Fault Diagnosis using eXplainable AI: A transfer learning-based approach for rotating machinery exploiting augmented synthetic data

Due to the growing interest for increasing productivity and cost reduction in industrial environment, new techniques for monitoring rotating machinery are emerging. 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. Three major problems hinder the application of AI models in industrial environments, being the motivation for the research: (i) impossibility or long time to obtain a sample of all operational conditions (since faults rarely happen), (ii) high cost of experts to label all acquired data (in a supervised scenario) and (iii) lack of interpretability of the models (black-boxes). 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, namely 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 post hoc 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 obtained on three datasets containing different mechanical faults suggest the method offers 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.