Abnormal vibration signal generation method and system

The synchronized training of a GAN's discriminator and classifier addresses the scarcity of abnormal signal samples, enhancing the generation and classification of diverse abnormal patterns in industrial equipment, achieving improved accuracy and efficiency in fault detection.

TWI934691BActive Publication Date: 2026-08-01ASUSTEK COMPUTER INC
7 Cites 0 Cited by

Patent Information

Application Number
TW114124581
Authority / Receiving Office
TW · TW
Patent Type
Patents
Current Assignee / Owner
Filing Date
2025-06-30
Publication Date
2026-08-01
Estimated Expiration
2045-06-29

AI Technical Summary

Technical Problem

Existing methods for fault detection and classification of abnormal vibration signals in industrial equipment face challenges due to the scarcity of real abnormal signal samples, limitations in data annotation, and the instability of generative adversarial networks, leading to ineffective generation and classification of diverse abnormal patterns.

Method used

A method involving a generative adversarial network (GAN) with a synchronized discriminator and classifier, trained using a labeled dataset, to generate and classify abnormal vibration signals, optimizing both components through a total loss value to enhance data representation and classification accuracy.

Benefits of technology

The approach effectively expands the number of abnormal vibration signal samples and improves classification accuracy by jointly optimizing the generator and classifier, enabling accurate simulation and automatic classification of abnormal states.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

This disclosure discloses a method and system for generating abnormal vibration signals. The method includes the following steps: A training dataset includes multiple actual vibration signals, and the training dataset includes labeled categories for each actual vibration signal. A classifier for identifying vibration anomalies is pre-trained based on the training dataset. A generative adversarial network model including a generator, a discriminator, and a classifier is trained based on the training dataset to obtain a trained generator and a trained classifier. The discriminator and classifier are updated synchronously based on a total loss value. An abnormal vibration signal generation task is performed using the generator, or a vibration signal classification task is performed using the classifier.
Need to check novelty before this filing date? Find Prior Art