动态负荷模型参数辨识的微分进化算法  

Differential Evolution Algorithm for Dynamic Load Model Parameter Identification

作　　者：黄玉龙[1] 陈迅[2] 刘明波[3] 陈晓科[2] 杨汾艳[2] 曾艳[2] 林舜江[3] 

机构地区：[1]暨南大学电气信息学院,珠海5190702 [2]广东电网公司电力科学研究院,广州510600 [3]华南理工大学电力学院,广州510640

出　　处：《电工技术学报》2013年第11期270-277,共8页Transactions of China Electrotechnical Society

基　　金：国家自然科学基金(50777021);暨南大学人才引进计划资助项目

摘　　要：参数辨识是基于量测负荷建模的一项关键技术。针对微分进化算法进行静态负荷模型参数辨识收敛速度慢、种群规模大的问题,提出一种适合动态负荷模型参数辨识的微分进化算法。所提算法鲁棒性好,收敛速度快,全局寻优能力强,种群规模缩小,计算量降低。对某大城市两个220kV变电站实测扰动数据进行动态负荷模型参数辨识表明,计算精度明显优于遗传算法和Levenberg-Marquardt(L-M)算法相结合混合学习算法的计算精度,验证了所提算法的实用性。Parameter identification is the key technology in measurement-based load modeling. A differential evolution algorithm is proposed in this paper to identify dynamic load model＇s parameter, which overcomes the slowly convergent speed and oversize population problem existing in the differential evolution algorithm for static load model parameter identification. The proposed differential evolution algorithm shows good robustness, fast convergence, strong global optimization searching capability, and reduces population size and computation burden. Dynamic load model parameter identification for field measurement data from two substations in a metropolitan shows that the accuracy of the proposed algorithm is obviously better than the genetic algorithm and the Levenberg-Marquardt algorithm combined hybrid learning algorithm, and the practicability of the proposed algorithm is validated accordingly.

关 键 词：微分进化 动态负荷模型 参数辨识 

分 类 号：TM714[电气工程—电力系统及自动化]