Position Sensorless Detection of SRM Based on BP Neural Network

Theory of SRM position detection based on exciting pulse voltage to no conduction phase is analysised, and a combination with BP (back propagation) neural network model to predict SRM rotor position without sensorless is proposed. Training sample and test sample can be obtained by SRM detection system with position sensor, with response current as input, and the rotor position as output, a single input and a single output of BP neural network model is built. Mapping between response current and rotor position are obtained by training BP neural network, and the rotor position prediction is realized, test sample is used to verify the BP neural network model. Simulation results show that this approach improves the detection accuracy compared with the method by injecting exciting pulse voltage.

[1]  Longya Xu,et al.  Accurate rotor position detection and sensorless control of SRM for super-high speed operation , 2002 .

[2]  W. B. Lee,et al.  An improved sensorless driving method of switched reluctance motors using impressed voltage pulse , 2002, Proceedings of the Power Conversion Conference-Osaka 2002 (Cat. No.02TH8579).

[3]  Binggang Cao,et al.  Neural network control of electric vehicle based on position-sensorless brushless DC motor , 2007, 2007 IEEE International Conference on Robotics and Biomimetics (ROBIO).

[4]  Wang Xu.  POSITIONS DETECTION AND FORECAST OF SENSORLESS SWITCHED RELUCTANCE MOTOR , 2000 .

[5]  Yu-Bo Wang,et al.  Position sensorless control of SRMs based on novel BP Neural Network , 2010, 2010 International Conference on Computer, Mechatronics, Control and Electronic Engineering.

[6]  Longya Xu,et al.  Accurate rotor position detection and sensorless control of SRM for super-high speed operation , 2000, APEC 2000. Fifteenth Annual IEEE Applied Power Electronics Conference and Exposition (Cat. No.00CH37058).