Scholarlink

<p>DC microgrids (DCMGs) are becoming very popular nowadays due to the number of advantages related to them compared to the existing AC systems. However, with the number of sources connected through power electronic converters in the DCMGs, maintaining the system’s stability is a tough challenge, especially when the system operates during the coupling mode. Hence, before DCMG is coupled, the resultant system stability must be guaranteed. This work proposes a novel system matrix for the resultant system, which does not require complete rebuilding when any new source is added. Furthermore, stability analysis of each source is presented by developing its transfer function and then using the eigenvalue method. Analysis of the proposed system is done by including line-to-ground and line-to-line faults under different loading conditions. Following the stability analysis, the machine learning techniques such as support vector machine (SVM), multilayer perceptron (MLP), cross-gradient boosting machine (XGBoost), and light gradient boosting machine (LGBM) with input features as voltage, current, power, and resistance values are utilized for the automated fault identification in two operational conditions such as isolated microgrid (MG) and integrated MG, respectively. The results reveal that the gradient boosting models have obtained the accuracy values of 98% which is higher than other machine learning techniques, such as SVM and MLP models, for recognizing faults in both MG conditions. The proposed approach provides a robust framework by combining the stability analysis and automated fault identification under variable conditions with load effect in MGs.</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

Integrated stability analysis and machine learning-based fault identification in microgrids during coupling operations and load effects

  • 期刊论文

  • 发表日期：

    2026年3月24日

查看全文
  • Rakesh Rajan Shukla,
  • Anup Kumar Panda,
  • Man Mohan Garg,
  • Rajesh Kumar Tripathy

摘要

DC microgrids (DCMGs) are becoming very popular nowadays due to the number of advantages related to them compared to the existing AC systems. However, with the number of sources connected through power electronic converters in the DCMGs, maintaining the system’s stability is a tough challenge, especially when the system operates during the coupling mode. Hence, before DCMG is coupled, the resultant system stability must be guaranteed. This work proposes a novel system matrix for the resultant system, which does not require complete rebuilding when any new source is added. Furthermore, stability analysis of each source is presented by developing its transfer function and then using the eigenvalue method. Analysis of the proposed system is done by including line-to-ground and line-to-line faults under different loading conditions. Following the stability analysis, the machine learning techniques such as support vector machine (SVM), multilayer perceptron (MLP), cross-gradient boosting machine (XGBoost), and light gradient boosting machine (LGBM) with input features as voltage, current, power, and resistance values are utilized for the automated fault identification in two operational conditions such as isolated microgrid (MG) and integrated MG, respectively. The results reveal that the gradient boosting models have obtained the accuracy values of 98% which is higher than other machine learning techniques, such as SVM and MLP models, for recognizing faults in both MG conditions. The proposed approach provides a robust framework by combining the stability analysis and automated fault identification under variable conditions with load effect in MGs.