Energy-saving optimization of air-conditioning water system based on data-driven and improved parallel artificial immune system algorithm. (1st May 2023)

• Record Type: Journal Article
• Authors: Yang, Siyuan; Yu, Junqi; Gao, Zhikun; Zhao, Anjun
• Is Part Of: Energy conversion and management. Volume 283(2023)
• Publication Date: 2023-05-01
• Subjects: Air-conditioning water system -- Energy-saving optimization -- Artificial immune system -- Data-driven -- Optimal control; Direct energy conversion -- Periodicals; Energy storage -- Periodicals; Energy transfer -- Periodicals; Énergie -- Conversion directe -- Périodiques; Direct energy conversion; Periodicals; 621.3105
• DOI: 10.1016/j.enconman.2023.116902
• Languages: English
• ISSNs: 0196-8904

Abstract

Abstract: As the air-conditioning water system is designed according to the maximum load, the system will deviate from its optimum state while operating under partial load. Therefore, it is critical that the numerous operating parameters of the various equipments in the system are dynamically adjusted in an effective and timely manner to maximize the operational energy efficiency of the system. To this end, an improved parallel artificial immune system (IPAIS) algorithm is proposed to determine the optimal operating parameters of the system under different loads. Before optimization, the power consumption model is developed using generalized regression neural network (GRNN) combined with mechanism model for each kind of equipment in the system. Afterwards, the optimal control problem is described with the objective of minimizing the total power consumption of all equipments and considering the relevant constraints. Subsequently, the IPAIS is developed to solve the problem by introducing four improvement strategies. Finally, a simulation experiment is conducted using an actual case of an air-conditioning water system. The results show that the developed power consumption model performs well in accuracy, robustness and generalization ability, and the total system energy consumption is reduced by 15.19% after optimization. Meanwhile, the IPAIS is extended to five variants to confirm the functionality and effectiveness of each improved strategy. Furthermore, the optimization performance of IPAIS in the actual system is comprehensively verified and analyzed using an experimental platform. Compared with the comparison algorithms, IPAIS is able to achieve superior optimization results and presents significant advantages in convergence, robustness and computational complexity.