ExplainabilityTransparencyTrustworthinesscs.AI
Fault Diagnosis using eXplainable AI: a Transfer Learning-based Approach for Rotating Machinery exploiting Augmented Synthetic Data
Lucas Costa Brito, Gian Antonio Susto, Jorge Nei Brito, Marcus Antonio Viana Duarte
October 6, 2022
Expert systems with applications
106 citations
3 influential
Key Takeaways
-
Synthetic data can address the scarcity problem in fault diagnosis: The paper demonstrates that artificially generated vibration signals can effectively train AI models for detecting equipment failures, overcoming the challenge that real faults occur too infrequently to collect sufficient labeled training data in production environments.
-
Explainability is critical for real-world AI adoption in industrial settings: By using Grad-CAM visualization with neural networks, the FaultD-XAI approach makes the AI's diagnostic decisions interpretable rather than "black-box," which the authors identify as essential for building user trust and confidence in automated fault diagnosis systems.
-
Transfer learning reduces labeling costs and expert dependency: The approach uses knowledge learned from synthetic data to diagnose real machinery faults, significantly reducing the need for expensive expert annotation of large datasets—a major barrier to implementing AI in production processes.
-
Combining multiple techniques improves both performance and trustworthiness: The integration of transfer learning, synthetic data augmentation, and explainable AI (Grad-CAM with 1D CNNs) addresses multiple practical challenges simultaneously—data scarcity, interpretability, and cost—rather than solving them in isolation.
Abstract
Artificial Intelligence (AI) is one of the approaches that has been proposed to analyze the collected data (e.g., vibration signals) providing a diagnosis of the asset's operating condition. It is known that models trained with labeled data (supervised) achieve excellent results, but two main problems make their application in production processes difficult: (i) impossibility or long time to obtain a sample of all operational conditions (since faults seldom happen) and (ii) high cost of experts to label all acquired data. Another limitating factor for the applicability of AI approaches in this context is the lack of interpretability of the models (black-boxes), which reduces the confidence of the diagnosis and trust/adoption from users. To overcome these problems, a new generic and interpretable approach for classifying faults in rotating machinery based on transfer learning from augmented synthetic data to real rotating machinery is here proposed, namelly FaultD-XAI (Fault Diagnosis using eXplainable AI). To provide scalability using transfer learning, synthetic vibration signals are created mimicking the characteristic behavior of failures in operation. The application of Gradient-weighted Class Activation Mapping (Grad-CAM) with 1D Convolutional Neural Network (1D CNN) allows the interpretation of results, supporting the user in decision making and increasing diagnostic confidence. The proposed approach not only obtained promising diagnostic performance, but was also able to learn characteristics used by experts to identify conditions in a source domain and apply them in another target domain. The experimental results suggest a promising approach on exploiting transfer learning, synthetic data and explainable artificial intelligence for fault diagnosis. Lastly, to guarantee reproducibility and foster research in the field, the developed dataset is made publicly available.