Intelligent controller based on pattern recognition

Abstract

Due to the increase in the number of household electrical appliances, electricity consumption accordingly increases. Therefore, there must be a reasonable way to reduce electricity consumption to minimize electrical problems that occur during high demand periods on the grid without compromising user’s comfort. The basic aim of this project is to design a real-time controller to organize electricity consuming in households. This controller gets its feedback from an artificial neural network (ANN) that classify devices according to their profiles. This research will reduce the number of sensors to single current sensor attached to the main electrical unit and appliances can be categorized accordingly. The ANN gives very accurate results to recognize appliances that have different current consumption pattern .