Answers are grounded in this paper and cite their sources.
Faults in induction motors will reduce the operational efficiency and stability of mechanical systems. In recent years, artificial intelligence (AI) models have become an important technology for diagnosing induction motor faults. These AI models can only accurately identify faults when trained on a complete dataset. However, it is difficult to obtain a complete dataset in practical engineering scenarios, which will worsen the performance of AI models. To solve the problem, a numerical model-aided intelligent diagnosis method is proposed. Firstly, a numerical model of the health induction motor is established, and the model is updated using measured health signals. Subsequently, various types of fault mathematical models are injected into the healthy numerical model to obtain simulated fault samples. Finally, the simulated and/or measured signals are used to train the CNN. The experimental results show that the model has achieved satisfactory results to construct a relative complete dataset for engineering applications.
No takes yet. Share an insight, caveat, or question.
Fang et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: