APPARATUS OF INDUCTION MOTOR SPEED CONTROL USING NEURAL NETWORK

摘要

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a control apparatus for an induction motor, and more particularly, to an induction motor apparatus that is robust against load variations by using a neural network algorithm.;Conventional controllers such as the PI controller used for controlling the induction motor have a problem that the gain constant value of the controller must be adjusted in order to change the operating point and maintain the optimum operation state when the load fluctuates or the system parameters are changed It is holding.;In order to solve the above problems, the present invention has been made to solve the above problems, and it is an object of the present invention to provide a control apparatus and a control method thereof, An inverter unit for converting a direct current into an AC signal in accordance with the output of the space vector converting unit, a motor unit controlled by an output of the inverter unit, a current value supplied to the motor unit, And a parameter estimator for estimating a parameter of the electric motor based on a predetermined control algorithm and supplying the parameter to the controller, An evaluation function for minimizing the error of the reference trajectory is used, It characterized in that to provide a speed control device of an induction motor using a neural network comprising the integral term of the response error in the sampling instant, increment of the control input, and an error response.;The controller according to the present invention can solve the problem that the conventional controller can not operate properly when the operating point of the load changes or the like due to the fixed gain at a specific operating point, It is effective.;Induction motor, neural network, parameter estimation, response error, integral