Efficiency Optimization of 3-Ø Induction Motor Using Ann

Induction motors especially squirrel cage type are the most widely used motors in the industry for variable frequency applications after the development of power electronic devices. These motors consume 75% of total industrial electrical energy and hence it is of prime importance to improve its efficiency, so that a large amount of saving of electrical energy is possible as well as contribution to environmental aspects can also be made. It is seen that in many variable speed applications like HVAC, mining, textile, wood cutting etc. induction motors operate at light-load conditions for prolonged time duration frequently and they have very low efficiency (around 20 - 30% due to low power factor). So there is a possibility of improving efficiency in a significant way by providing optimal flux instead of rated flux especially in steady state. This paper presents an application of Artificial Neural Network (ANN) for efficiency improvement in steady-state and light-load condition in an IM drive system based on Nola theory. The ANN controller generates the optimal voltage and frequency which runs the induction motor at its optimal efficiency condition point for a particular speed and torque requirement of the load. Optimal voltage values are shown in results. A significant improvement in efficiency is seen.