A QoE-Aware Approach for Smart Home Energy Management

Abstract

In this paper, a Quality of Experience (QoE)-aware Smart Home Energy Management (SHEM) system is proposed. Firstly, a survey has been conducted on 64 people to investigate the degree of satisfaction perceived when the starting time of appliances was postponed or anticipated with respect to the preferred time. Secondly, the results were clustered in different profiles using the k-means algorithm to control appliances' working time according to the detected user profile. Thirdly, a SHEM system is run that relies on two algorithms: the QoE-aware Cost Saving Appliance Scheduling (Q-CSAS) and the QoE-aware Renewable Source Power Allocation (Q-RSPA). The former is aimed at scheduling controllable loads based on users' profile preferences and Time-of-Use (TOU) electricity prices, thus taking into account the level of annoyance perceived when a task is postponed or anticipated. The latter re-allocates the starting time of appliances whenever a surplus of energy has been made available by Renewable Energy Sources (RES). This re-allocation takes place using a distributed max-consensus negotiation algorithm. The objective is that of scheduling the appliances starting time so that a trade-off between cost saving and annoyance perceived is achieved. As demonstrated by simulation results, the two algorithms ensure a cost saving that goes from 19% to 84% depending on the presence of RES, with a resulting average annoyance factor value of 1.01 to 1.03.