Scholarlink

The gearbox, critical for wind turbines, often develops faults, underscoring the importance of research into fault diagnosis methods. Addressing Neutrosophic K-Nearest Neighbor (NKNN) algorithm limitations, like uncertain membership weights and low diagnostic accuracy, we propose PSO-NKNN. This novel approach integrates wavelet packet analysis and particle swarm optimization, involving fault feature extraction, outlier noise elimination, and model development. Experimental validation with real QPZZ-II platform data shows PSO-NKNN rectifies NKNN's limitations and markedly improves classification accuracy, enhancing noise resistance.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Optimized Fault Diagnosis Method for Wind Turbine Gearbox Using PSO-Based Neutrosophic K-Nearest Neighbor Algorithm

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Kun Tian,
  • Yunfei Ding,
  • Qifan Chen,
  • Qiancheng Sun

摘要

The gearbox, critical for wind turbines, often develops faults, underscoring the importance of research into fault diagnosis methods. Addressing Neutrosophic K-Nearest Neighbor (NKNN) algorithm limitations, like uncertain membership weights and low diagnostic accuracy, we propose PSO-NKNN. This novel approach integrates wavelet packet analysis and particle swarm optimization, involving fault feature extraction, outlier noise elimination, and model development. Experimental validation with real QPZZ-II platform data shows PSO-NKNN rectifies NKNN's limitations and markedly improves classification accuracy, enhancing noise resistance.