Scholarlink

<p>To address the challenges associated with gearboxes of underground belt conveyors—namely, their susceptibility to failure under high-impact loads, strong background noise, and non-stationary operating conditions, as well as the limitations of traditional diagnostic methods in feature extraction and model accuracy—this study proposes a fault diagnosis approach based on variational mode decomposition (VMD) integrated with an improved particle swarm optimization least squares support vector machine (IPSO-LSSVM). First, the original vibration signals are adaptively decomposed using VMD, and key mode components are selected and reconstructed based on kurtosis and permutation entropy criteria, achieving effective denoising and feature enhancement under complex operating conditions. Next, an LSSVM-based fault classification model is constructed, where an improved particle swarm optimization (IPSO) algorithm is employed to adaptively optimize the regularization and kernel parameters, thereby enhancing the model’s generalization and classification performance. Finally, a gearbox fault diagnosis test platform is established to collect vibration data under various representative fault scenarios for model validation. Experimental results demonstrate that the proposed method achieves a fault recognition accuracy of 96.85% across multiple fault modes, significantly outperforming traditional methods such as support vector machines (SVM), random forest (RF), and backpropagation neural network (BPNN), confirming its robustness and practical applicability in engineering environments.</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

Fault diagnosis of belt conveyor gearboxes based on Vibration signal processing and IPSO-optimized LSSVM

  • 期刊论文

  • 发表日期：

    2025年9月10日

查看全文
  • Qiang Zhang,
  • Jigeng Bing,
  • Ying Tian,
  • Yang Wang,
  • Xinye Liu

摘要

To address the challenges associated with gearboxes of underground belt conveyors—namely, their susceptibility to failure under high-impact loads, strong background noise, and non-stationary operating conditions, as well as the limitations of traditional diagnostic methods in feature extraction and model accuracy—this study proposes a fault diagnosis approach based on variational mode decomposition (VMD) integrated with an improved particle swarm optimization least squares support vector machine (IPSO-LSSVM). First, the original vibration signals are adaptively decomposed using VMD, and key mode components are selected and reconstructed based on kurtosis and permutation entropy criteria, achieving effective denoising and feature enhancement under complex operating conditions. Next, an LSSVM-based fault classification model is constructed, where an improved particle swarm optimization (IPSO) algorithm is employed to adaptively optimize the regularization and kernel parameters, thereby enhancing the model’s generalization and classification performance. Finally, a gearbox fault diagnosis test platform is established to collect vibration data under various representative fault scenarios for model validation. Experimental results demonstrate that the proposed method achieves a fault recognition accuracy of 96.85% across multiple fault modes, significantly outperforming traditional methods such as support vector machines (SVM), random forest (RF), and backpropagation neural network (BPNN), confirming its robustness and practical applicability in engineering environments.