Optimal chiller loading by improved sparrow search algorithm for saving energy consumption. (15th May 2023)

• Record Type: Journal Article
• Authors: Xue, Zhilu; Yu, Junqi; Zhao, Anjun; Zong, Yue; Yang, Siyuan; Wang, Meng
• Is Part Of: Journal of building engineering. Volume 67(2023)
• Publication Date: 2023-05-15
• Subjects: Optimal chiller loading -- Parallel chillers system -- Sparrow search algorithm -- Energy saving; Building -- Periodicals; 690.05
• DOI: 10.1016/j.jobe.2023.105980
• Languages: English
• ISSNs: 2352-7102

Abstract

Abstract: Aims to reduce the energy consumption in a heating, ventilation, and air conditioning (HVAC) system with improper load distribution. An efficient optimization method, the improved sparrow search algorithm, has been developed to address the optimal chiller loading (OCL) problem for parallel chillers system. In general, the optimization goal is to minimize the system's energy consumption subject to meeting load demands, and the partial load rate of each chiller is taken as the optimization variable. This algorithm utilizes the Circle chaotic mapping to initialize the position to improve the quality and diversity of the initial solution. Meanwhile, to improve the algorithm's optimization accuracy, the information exchange reinforcement mechanism of the Gray Wolf Optimizer is used to update the producer position. Besides, the chaotic sine cosine mechanism is combined with the scrounger position update to enhance the convergence speed of the algorithm. Eventually, three case studies are selected to evaluate the performance of ISSA in detail and compared with other optimization algorithms. Experimental simulation verified that the improved sparrow search algorithm is more energy-saving and has the advantages of fast convergence, short running time and good robustness. Highlights: An ISSA algorithm is applied to solve the optimal chiller loading problem. The three improved strategies are introduced to obtain better operating results. Detailed experiments have been conducted for the parameter setting. The experimental results indicate that the ISSA can find a better optimal solution. The convergence, algorithmic complexity and robustness of the ISSA are tested.