Scholarlink

Conveyor belts play a critical role in material transportation, particularly in mines, where operational efficiency and reliability are paramount. Accurate prediction of conveyor belt degradation is essential for minimizing maintenance costs and preventing unexpected failures. This study uses pre-trained machine learning models, trained on historical failure data from a limestone surface mine in Poland, to predict conveyor belt degradation caused by material transportation at a mine. Historical records of selected input and output variables were analyzed to predict future belt damage-behavior. Four models were designed to predict five output variables associated with conveyor belt failures. Artificial neural networks (ANNs) were used to model failure size and maintenance requirements, leveraging past repair data, belt wear characteristics, and operational conditions. The results demonstrate the effectiveness of AI-based forecasting in optimizing maintenance schedules and enhancing conveyor belt longevity, contributing to more reliable predictive maintenance strategies.

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

Forecasting Conveyor Belt Performance in Future Using Pre-trained Models Based on Historical Failure Data

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Parthkumar Parmar,
  • Aleksandra Rzeszowska,
  • Anna Burduk,
  • Leszek Jurdziak

摘要

Conveyor belts play a critical role in material transportation, particularly in mines, where operational efficiency and reliability are paramount. Accurate prediction of conveyor belt degradation is essential for minimizing maintenance costs and preventing unexpected failures. This study uses pre-trained machine learning models, trained on historical failure data from a limestone surface mine in Poland, to predict conveyor belt degradation caused by material transportation at a mine. Historical records of selected input and output variables were analyzed to predict future belt damage-behavior. Four models were designed to predict five output variables associated with conveyor belt failures. Artificial neural networks (ANNs) were used to model failure size and maintenance requirements, leveraging past repair data, belt wear characteristics, and operational conditions. The results demonstrate the effectiveness of AI-based forecasting in optimizing maintenance schedules and enhancing conveyor belt longevity, contributing to more reliable predictive maintenance strategies.