Scholarlink

The use of anomaly detection in solar power systems is critical for ensuring the reliability of solar energy. This paper compares several machine learning approaches, including SVM, K-Means, and HMM, using large datasets from two solar power stations in India. These techniques help detect issues like reduced power generation, unfavorable weather conditions, and equipment failure. Performance characteristics such as accuracy are discussed for each anomaly type, with SVM used for classification, K-Means for pattern recognition, and HMM for time-series analysis. The study aims to improve the reliability and optimize the performance of solar energy systems.

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

Study on Anomaly Detection in Solar Power Conversion Systems Using Machine Learning Technique

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • N. Deeraj,
  • R. S. Subhashree,
  • Solasa Venkata Naga Mohana Krishna,
  • M. VenkateshKumar

摘要

The use of anomaly detection in solar power systems is critical for ensuring the reliability of solar energy. This paper compares several machine learning approaches, including SVM, K-Means, and HMM, using large datasets from two solar power stations in India. These techniques help detect issues like reduced power generation, unfavorable weather conditions, and equipment failure. Performance characteristics such as accuracy are discussed for each anomaly type, with SVM used for classification, K-Means for pattern recognition, and HMM for time-series analysis. The study aims to improve the reliability and optimize the performance of solar energy systems.