Development On Intelligent Drive Control System Of Switched Reluctance Motor

Switched reluctance motor (SRM) drive system has become the powerfulcompetitor against the traditional AC and DC drive systems, due to its simplestructure, low cost and high efficiency in the broad range of speed. But the strongnonlinear electromagnetic characteristic of switched reluctance motor increases itscontrol difficulty,and satisfying characteristics can’t be easily obtained withconventional methods. So the study of intelligent control strategies for switchedreluctance motor drive system is focused on the field of switched reluctance motorsin recent years.In this paper, some materials about switched reluctance motor and its drivesystem are studied deeply. Aiming the changes of its inductance, winding currentsand torque at different stages, PWM Control stragery is adopted when high speed,and Chopped Current Control stragery is adopted when low speed, RBF NeuralNetworks control stragery is applied to the drive system of switched reluctancemotor. RBF neural network has many outstanding merits, for example, quicklearning speed and strong generalization ability. Satisfying control results call beobtained by combining RBF neural network with quick training arithmetic andappropriate control mode.At the beginning, this paper establishes the dynamic simulation models of the4-phase (8/6poles) SRM and its drive system based on MATLAB/SIMULlNK, andthen the research on intelligent control strategies is processed on it. A new solutionthat adaptive PWM speed control for switched reluctance motors based on RBFneural network is presented in the research of speed regulation. This method builds up a speed controller which is trained off-line in advance, making use of thepowerful approximating ability and fast convergence property of RBF neuralnetwork, Via combining with the on-line identification network, this controller canadapt its parameters to the change environment. The results of simulink experimentsshow that with this method, quick response speed, high control precision and goodadaptability can be achieved.Then, according to the control features of SRD system and control strategies thehardware circuits of SRD system are designed. The main hardware circuits includepower converter and drive circuit, position detection circuit, current detection circuit,power circuit, fault detection circuit and the minimal system circuit. On this basis,according to the thought of foreground/background modular programming and themain functions of SRD system, the software of SRD system is designed and the flowchart of main program and subprogram are provided.In this paper, the hardware design of the SRD based on TMS320LF2407DSP isalso introduced. Detection of the phase current and rotor position, as well as control ofspeed are realized by software. The experiment results show that the design of thesystem is rational, the motor runs stable and reliable, and meets the requirement ofdesign.