Bi-objective Optimal Scheduling of Smart Homes Appliances Using Artificial Intelligence

Abstract

The residential load shares a large portion of the total electricity consumption (about 27%). Management of power demand by residential consumers is considered difficult because it depends on consumers’ behavior. Dynamic pricing with demand side management programs can prove most suitable to manage home appliances. Two major objectives of residential load management are reducing electricity bills as well as peak load. But these two objectives are conflicting in nature, so it becomes difficult and confusing for consumers to manually decide the desired scheduling of appliances. In this research work, an algorithm is suggested for the desired scheduling of smart home appliances utilizing artificial intelligence-based approaches, namely, Cuckoo search (CS), particle swarm optimization (PSO), and hybrid GA-PSO. In this research work, customer baseline load (CBL)-centered electricity cost optimization model is implemented to achieve both the objectives simultaneously. The proposed algorithm has been successfully validated on smart home residential loads of various categories, as well as real data of static and dynamic electricity pricing applicable in two utilities, Tehran Power Distribution Company, Iran, and Kerala State Electricity Board (KSEB), India. The proposed algorithm validates its performance to minimize electricity cost and peak power demand simultaneously, and the optimal scheduling of appliances facilitates both the residential consumer and utilities.