Design and implementation of a variable speed drive system for single-phase induction motors

Single-phase induction motors are widely used in low power industrial and domestic applications where three phase is unavailable. It is required to control the speed of the motor when the application need variable speed. In this study, how the speed of a single phase induction motor is examined and the methods for speed change is given by considering its structure. Inverter topologies used in a single-phase motor drives and the modulation methods have been examined. For a variable-speed induction motor drive system, extreme learning machine based selective harmonic elimination pulse width modulation has been proposed. Extreme learning machine is a very fast learning algorithm developed for single hidden layer feed forward neural networks. The modulation index is used as an input to the extreme learning machine. The outputs of the extreme learning machine are the switching angles corresponding to this modulation index. The number of the outputs of the extreme learning machine depends on the number of harmonics to be eliminated. The proposed system has been simulated and and extensive tests have been made. In all cases selected low-order harmonics have been successfully eliminated. Simulation results were confirmed by experimental results. Finally, real time open loop V/f speed control of single-phase induction motor with the proposed extreme learning machine based selective harmonic elimination pulse width modulation was implemented and the performance was investigated and results were presented. A good speed tracking performance has been obtained for different reference speeds including change of direction of rotation at any speed. With the proposed system, both the speed control and the harmonic elimination were made for all speed range of single phase induction motor.