Abstract
The energy usage of parallel chillers systems accounts for 25-40 % of the total energy cost of a building. In light of global warming concerns and the need for energy conservation, it is essential to distribute the load of the parallel chillers systems effectively to achieve energy savings in buildings. Accordingly, this study presents a multi-strategy improved sparrow search algorithm (MSSA) to address optimization of the optimal chillers loading (OCL) problem. The proposed algorithm augments the basic sparrow search algorithm (SSA) by introducing the Sine chaotic map, Levy flight method, and Cauchy variation to enhance diversity, avoid local optima, and increase global and local search capacities. We use 9 benchmark functions to check the performance of the proposed MSSA algorithm, and the results are better than the selected algorithms such as particle swarm algorithm (PSO), harris hawks optimization (HHO), artificial rabbit optimization (ARO) and sparrow search algorithm (SSA). In addition, MSSA is applied to two typical cases to demonstrate its performance to optimal chillers loading and the results indicate that the MSSA outperforms similar algorithms. This study validates that MSSA can provide a promising solution to resolve the OCL problem.