Research On Double-fed Speed Regulation System Based On Neural Network Inverse Control Strategy And Simulation

The double-fed motor is a multivariable, nonlinear and strong coupling system, on the basis of vector control theory, this paper combined neural networks with inverse system, created a non-analytic forms of the inverse system, and this approach can play a role in feedback linearization and multi-variable decoupling, it is also simple and easy to understand.in this paper, with stator flux oriented vector control method, the neural network inverse system control strategy is applied to control the double-fed motor, which greatly improved the control performance of double-fed motor. After the analysis of mathematical model of asynchronous motor in three-phase stationary reference frame, two-phase stationary reference frame and the two-phase synchronous rotating reference frame, with stator flux oriented vector control, the mathematical model of double-fed motor on the two-phase synchronous rotating reference frame is given, and put speed and the stator flux as state variables, also the excitation current component and torque component of rotor current as control variables to prove the reversibility of double-fed motor, then model of double-fed motor inverse system based on BP neural network is constructed with MATLAB, and the inverse system is connected with original system to get pseudo-linear composite system, Thus the double-fed speed regulation system is controlled just by controlling the two first-order integral subsystem, then additional controllers are designed for the decoupled subsystems to achieve dynamic decoupling of speed and flux. On this basis, an overall model double-fed speed regulation system is built, and function of each sub-module is illustrate detail. The simulation result is proved that the method can achieve dynamic decoupling with speed and flux of the double-fed motor, thus achieving high performance of speed control.This paper designs various hardware devices of the double-fed speed regulation system on the basis of theoretical analysis. Including the main circuit, DSP control circuit board, signal detection circuit, drive circuit and fault signal processing circuit, and the function of each circuit is given, and parameters of the main device are calculated. Control chip is DSP TMS320F2808 with high performance by TI, to achieve full digital control for the inverter. The results of this paper provides reference value for further research and development of double-fed speed regulation system.