Model-based optimal load control of inverter-driven air conditioners responding to dynamic electricity pricing

Abstract

Dynamic electricity pricing provides a great opportunity for residential consumers to participate into demand response (DR) programs to reduce the electricity bills. The lack of automatic response to time-varying electricity prices is one of the challenges faced by the residential electric appliances. Most of the existing studies on DR of residential air conditioner (ACs) focus on the single-speed ACs, rather than the inverter-driven ACs which are more energy efficient and extensively installed in today's residential buildings. This paper presents a novel model-based optimal load scheduling method for residential inverter-driven ACs to realize automatic DR to the day-ahead dynamic electricity prices. The models of the inverter-driven ACs and room thermal dynamics are firstly developed, identified and integrated for the development of the model-based scheduling. The tradeoff problem between the electricity costs, resident's comfort and peak power reductions is formulated as a mixed-integer nonlinear programming problem with adjustable weightings. The optimal solution of the nonlinear programming problem is searched by the genetic algorithm (GA). Simulation results show that multiple goals can be achieved via GA optimization and the regulations of the weights in the objective function. The developed framework can be implemented in the programmable communicating thermostats (PCTs) or the smart home energy manage systems (HEMSs) to enable residential inverter-driven ACs automatically respond to day-ahead dynamic electricity pricing.