{"title":"Data-driven optimization of asymmetric curved winglets in fin-and-tube heat exchangers","authors":"Rishikesh Sharma, D.P. Mishra, Lakhbir Singh Brar","doi":"10.1016/j.swevo.2025.102056","DOIUrl":null,"url":null,"abstract":"<div><div>The drive to reduce energy consumption and enhance heat transfer capability makes the optimization of heat exchangers (HEs) crucial in modern thermal management systems. Recognizing the potential for significant energy savings, this study focuses on the novel asymmetrically placed curved winglets in fin-and-tube heat exchangers. A total of 2320 simulations, designed using a Latin hypercube sampling plan, have been performed. The dependent variables are calculated using computational fluid dynamics to train artificial neural networks that serve as a surrogate model for genetic algorithm (GA) to perform multi-objective optimization. The GA aims to maximize the enhancement factor (<em>η</em>) – a ratio of the Colburn and friction factors.</div><div>Since HEs operate over a range of Reynolds numbers (<em>Re</em>), all the previous optimization-based studies have been based on a single <em>Re</em> that raises several questions about the HE’s performance at off-design conditions (over a range of <em>Re</em>). Hence, the present optimization-based study considers three different (but fixed) <em>Re</em> values, followed by optimizing <em>Re</em> against each optimized winglet geometry. All the results are compared at design and off-design conditions. The cross-validation of the Pareto front points using CFD reveals a deviation of <5 %, indicating good predictive performance and consistency of the optimized datasets within the defined simulation framework. Compared to the baseline model without winglets, the optimized designs achieved <em>η<sub>max</sub></em> = 100.17 % and also outperformed under varied operating conditions. Hence, besides introducing a novel asymmetric design, this research provides guidelines on <em>Re</em>-based optimizations that could significantly improve HE performance in energy-dependent industries.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"97 ","pages":"Article 102056"},"PeriodicalIF":8.2000,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Swarm and Evolutionary Computation","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2210650225002147","RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE","Score":null,"Total":0}
引用次数: 0
Abstract
The drive to reduce energy consumption and enhance heat transfer capability makes the optimization of heat exchangers (HEs) crucial in modern thermal management systems. Recognizing the potential for significant energy savings, this study focuses on the novel asymmetrically placed curved winglets in fin-and-tube heat exchangers. A total of 2320 simulations, designed using a Latin hypercube sampling plan, have been performed. The dependent variables are calculated using computational fluid dynamics to train artificial neural networks that serve as a surrogate model for genetic algorithm (GA) to perform multi-objective optimization. The GA aims to maximize the enhancement factor (η) – a ratio of the Colburn and friction factors.
Since HEs operate over a range of Reynolds numbers (Re), all the previous optimization-based studies have been based on a single Re that raises several questions about the HE’s performance at off-design conditions (over a range of Re). Hence, the present optimization-based study considers three different (but fixed) Re values, followed by optimizing Re against each optimized winglet geometry. All the results are compared at design and off-design conditions. The cross-validation of the Pareto front points using CFD reveals a deviation of <5 %, indicating good predictive performance and consistency of the optimized datasets within the defined simulation framework. Compared to the baseline model without winglets, the optimized designs achieved ηmax = 100.17 % and also outperformed under varied operating conditions. Hence, besides introducing a novel asymmetric design, this research provides guidelines on Re-based optimizations that could significantly improve HE performance in energy-dependent industries.
期刊介绍:
Swarm and Evolutionary Computation is a pioneering peer-reviewed journal focused on the latest research and advancements in nature-inspired intelligent computation using swarm and evolutionary algorithms. It covers theoretical, experimental, and practical aspects of these paradigms and their hybrids, promoting interdisciplinary research. The journal prioritizes the publication of high-quality, original articles that push the boundaries of evolutionary computation and swarm intelligence. Additionally, it welcomes survey papers on current topics and novel applications. Topics of interest include but are not limited to: Genetic Algorithms, and Genetic Programming, Evolution Strategies, and Evolutionary Programming, Differential Evolution, Artificial Immune Systems, Particle Swarms, Ant Colony, Bacterial Foraging, Artificial Bees, Fireflies Algorithm, Harmony Search, Artificial Life, Digital Organisms, Estimation of Distribution Algorithms, Stochastic Diffusion Search, Quantum Computing, Nano Computing, Membrane Computing, Human-centric Computing, Hybridization of Algorithms, Memetic Computing, Autonomic Computing, Self-organizing systems, Combinatorial, Discrete, Binary, Constrained, Multi-objective, Multi-modal, Dynamic, and Large-scale Optimization.