电气工程学报 ›› 2018, Vol. 13 ›› Issue (1): 16-22.doi: 10.11985/2018.01.003

• • 上一篇    

基于动态柯西蜂群算法优化支持向量机的风机叶片故障诊断

王宇鹏1,王致杰1,刘琦2,徐莉莉2,王鸿3,程亚丽1   

  1. 1. 上海电机学院电气学院上海 201306
    2. 上海电气风电设备有限公司上海 200241
    3. 山东科技大学电气与自动化工程学院青岛 266590
  • 收稿日期:2017-08-15 出版日期:2018-01-25 发布日期:2020-04-10
  • 作者简介:王宇鹏 男 1992年生,硕士研究生,研究方向为智能电网技术与故障诊断。|王致杰 男 1964年生,博士研究生,研究方向为风电机组、电力设备故障诊断。
  • 基金资助:
    自然科学基金资助项目(51477099);上海市自然科学基金资助项目(15ZR1417300);上海市闵行区科委产学研基金资助项目(V14MH166)

Fault Diagnosis of Wind Turbine BladeBased on Cauchy Artificial Bee Colony Algorithm Optimized Support Vector Machine

Yupeng Wang1,Zhijie Wang1,Qi Liu2,Lili Xu2,Hong Wang3,Yali Cheng1   

  1. 1. Shanghai Dianji University Shanghai 201306 China
    2. Shanghai Electric Wind Power Equipment Co. Ltd Shanghai 200241 China
    3. Shandong Science and Technology University Qingdao 266590 China
  • Received:2017-08-15 Online:2018-01-25 Published:2020-04-10

摘要:

为了提高风力发电机叶片故障的识别率,利用支持向量机建立风力发电机叶片故障和特征参数之间的非线性关系。在蜂群算法中引入一种动态柯西因子,动态调节蜂群寻优过程中的搜索步长,提高蜂群算法的扰动能力,避免蜂群陷入局部搜索,采用这种动态柯西蜂群算法对支持向量机的参数寻优,建立动态柯西蜂群算法优化的支持向量机模型。采集南方某风场风力发电机叶片的四种工况下的特征数据训练此模型并进行故障诊断,诊断结果表明改进后的蜂群算法优化支持向量机模型能够提高风力发电机叶片的故障识别率,具有一定的工程参考意义。

关键词: 动态柯西因子, 蜂群算法, 支持向量机, 风机叶片, 故障诊断

Abstract:

In order to improve the recognition rate of wind turbine blade fault diagnosis, the nonlinear relationship between fault diagnosis and characteristic parameters of wind turbine blade is established by using support vector machine. In the Cauchy artificial bee colony algorithm, a dynamic Cauchy factor is introduced to dynamically adjust the search step in the optimization process of the colony, to improve the perturbation ability of the colony algorithm, and to avoid the colony into the local search. A dynamic support vector machine model for dynamic Cauchy bee colony optimization is established. The model is trained and diagnosed by four conditions of a wind farm in the south. The results show that the support vector machine model can be improved based on the dynamic Cauchy artificial bee colony algorithm. Wind turbine generator fault recognition rate, with a certain engineering reference significance.

Key words: Dynamic Cauchy factor, artificial bee colony algorithm, support vector machine, wind turbine, bladefault diagnosis

中图分类号: