A domain adaptation neural network for digital twin-supported fault diagnosis

TL;DR

This paper introduces a domain adaptation neural network framework using DANN to improve fault diagnosis accuracy in digital twin systems by effectively transferring knowledge from simulated to real-world data.

Contribution

It proposes a novel fault diagnosis framework based on DANN that enhances model performance across simulated and real data domains, addressing the sim-to-real gap.

Findings

DANN improves fault diagnosis accuracy from 70% to over 80%.

The framework outperforms traditional lightweight models like CNN, TCN, Transformer, and LSTM.

Experimental results validate the effectiveness of domain adaptation in digital twin applications.

Abstract

Digital twins offer a promising solution to the lack of sufficient labeled data in deep learning-based fault diagnosis by generating simulated data for model training. However, discrepancies between simulation and real-world systems can lead to a significant drop in performance when models are applied in real scenarios. To address this issue, we propose a fault diagnosis framework based on Domain-Adversarial Neural Networks (DANN), which enables knowledge transfer from simulated (source domain) to real-world (target domain) data. We evaluate the proposed framework using a publicly available robotics fault diagnosis dataset, which includes 3,600 sequences generated by a digital twin model and 90 real sequences collected from physical systems. The DANN method is compared with commonly used lightweight deep learning models such as CNN, TCN, Transformer, and LSTM. Experimental results show that incorporating domain adaptation significantly improves the diagnostic performance. For example, applying DANN to a baseline CNN model improves its accuracy from 70.00% to 80.22% on real-world test data, demonstrating the effectiveness of domain adaptation in bridging the sim-to-real gap.