Direct torque adaptive vector neural control of a three-phase induction motor

The paper proposed a neural solution to the direct torque vector control of three phase induction motor including real-time trained RNN velocity controller and a hysteresis flux and torque controllers, which permitted the speed up reaction to the variable load. The basic equations and elements of the direct field oriented torque control scheme are given. The control scheme is realized by one RNN learned by a real-time BP algorithm and three FFNNs learned off-line by Levenberg-Marquardt algorithm with data taken from PI-control scheme simulations.