摘要
提出一种基于群体相似性组合模型的故障预警方案。利用动态时间规整算法分析风电机群的相似性,得到工作状态相似的风电机群;采用遗传算法优化的软模糊聚类划分工况,构建非线性状态估计模型;依据最大信息系数确定模型的辅助变量,基于诱导有序加权平均(induced ordered weighted averaging,IOWA)算子构建组合预测模型。实验结果表明,所提出的组合模型对风电机组齿轮箱早期温度预警是有效的。
A fault warning scheme based on group similarity combination model was proposed.The dynamic time warping algorithm was used to analyze the similarity of wind turbine cluster,and the wind turbine cluster with similar working state was obtained.Then the nonlinear state estimate technique model was constructed by using the soft fuzzy clustering method optimized by genetic algorithm.The auxiliary variables of the model were determined according to maximal information coefficient,and the combined prediction model was constructed based on induced ordered weighted averaging(IOWA)operator.The experimental results showed that the proposed combined model was effective for early temperature warning of wind turbine gearbox.
作者
甄成刚
张争鹏
郭东庆
刘贞辉
ZHEN Chenggang;ZHANG Zhengpeng;GUO Dongqing;LIU Zhenhui(School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China;Information Communication Branch,State Grid Shanxi Electric Power Company,Taiyuan 030021,China)
出处
《郑州大学学报（理学版）》
CAS
北大核心
2023年第2期88-94,共7页
Journal of Zhengzhou University:Natural Science Edition
基金
北京市自然科学基金项目(4182061)。
