Fpga-based Sensorless Srd System Research

Switched Reluctance Motor (SRM for short) features simple structure, high reliability, etc. Composed of SRM, the speed governing system (Switched Reluctance Motor Device, SRD for short) has wide range of speed regulation, advanced performance in low speed, high torque output and good efficiency in the wide range of speed and power. These make SRD a kind of new promising speed governing system with high efficiency and energy saving ability. However, SRM has to be controlled under the precondition that the instantaneous rotor position is known, thus the extra position sensor complicates the structure of SRD and decreases the reliability of the system. Therefore, exploring useful sensorless-position-estimation SRD is an important and significant research work.On the basis of this background, this paper puts forward a SRM rotor position estimation method based upon Radial Basis Function (RBF for short) Neural Network. With the foundation of flux-link theory, this method establishes a flux calculation model by applying trapezoid formula, and through simulation by MATLAB/Simulink, obtains a RBF Neural Network based rule of mapping relation between rotor current, flux-link and rotor position. It is also implemented on the platform of the Field Programmable Gate Array (FPGA for short). The simulation and experiments show that this method has small error and high accuracy in position estimation, the control signal can punctually reflect the rotor position in real time, validating the feasibility and real time ability of the indirect position estimation method.Finally, this paper develops the hardware and software for the SRD control system using FPGA chip EP3C25. By making use of high-speed A/D converter AD9054BST-135 to measure current and voltage, regarding a 2.2 KW four phase 8/6 pole SRM as control plant, a sensorless-position-estimation SRD control system is realized on the basis of double loop of ASR (Automatic Speed Regulator) and ACR (Automatic Current Regulator) with PI algorithm. The experiment results demonstrate that:the system has big starting torque, hard mechanical characteristic and good convenience in speed regulation. The system also contains over-voltage and over-current protection circuit which greatly improves the stability and reliability of the system.