Research Of Switched Reluctance Motor Drives And Novel Control System

Switched reluctance motor (SRM) take on the merits such as simple structure, low-cost, flexible control, etc. Switched reluctance drives system (SRD) has the advantage that conventional AC or DC drives system may have not, accordingly, as a emerging technology, theory and application research for SRD need to be studied. This dissertation is focused on the theory of intelligent control and control system design for SRM.Firstly, This thesis analyzes PWM control method of SRM and its convert circuit characteristics, and designed the control system based on TMS320F240 DSP controller. Then, the thesis analyzes the nonlinear magnetic characteristics of SRM based on the local approach Artificial Neural Networks (ANN) theory, and established the model for SRM using Radial Basis Function (RBF). Combined with the sampling dada for a SRM prototype, the simulation results is given and proved that the ANN model of SRM has higher precision and robustness compared with conventional method, while possessing the rapid and simple characteristics. In order to minimize the torque ripple of SRD, the dissertation further design the intelligent torque control system, introduce the instantaneous torque control strategy based on torque distribute function, which is implemented by the flux linkage controller of SRM. Furthermore, considering the multi control parameter, the article also found the rotor position angle controller based fuzzy logic so as to optimize the whole control system and effectively reduce the torque ripple. With regard to the converter of SRD system, the thesis firstly gives the study of the topology of converters that have the least main switches, and then lucubrates the single switch per phase converter with C-dump energy recovery circuit. The thesis analyzes the operation mode of the converter and gives the formula for estimation of the converter-related parameters. All these works guide the design for converter of SRD.Finally, the dissertation gives the research about the sensorless method for estimate the rotor position for SRD system. A sliding-mode observer (SMO) for estimate rotor position is discussed in the study, the SRD control system based on the SMO is designed. The simulation and experiment have been set up using DSP solution, the result validates the robustness and precision of the scheme for estimation of rotor position, while minimizing the torque ripple and optimizing other performance.