Optimal Operation of Smart Home Appliances using Deep Learning

Abstract

This paper discusses an optimal operation of smart home appliances using deep learning. A yearly dataset was used to predict the day-ahead energy consumption pattern of household appliances. The preliminary findings indicate promising improvement in forecasting accuracies for smart home appliances. The forecasted result is integrated with Linear Programming based optimization model to make an appliance management system suitable for demand response. In addition, constraints for price, demand and equipment rating were used in the optimization model to generate the appliance schedule.