Online Inductance Estimation of PM-Assisted Synchronous Reluctance Motor Using Artificial Neural Network

A Permanent Magnet-Assisted Synchronous Reluctance Motor (PMASynRM) is favored because of its lower Permanent Magnet (PM) amount. They utilize more reluctance than PM torque compared to Interior Permanent Magnet Synchronous Motor (IPMSM) and Surface Permanent Magnet Synchronous Motor (SPMSM). Understanding the motor parameters at each operating point is crucial to achieving maximum efficiency. Variation of motor parameters due to temperature using offline models has been reported in the literature. These methods are computationally intensive, especially when the effect of cross-saturation is included in the models. Online parameter estimation is more impressive in real applications to develop a high-performance control technique as a result of these limitations, particularly when motors confront highly nonlinear structures such as PMASynRMs. An Artificial Neural Network (ANN) is presented in this paper for online estimation of inductances at the Rotor Reference Frame (RRF) to ensure optimum performance for model-based control systems. An online-tuned dynamic model is implemented by the proposed ANN and compared with Finite Element Analysis (FEA) data and experimental validation tests.