This paper proposes a method for the modeling of induction motor using neural networks. The use
of artificial neural networks (ANN) in this paper is due to their benefits such as fast computation,
ability of implementation, and easy interpolation. The neural network is trained by an experimental
data over a wide range of voltages, frequencies and load torques. At each value of frequency and
applied voltage the motor is loaded from no-load to full load. Therefore, the training data cover all
the possible operating range of the motor under different frequencies and voltages. The advantages
of the proposed model using ANN, is that the knowledge of the electrical parameters of the motor
is not necessary, and the change in these parameters with operating conditions doesn't effect on
performance of motor model. The inputs of the ANN motor model are voltage, frequency and
current and the model gives the motor speed, torque and efficiency as the outputs. By these three
values another motor performance characteristics can be calculated. The proposed model with
(ANN) can be used in speed and torque sensorless estimation depending on the motor inputs,
which will give an efficient and better performance for an induction motor.