Pareto local search for a multi-objective demand response problem in residential areas with heat pumps and electric vehicles

In future energy systems characterized by significant shares of fluctuating renewable energy sources, there is a need for a fundamental change in electricity consumption. The energy system must adapt to the intermittent generation of renewable energy sources. This can be achieved by using flexible electrical loads, such as heat pumps and electric vehicles, with efficient control methods. In this paper, we introduce the Pareto local search method PALSS which employs heuristic search operations to solve the multi-objective demand response problem in residential areas, resulting in superior performance to existing approaches. Flexible electrical loads are shifted with the objective of minimizing the electricity cost and peak load while maintaining the inhabitants’ comfort in favorable ranges. Furthermore, we extend PALSS by incorporating reinforcement learning into the search operations in the approach RELAPALSS. For the evaluation, we employ the dichotomous method to obtain solutions that are guaranteed to be Pareto-optimal, serving as benchmarks. The results demonstrate that PALSS outperforms state-of-the-art multi-objective evolutionary algorithms by 16% (18% for RELAPALSS) regarding the performance indicator Generational Distance and by 128% (130% for RELAPALSS) for the indicator Hypervolume. The runtime for PALSS and RELAPALSS is reduced by up to 92% compared to exact methods. • Novel local search method for controlling heat pumps and electric vehicles. • Multiobjective optimization for demand response in residential areas. • Integrating reinforcement learning into the local search operations. • Developed methods strongly outperform state-of-the-art metaheuristics.