Smart Home Energy Management System Using Least Square Regression Analysis

Abstract

Smart home is a residence with several electrical and electronic appliances that are capable of communicating with each other and can be controlled remotely from any room in the home or from any location in the world. Easy control of home appliances/devices and energy management has been the main goal that leads to the invention of smart homes. However, most of the systems developed for these homes are either complex or could not manage energy wastage efficiently which incurring more electricity bills cost. In this work, an intelligent home energy management system that is based on Least Square Regression (LSR) analysis is presented. The system is trained based on the historical data of occupant’s interaction with the appliances over a period of time. It monitors and computes the power consumption of home user over a period of time. This system takes decision and controlled the output using LSR based on what it learnt by alerting the home user on condition of accept or reject response through Android GUI Apps. The system performance evaluation based on the frequency prediction which is given as 0.77 RMSE, the activation time prediction is given as 127.89 seconds RMSE which is slightly above 2 minutes with a regression coefficient of (R=0.999988). The RMSE of 257.90 seconds for activation of duration prediction with regression coefficient analysis of (R= 0.989071).