An opposition-based learning enhanced artificial hummingbird algorithm for residential load scheduling considering various consumer behaviour models

Demand side management (DSM) is an essential measure to balance power supply and demand in present smart grid environment. Modern communication technologies and advanced metering infrastructure have made it possible for residential customers to take part in demand side control programs. The proposed work explores the use of Opposition based learning enhanced artificial hummingbird algorithm for implementing an intelligent residential load scheduling which is one of the DSM practices. To effectively address the challenging multi-objective optimization problem, the work uses an optimization strategy inspired by biology and strengthened by the opposition-based learning algorithm. It respects the fixed operation patterns of critical appliances while intelligently shifting the operation of flexible appliances within user-defined time limitations. Time-of-use electricity pricing, maximum power limitations and user comfort metrics are the objectives considered during the optimization process. Analysis of four different consumer types in a locality shows distinct optimization potential and adaptability to real life scenarios. The proposed algorithm achieves a reduction of 22.7% in peak to average ratio (PAR) and 35.7% reduction with integration of solar photovoltaic source. Furthermore, an average economic benefit of 22.39% is achieved across all the groups. The results demonstrate the efficiency of the algorithm through increased economic gains while preserving consumer satisfaction and reduced PAR.