Scholarlink

This research paper endeavors to develop and compare predictive models for estimating the Remaining Useful Life (RUL) of manufacturing and engineering systems through the utilization of sensor data. The datasets employed in this study encompass simulated operational and sensor measurements gathered from machinery subjected to diverse operational conditions and fault modes. The primary objective of this investigation is to ascertain the efficacy of conventional machine learning (ML) techniques, specifically Long Short-Term Memory networks (LSTMs), for time series analysis and RUL prediction. Furthermore, generative AI models, such as Generative pretrained transformers (GPTs), are explored for their potential to augment fault detection and RUL estimation accuracy.

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

Enhancing Remaining Useful Life Prediction: A Comparative Study of Classical Machine Learning and Generative AI

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Gundelly Siddartha Yadav,
  • Ravi Katukam

摘要

This research paper endeavors to develop and compare predictive models for estimating the Remaining Useful Life (RUL) of manufacturing and engineering systems through the utilization of sensor data. The datasets employed in this study encompass simulated operational and sensor measurements gathered from machinery subjected to diverse operational conditions and fault modes. The primary objective of this investigation is to ascertain the efficacy of conventional machine learning (ML) techniques, specifically Long Short-Term Memory networks (LSTMs), for time series analysis and RUL prediction. Furthermore, generative AI models, such as Generative pretrained transformers (GPTs), are explored for their potential to augment fault detection and RUL estimation accuracy.