Scholarlink

<p>Predictive maintenance has become a vital tool in minimizing expenses and operational setbacks while proactively averting potential failures. Its scope extends across various sectors, encompassing critical component upkeep crucial for ensuring public safety. Addressing the challenge of preempting catastrophic failures in diesel engines, this study uses a simulated dataset featuring 3500 realistic failure scenarios considering the engine cylinder, coupled with a crankshaft torsional vibration model. The research proposes employing artificial intelligence regression techniques, specifically support vector regression and Gaussian processes, to forecast diesel engine faults. This methodology is applied in conjunction with an engine simulator to evaluate its efficacy and precision. Notably, the Gaussian process regressor exhibits superior performance, achieving an RMSE value of 0.015&#xa0;±&#xa0;0.001%.</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

Predicting faults in diesel engines with kernel machines regression techniques

  • 期刊论文

  • 发表日期：

    2025年7月22日

查看全文
  • Denys P. Viana,
  • Dionísio H. C. de S. S. Martins,
  • Diego B. Haddad,
  • Fabrício L. e Silva,
  • Milena F. Pinto,
  • Ricardo H. R. Gutiérrez,
  • Ulisses A. Monteiro,
  • Luiz Vaz,
  • Thiago de M. Prego,
  • Fabio A. A. Andrade,
  • Luís Tarrataca,
  • Amaro A. de Lima

摘要

Predictive maintenance has become a vital tool in minimizing expenses and operational setbacks while proactively averting potential failures. Its scope extends across various sectors, encompassing critical component upkeep crucial for ensuring public safety. Addressing the challenge of preempting catastrophic failures in diesel engines, this study uses a simulated dataset featuring 3500 realistic failure scenarios considering the engine cylinder, coupled with a crankshaft torsional vibration model. The research proposes employing artificial intelligence regression techniques, specifically support vector regression and Gaussian processes, to forecast diesel engine faults. This methodology is applied in conjunction with an engine simulator to evaluate its efficacy and precision. Notably, the Gaussian process regressor exhibits superior performance, achieving an RMSE value of 0.015 ± 0.001%.