Scholarlink

The aircraft engine plays a vital role in the aerospace field; any malfunction in an engine can lead to serious consequences, hence ensuring the safety and reliability of an aircraft engine is necessary for preventing accidents. The engine RUL (Remaining Useful Life) prediction helps in identifying potential issues and abnormalities in engines before they fail. This early detection allows timely maintenance and reduces the risk of unexpected engine failures so it is necessary to enhance overall safety and efficient usage of engines. The prediction of RUL is carried out using machine learning. First, the statistical formulas are used to increase the variability of the data and then sensors are selected using two majorly known methods—RReliefF and Mutual Information. They are then trained for four different ML models for predicting RUL. The Root Mean Squared Error (RMSE) and Maximum Amplitude Error (MAE) are then used to compare the ML models for RUL prediction.

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

Prediction of Remaining Useful Life (RUL) for an Aircraft Engine Using Various Machine Learning Algorithms

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Dhruvi Shrimali,
  • Swapnil Vegad,
  • Keval Bhavsar,
  • Umang Parmar

摘要

The aircraft engine plays a vital role in the aerospace field; any malfunction in an engine can lead to serious consequences, hence ensuring the safety and reliability of an aircraft engine is necessary for preventing accidents. The engine RUL (Remaining Useful Life) prediction helps in identifying potential issues and abnormalities in engines before they fail. This early detection allows timely maintenance and reduces the risk of unexpected engine failures so it is necessary to enhance overall safety and efficient usage of engines. The prediction of RUL is carried out using machine learning. First, the statistical formulas are used to increase the variability of the data and then sensors are selected using two majorly known methods—RReliefF and Mutual Information. They are then trained for four different ML models for predicting RUL. The Root Mean Squared Error (RMSE) and Maximum Amplitude Error (MAE) are then used to compare the ML models for RUL prediction.