Work

Development of Quantitative Autoencoder-Based Fault Diagnosis Method for Application in Conveyor System

Year: 2025

Type: conference-paper

Abstract: In Industry 4.0, efficient status monitoring, early warnings, failure diagnosis, and residual life prediction are crucial. Conveyor systems, as crucial components of many manufacturing processes, play a significant role in ensuring efficient and reliable production. Malfunctions or failures in conveyor systems can lead to decreased efficiency. To solve this problem, a quantitative autoencoder method is proposed to accurately assess the diagnostic performance of the model. Through multi-sensor experiments on a general conveyor platform, the effectiveness of the model in real factory environments is verified. Experimental results demonstrate that the model can accurately diagnose equipment failures and generate reliable equipment remaining useful life prediction curves.

Source: IIAE International Conference on Industrial Application Engineering (ICIAE)

DOI: 10.12792/iciae2025.072

Authors: Cheng-Hao Haung, Po-Cheng Yang, Yu-Ching Cheng, Chia-Ming Chang, Kuo-Shen Chen, Yun-Hui Liu

Institutions: National Cheng Kung University, Southern Taiwan University of Science and Technology

Language: English

Topic: Machine Fault Diagnosis Techniques

Subfield: Control and Systems Engineering

Field: Engineering

Domain: Physical Sciences

SDG: Industry, innovation and infrastructure

Open Access status: hybrid

Funder: National Science and Technology Council