Artificial Monitoring of Eccentric Synchronous Reluctance Motors Using Neural Networks

This paper proposes an artificial neural network for monitoring and detecting the eccentric error of synchronous reluctance motors. Firstly, a 15 kW motor is introduced took as case study to investigate effects rotor. Then, equivalent magnetic circuits studied are analyzed developed, in cases dynamic rotor static condition, respectively. After that, analytical equations derived, terms its air-gap flux density, electromagnetic torque, force, followed by finite element analyses. modal analyses stator whole performed, respectively, explore natural frequency shape motor, which further vibrational analysis possible be conducted. The vibration level housing furtherly relationship with eccentricity, validated prototype test. Furthermore, network, has 3 layers, proposed. By taking inputs, distance output, proposed trained till smaller than 5%. Therefore, this obtaining input parameters tested based on it automatically reporting upper-level control center.