to write non-feasible domain to the target function as the penalty function, and establishes the search range and convergence speed of the solution process of variable penalty coefficient, and the variation strategy to improve the global search capability.The simulation results show that the improved PSO algorithm converges faster, has less dispersion and higher convergence accuracy than the traditional genetic algorithm and particle group algorithm.
keywords:particle group algorithm  group intelligent framework  penalty coefficient  economic load distribution  convergence accuracy
查看全文   查看/发表评论   下载pdf阅读器