Towards a thermodynamically favorable dew point evaporative cooler via optimization. (1st January 2020)

• Record Type: Journal Article
• Authors: Lin, Jie; Wang, Ruzhu; Li, Chuan; Wang, Sibao; Long, Jianyu; Chua, Kian Jon
• Is Part Of: Energy conversion and management. Volume 203(2020)
• Publication Date: 2020-01-01
• Subjects: Dew point evaporative cooling -- Multi-objective optimization -- Thermodynamic performance -- Genetic algorithm; 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.2019.112224
• Languages: English
• ISSNs: 0196-8904

Abstract

Highlights: A robust optimization framework is proposed for the dew point evaporative cooler. Two optimization methods are developed using genetic algorithm. A 2-D thermodynamic model is coupled with the optimization algorithms. The optimization results from the two methods are discussed and compared. The multi-to-single-objective optimization simplifies the optimization complexity. Abstract: Dew point evaporative cooling is an essential alternative to conventional vapor compression chillers in air conditioning systems to reduce electricity consumption and carbon emission. In this paper, we propose a robust optimization framework of the dew point evaporative cooler towards favorable dew point effectiveness, cooling capacity and Coefficient of Performance (COP). Two optimization algorithms, i.e., multi-to-single-objective and multi-objective optimizations, are developed using genetic algorithm. Concurrently, a 2-D thermodynamic model is established for a counter-flow dew point evaporative cooler and coupled with the optimization algorithms. The model agrees well with the experimental tests on a cooler prototype, and the maximum discrepancy is within ±4.7%. The proposed optimization study is then carried out to investigate the ultimate objective functions and their corresponding decision variables. Key findings that have emerged from this study reveal that the multi-to-single-objective optimization is able to obtain appropriate objective functions according to predefined preference with less complexity and faster response, compared to a multi-objective optimization. The optimal channel length and working ratio are found to be constant in different scenarios, i.e., at 0.50 m and 0.40, respectively, hence they can be eliminated to reduce the number of decision variables.