Scholarlink

<p>In many different industries, converters are essential, which emphasizes the need for a reliable monitoring system. A converter monitoring system needs to be able to identify the deterioration of crucial converter components and diagnose errors in real time. In this paper, a novel method for fault prediction in a dual boost rectifier converter is presented. In the dual boost rectifier, various fault scenarios are simulated, resulting in different voltage signatures. For the purpose of fault prediction, these signals are then fed into a Support Vector Machine (SVM). Green Anaconda Optimization (GAO) is an algorithm used to minimize errors in the SVM weight parameters. The proposed methodology is implemented in MATLAB/SIMULINK, and performance is compared with existing techniques. Notably, the proposed method achieves an accuracy, precision, recall, and F1-score of 99.85, 99.55, 99.56, and 99.56%, respectively. In comparison to existing methods, the proposed approach exhibits a minimal mean absolute error (MAE) of 0.009% and root mean square error (RMSE) of 0.16%. Conversely, the KNN method among the existing ones records a lower accuracy of 99.65%, accompanied by a higher error rate with a maximum MAE of 0.028% and RMSE of 0.062%. The results conclusively establish the efficacy of the GAO-optimized SVM in detecting faults within the dual boost rectifier system.</p>

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

Predictive maintenance of dual boost rectifier: a green anaconda driven machine learning approach

  • 期刊论文

  • 发表日期：

    2025年2月22日

查看全文
  • Soumya Rani Mestha,
  • Nagesh Prabhu

摘要

In many different industries, converters are essential, which emphasizes the need for a reliable monitoring system. A converter monitoring system needs to be able to identify the deterioration of crucial converter components and diagnose errors in real time. In this paper, a novel method for fault prediction in a dual boost rectifier converter is presented. In the dual boost rectifier, various fault scenarios are simulated, resulting in different voltage signatures. For the purpose of fault prediction, these signals are then fed into a Support Vector Machine (SVM). Green Anaconda Optimization (GAO) is an algorithm used to minimize errors in the SVM weight parameters. The proposed methodology is implemented in MATLAB/SIMULINK, and performance is compared with existing techniques. Notably, the proposed method achieves an accuracy, precision, recall, and F1-score of 99.85, 99.55, 99.56, and 99.56%, respectively. In comparison to existing methods, the proposed approach exhibits a minimal mean absolute error (MAE) of 0.009% and root mean square error (RMSE) of 0.16%. Conversely, the KNN method among the existing ones records a lower accuracy of 99.65%, accompanied by a higher error rate with a maximum MAE of 0.028% and RMSE of 0.062%. The results conclusively establish the efficacy of the GAO-optimized SVM in detecting faults within the dual boost rectifier system.