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

2025-05-27

AI Summary

In digital twin–based fault diagnosis, performance degradation arises from distributional shift (sim-to-real gap) between simulation and real-world data. To address this, this paper proposes a domain-adversarial neural network (DANN)–based cross-domain diagnosis framework—the first application of DANN to digital twin–driven fault diagnosis. The framework features a lightweight, embeddable architecture compatible with mainstream time-series models, including CNN, TCN, Transformer, and LSTM. Evaluated on a robotic fault dataset, the DANN-enhanced CNN improves real-world diagnostic accuracy from 70.00% to 80.22%, outperforming all baseline methods. This work effectively mitigates domain shift and establishes a novel paradigm for reliable transfer of simulation-derived knowledge to physical systems within digital twin frameworks.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Web applications in cross-disciplinary domains and verticals such as mixed reality, smart cities, and digital health
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Web Mining and Content Analysis: Web data generation and simulation

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.

Problem

• Bridges performance gap between simulated and real-world fault diagnosis
• Enables knowledge transfer from digital twin to physical system data
• Improves accuracy of deep learning models in real scenarios

Innovation

• Domain-Adversarial Neural Networks for fault diagnosis
• Digital twin-generated data for model training
• Bridging sim-to-real gap with domain adaptation