电网技术2016，Vol.40Issue(8)：2376-2382,7.DOI:10.13335/j.1000-3673.pst.2016.08.019
同步发电机非线性模型的期望最大辨识
Maximum Expectation Identification of Nonlinear Model of Synchronous Generator
摘要
Abstract
In order to realize on-line identification of synchronous generator nonlinear model under noise disturbance,maximum expectation identification method of the nonlinear model of was proposed.Firstly,nonlinear discrete state space model was constructed,including voltage regulator,and synchronous generator parameter identification problem was transferred to maximum expectation problem.Secondly,quantitative evaluation coefficient was proposed to compare difference between measured and output values of parameter variation.Particle filter and particle smoother was used to construct conditional data of synchronous generator,and gradient search method was used to get parameter identification algorithm of synchronous generator.Finally,simulation was carried out to identify experiment parameters.Simulation results verify effectiveness of the proposed method.关键词
同步发电机/参数辨识/期望最大/梯度搜索/轨迹灵敏度/主导参数Key words
synchronous generator/parameter identification/expectation maximization/gradient search/trajectory sensitivity/dominant parameter分类
信息技术与安全科学引用本文复制引用
邱健,于海承,环加飞,牛琳琳,严剑峰,于之虹,田芳,李亚楼..同步发电机非线性模型的期望最大辨识[J].电网技术,2016,40(8):2376-2382,7.基金项目
国家自然科学基金项目(61471328).Project Supported by National Natural Science Foundation of China (61471328). （61471328）
