Communication Signal Identification For Switched Reluctance Motors Based On Fuzzy Neural Networks

In the last decade, the switched reluctance motor drive (SRD) is rapidly gaining importance application in variable-speed drive systems. It possesses many specific advantages such as simple structure, a wide speed range, superior performances of speed tuning, good reliability and high efficiency. SRD is a system of position closed loop. The accurate and real-time position information is very important for high performance operating of switched reluctance motor (SRM). Traditionally, the position information is provided by a mechanical rotor position sensor. However, the existence of position sensor not only brings on structure complication, but also weakens the reliability of the system at high speed. So, position sensorless operation has become main topic in the research field of SRM.Because of SRM's inherent nonlinearity, it is very difficult to derive a accurate analytic mathematical model for SRM. The applications of Fuzzy Neural Networks provide an efficient method for the modeling of the nonlinear system. FNN is a kind of neural network built by fuzzy rules. FNN possesses not only the advantages of fuzzy logic like clear structure and exact physical meaning, but also the merits of neural networks like strong ability to learning and rapid convergence. Also, FNN has an ability of excellent nonlinear identification. It has been applied to fields of controlling and modeling.In this paper, the position sensorless control methods of SRM are further discussed and summarized. Making reference to the theory of hall sensor, this paper presents a novel position sensorless control for SRM. The proposed method is based on FNN, which are built to directly map the relationship of phase current, flux linkage and phase commutation signals. After training FNN with off-line, phase commutation signals are identified by FNN. Then, mechanical sensor is expected to be substituted by FNN. To verify the feasibility and effectiveness of the method, the nonlinear inductance model of SRM is deeply analyzed. With measured flux linkage characteristic of prototype, the parameters of inductance model are identified by nonlinear least square method. Then, the dynamic simulation model of the 4-phase (8/6 poles) SRM and its drive system is established based on the environment of MATLAB/SIMULINK. The simulation results show that the dynamic model has high precision and is suitable for the research of controlling strategies. Secondly based on the dynamic simulation model, the presented method of position sensorless control is simulated during the starting period and steady running period. The simulation results prove that this method can exactly achieve phase commutation identification, and the system can operate steadily with the proposed position sensorless control method.Finally, the experiment system of SRM based on TMS320F2812 as the core of controller is introduced and supplies good foundation for further research.