Bobechko Y. Synthesis and analysis of neurocontroller for a switched reluctance motor control system

The thesis is devoted to solving a scientific and applied problem of synthesis of the sensorless rotor position identification system and control of a switched reluctance motor based on the theory of artificial neural networks. For this purpose a series of neural networks, the common feature of which is a set of input variables - phase currents and sometimes supply voltage, was synthesized. Structures of the neuro-estimator of the rotor position angle, the neurocontroller to determine the signals to the control keys and the neurocontroller to identify the rotor position, sections switching-on and commutation angles, the values of which allow to form a flat speed-torque characteristics, were proposed.