一种基于STL-PSO-SVDD算法的风机发电机轴承健康指标算法  

A STL-PSO-SVDD algorithm of wind turbine generator bearing

作　　者：成骁彬 Cheng Xiaobin(Shanghai Electric Wind Power Group,Shanghai 200241,China)

机构地区：[1]上海电气风电集团股份有限公司,上海200241

出　　处：《无线互联科技》2023年第22期108-111,共4页Wireless Internet Technology

摘　　要：风电作为可再生资源的重要组成部分,已获得了广泛的关注。风力发电量多寡取决于风机的状态运维,而风机发电机轴承更换占风机停机维保项目的比例较大。文章对于风机发电机轴承进行监测,提出了一种新的发电机轴承健康指标算法。该算法分为3步:第一步对原始数据进行时域特征的提取;第二步将已提取的时域特征构造成时间序列数组并提取时间序列特征;第三步,使用SVDD算法(Support Vector Domain Description)将时域特征和时间序列特征拟合成一个的健康区间,并对其进行监测。同时,使用粒子群优化算法(PSO)优化SVDD算法和STL参数。以来自风场的实际案例来验证所提出的方法,同业界振动均方根指标相比,本算法能提前得到报警信息。As one of the most crucial parts in renewable energy,wind power has widely aroused concern.The power output of wind turbine is based on the health state of wind turbine.The majority of wind turbine shut-down is in generator bearing.In this study,we propose a novel health index algorithm to monitor wind turbine generator.Firstly,we calculate time-domain indices from raw data.Then,time-series indices are from time-domain indices.Finally,SVDD(Support vector domain description)is utilized to get health index from time-domain and time-series indices.The parameters of SVDD algorithm and STL algorithm are optimized via Particle Swarm Optimization algorithm(PSO).A case study is discussed to validate the proposed health index to monitor wind turbine generator bearing.

关 键 词：时域 时间序列 支持向量描述 发电机轴承 粒子群优化算法 

分 类 号：TK8[动力工程及工程热物理—流体机械及工程]