{"title":"液体射流冲击冷却的机器学习回归建模:基于计算流体力学(CFD)","authors":"Amirhossein Kholghi , Farzad Azizi Zade , Hamid Niazmand , Mohammad Sardarabadi","doi":"10.1016/j.ijthermalsci.2025.110086","DOIUrl":null,"url":null,"abstract":"<div><div>A comprehensive CFD and Machine Learning Regression Models (MLRM) investigation optimized circular Jet Impingement cooling. Jet impingement, influenced by several parameters, is widely studied in industry. This study used a design of experiments based on the Taguchi Method (TM) to determine the efficient number of CFD simulations. Numerical simulations generated a dataset to analyze nozzle diameter, nozzle height, flow rate, and different fluids (water, nanofluids, and Microencapsulated PCM), validated with experimental data. Data is cleaned and split into training, validation, and test sets. Validation and training data are augmented, while test data remained unchanged. After feature selection, 6 singular and 6 ensemble RMs are trained to identify the best models, followed by developing a novel hybrid model. The hybrid model achieved a total R<sup>2</sup> = 0.90 and test R<sup>2</sup> = 0.84. Applicability and sensitivity analysis validated the hybrid model, followed by a TM-based analysis of variance. Results revealed that flow rate is the most crucial factor (51.5%) followed by the fluid type (45.8%). Finally, several optimization methods are applied, with the Nelder-Mead method predicting the optimum case (with total error = 10%): Re = 5961.6 to result in an h<sub>average</sub> = 4.6 (W/cm<sup>2</sup> °C) given a heat flux of 64 W/cm<sup>2</sup>.</div></div>","PeriodicalId":341,"journal":{"name":"International Journal of Thermal Sciences","volume":"217 ","pages":"Article 110086"},"PeriodicalIF":4.9000,"publicationDate":"2025-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Machine learning regression modeling of liquid jet impingement cooling: Based on computational fluid dynamics (CFD)\",\"authors\":\"Amirhossein Kholghi , Farzad Azizi Zade , Hamid Niazmand , Mohammad Sardarabadi\",\"doi\":\"10.1016/j.ijthermalsci.2025.110086\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>A comprehensive CFD and Machine Learning Regression Models (MLRM) investigation optimized circular Jet Impingement cooling. Jet impingement, influenced by several parameters, is widely studied in industry. This study used a design of experiments based on the Taguchi Method (TM) to determine the efficient number of CFD simulations. Numerical simulations generated a dataset to analyze nozzle diameter, nozzle height, flow rate, and different fluids (water, nanofluids, and Microencapsulated PCM), validated with experimental data. Data is cleaned and split into training, validation, and test sets. Validation and training data are augmented, while test data remained unchanged. After feature selection, 6 singular and 6 ensemble RMs are trained to identify the best models, followed by developing a novel hybrid model. The hybrid model achieved a total R<sup>2</sup> = 0.90 and test R<sup>2</sup> = 0.84. Applicability and sensitivity analysis validated the hybrid model, followed by a TM-based analysis of variance. Results revealed that flow rate is the most crucial factor (51.5%) followed by the fluid type (45.8%). Finally, several optimization methods are applied, with the Nelder-Mead method predicting the optimum case (with total error = 10%): Re = 5961.6 to result in an h<sub>average</sub> = 4.6 (W/cm<sup>2</sup> °C) given a heat flux of 64 W/cm<sup>2</sup>.</div></div>\",\"PeriodicalId\":341,\"journal\":{\"name\":\"International Journal of Thermal Sciences\",\"volume\":\"217 \",\"pages\":\"Article 110086\"},\"PeriodicalIF\":4.9000,\"publicationDate\":\"2025-06-24\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"International Journal of Thermal Sciences\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S1290072925004090\",\"RegionNum\":2,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"ENGINEERING, MECHANICAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Thermal Sciences","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1290072925004090","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, MECHANICAL","Score":null,"Total":0}
Machine learning regression modeling of liquid jet impingement cooling: Based on computational fluid dynamics (CFD)
A comprehensive CFD and Machine Learning Regression Models (MLRM) investigation optimized circular Jet Impingement cooling. Jet impingement, influenced by several parameters, is widely studied in industry. This study used a design of experiments based on the Taguchi Method (TM) to determine the efficient number of CFD simulations. Numerical simulations generated a dataset to analyze nozzle diameter, nozzle height, flow rate, and different fluids (water, nanofluids, and Microencapsulated PCM), validated with experimental data. Data is cleaned and split into training, validation, and test sets. Validation and training data are augmented, while test data remained unchanged. After feature selection, 6 singular and 6 ensemble RMs are trained to identify the best models, followed by developing a novel hybrid model. The hybrid model achieved a total R2 = 0.90 and test R2 = 0.84. Applicability and sensitivity analysis validated the hybrid model, followed by a TM-based analysis of variance. Results revealed that flow rate is the most crucial factor (51.5%) followed by the fluid type (45.8%). Finally, several optimization methods are applied, with the Nelder-Mead method predicting the optimum case (with total error = 10%): Re = 5961.6 to result in an haverage = 4.6 (W/cm2 °C) given a heat flux of 64 W/cm2.
期刊介绍:
The International Journal of Thermal Sciences is a journal devoted to the publication of fundamental studies on the physics of transfer processes in general, with an emphasis on thermal aspects and also applied research on various processes, energy systems and the environment. Articles are published in English and French, and are subject to peer review.
The fundamental subjects considered within the scope of the journal are:
* Heat and relevant mass transfer at all scales (nano, micro and macro) and in all types of material (heterogeneous, composites, biological,...) and fluid flow
* Forced, natural or mixed convection in reactive or non-reactive media
* Single or multi–phase fluid flow with or without phase change
* Near–and far–field radiative heat transfer
* Combined modes of heat transfer in complex systems (for example, plasmas, biological, geological,...)
* Multiscale modelling
The applied research topics include:
* Heat exchangers, heat pipes, cooling processes
* Transport phenomena taking place in industrial processes (chemical, food and agricultural, metallurgical, space and aeronautical, automobile industries)
* Nano–and micro–technology for energy, space, biosystems and devices
* Heat transport analysis in advanced systems
* Impact of energy–related processes on environment, and emerging energy systems
The study of thermophysical properties of materials and fluids, thermal measurement techniques, inverse methods, and the developments of experimental methods are within the scope of the International Journal of Thermal Sciences which also covers the modelling, and numerical methods applied to thermal transfer.