Eccentricity Fault Diagnosis System in Three-Phase Permanent Magnet Synchronous Motor (PMSM) Based on the Deep Learning Approach.

This project framework is built around creating and testing a deep‑learning diagnostic system for eccentricity faults in three‑phase permanent‑magnet synchronous motors (PMSMs). Eccentricity faults—static, dynamic, or mixed—create uneven air‑gap spacing. That unevenness lowers motor performance and can eventually cause the stator and rotor to hit each other. The proposed method uses one or more of the following as input signals: stator current waveforms, vibration data, or changes in electromagnetic torque. A convolutional neural network (CNN) or a long short‑term memory (LSTM) network is then trained to pull out useful features directly from the raw or lightly pre‑processed time‑series data. This avoids the need for manually designed feature extraction. The framework

also shows how to model both healthy and faulty motor states. You can use finite‑element analysis (FEA) or a mathematical d‑q model for this step, and then feed the simulated data into the deep‑learning classifier. Performance is judged with classification accuracy, precision, and computational latency. These metrics are examined under different load levels and noise conditions to prove that the system stays reliable. Overall, the work aims to enable dependable predictive‑maintenance strategies for industrial motor drives.

Message: Hi FE, I need help with “Eccentricity Fault Diagnosis System in Three-Phase Permanent Magnet Synchronous Motor (PMSM) Based on the Deep Learning Approach.” in “Electrical Engineering”