Development of AI-based Diagnosis Model for On-line Fault Detection for Washing Machines

Quality inspection in the production lines of washing machines is very important since even minor defects inside a washing machine can escalate into major issues such as leaks and loud noises. Previous studies have explored various methods for fault detection in washing machines, including vibration signal analysis. In addition, artificial intelligence (AI) diagnostic models have been widely adopted in many industry fields, and significant research is being conducted to improve the performance of these models. In this study, we propose the use of AI models and their enhancements for fault detection in washing machines for quality assurance in the manufacturing stage. We extract damage-sensitive features from sound recordings of these machines on the manufacturing line and utilize supervised learning to train an AI model to detect faults. Performance enhancement via feature selection is also performed. The proposed methods are validated by applying and comparing the performance of various AI models on the data obtained from the manufacturing line.