Scholarlink

<p>The coal-handling conveyor belt is a vital component in the maritime supply chain, which serves the crucial role of transporting materials. Failures in the components of the conveyor belt such as Induction Motor (IM) can result in increased safety hazards, extensive downtime, and substantial costs. Consequently, diagnosing the fault in the IM is crucial to avoiding disruptions in the operation. This study proposes an effective method of condition monitoring using the Dynamic Bond Graph (BG) -based IM modeling of the healthy and faulty IMs through the OpenModelica tool. The stator current obtained from the model is analyzed in the frequency domain using Fast Fourier Transform (FFT). The extracted features from the stator current are labelled into four different categories: healthy, Broken Rotor Bar, Stator Inter-turn short circuit (ITSC), and Dynamic Eccentricity (DE) are fed to an Extreme Learning Machine (ELM) for the classification of fault types. The ELM classifier demonstrates high classification accuracy of 95.92%, showing its effectiveness in fault classification. To validate its efficacy, the performance metrics of classifier is presented using violin plot. The proposed technique aims to improve predictive maintenance capabilities, reducing unplanned downtime and enhancing operational efficiency.</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

Health monitoring of industrial conveyor systems using causality-based fault detection and ELM classification

  • 期刊论文

  • 发表日期：

    2025年11月26日

查看全文
  • Lavanya Aruldass,
  • Revathi S

摘要

The coal-handling conveyor belt is a vital component in the maritime supply chain, which serves the crucial role of transporting materials. Failures in the components of the conveyor belt such as Induction Motor (IM) can result in increased safety hazards, extensive downtime, and substantial costs. Consequently, diagnosing the fault in the IM is crucial to avoiding disruptions in the operation. This study proposes an effective method of condition monitoring using the Dynamic Bond Graph (BG) -based IM modeling of the healthy and faulty IMs through the OpenModelica tool. The stator current obtained from the model is analyzed in the frequency domain using Fast Fourier Transform (FFT). The extracted features from the stator current are labelled into four different categories: healthy, Broken Rotor Bar, Stator Inter-turn short circuit (ITSC), and Dynamic Eccentricity (DE) are fed to an Extreme Learning Machine (ELM) for the classification of fault types. The ELM classifier demonstrates high classification accuracy of 95.92%, showing its effectiveness in fault classification. To validate its efficacy, the performance metrics of classifier is presented using violin plot. The proposed technique aims to improve predictive maintenance capabilities, reducing unplanned downtime and enhancing operational efficiency.