{"title":"基于 CFD 和人工神经网络的单坡单盆太阳能蒸发器年度性能评估建模方法","authors":"Ashutosh Verma, Hardial Singh","doi":"10.1002/htj.23049","DOIUrl":null,"url":null,"abstract":"<p>Thermal performance modeling and performance prediction of a solar still which is single basin single slope (SBSS) for the typical climatic condition of India at Jalandhar (31.3260° N, 75.5762° E) is analyzed in the present work. A numerical investigation of an SBSS solar still is conducted during the month of June 2022 using the ANSYS Fluent 2021 computational fluid dynamics (CFD) package and artificial neural network (ANN) prediction model. A user define function is written and used in fluent to formulate the problem with 9-h solar radiation flux, on solar still glass surface. The simulation outcomes for surface temperature at three different water depths were compared with the existing experimental study. Water temperature and productivity of freshwater were well aligned with experimental results. Three-dimensional domain is used with a two-phase volume of fluid model for the condensation and evaporation processes in a solar still. The performance evaluation parameters, that is, coefficients of convective, evaporative, and radiative heat transfer, different temperature values, distillation output, and system efficiency were calculated numerically. The parametric analysis is expanded, and an ANN model in MATLAB R2020a is utilized to estimate yearly performance and reduce the high computational cost of numerical analysis. The data for solar radiation, design, and operational parameters are fed into the ANN model, and as a result, the water temperature at various depths is computed. The ANN model was trained using numerical results computed for the month of June and tested, showing 99.7% accuracy with CFD results. The annual performance of the solar still was evaluated using ANN models for 9 h of the day and with different boundary conditions. To decrease computational and experimental costs, the recommended technique of combined CFD and ANN models for computing the annual performance of (SBSS) solar still is the most effective option.</p>","PeriodicalId":44939,"journal":{"name":"Heat Transfer","volume":null,"pages":null},"PeriodicalIF":2.8000,"publicationDate":"2024-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"CFD and artificial neural network-based modeling approach for the annual performance assessment of single slope single basin solar still\",\"authors\":\"Ashutosh Verma, Hardial Singh\",\"doi\":\"10.1002/htj.23049\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p>Thermal performance modeling and performance prediction of a solar still which is single basin single slope (SBSS) for the typical climatic condition of India at Jalandhar (31.3260° N, 75.5762° E) is analyzed in the present work. A numerical investigation of an SBSS solar still is conducted during the month of June 2022 using the ANSYS Fluent 2021 computational fluid dynamics (CFD) package and artificial neural network (ANN) prediction model. A user define function is written and used in fluent to formulate the problem with 9-h solar radiation flux, on solar still glass surface. The simulation outcomes for surface temperature at three different water depths were compared with the existing experimental study. Water temperature and productivity of freshwater were well aligned with experimental results. Three-dimensional domain is used with a two-phase volume of fluid model for the condensation and evaporation processes in a solar still. The performance evaluation parameters, that is, coefficients of convective, evaporative, and radiative heat transfer, different temperature values, distillation output, and system efficiency were calculated numerically. The parametric analysis is expanded, and an ANN model in MATLAB R2020a is utilized to estimate yearly performance and reduce the high computational cost of numerical analysis. The data for solar radiation, design, and operational parameters are fed into the ANN model, and as a result, the water temperature at various depths is computed. The ANN model was trained using numerical results computed for the month of June and tested, showing 99.7% accuracy with CFD results. The annual performance of the solar still was evaluated using ANN models for 9 h of the day and with different boundary conditions. To decrease computational and experimental costs, the recommended technique of combined CFD and ANN models for computing the annual performance of (SBSS) solar still is the most effective option.</p>\",\"PeriodicalId\":44939,\"journal\":{\"name\":\"Heat Transfer\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":2.8000,\"publicationDate\":\"2024-03-28\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Heat Transfer\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://onlinelibrary.wiley.com/doi/10.1002/htj.23049\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"THERMODYNAMICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Heat Transfer","FirstCategoryId":"1085","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1002/htj.23049","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"THERMODYNAMICS","Score":null,"Total":0}
引用次数: 0
摘要
本研究分析了印度贾朗达尔(北纬 31.3260°,东经 75.5762°)典型气候条件下单盆单坡(SBSS)太阳能蒸发器的热性能建模和性能预测。在 2022 年 6 月期间,使用 ANSYS Fluent 2021 计算流体动力学(CFD)软件包和人工神经网络(ANN)预测模型对 SBSS 太阳能蒸发器进行了数值研究。在 Fluent 中编写并使用了一个用户定义函数,以 9 小时的太阳辐射通量来计算太阳能电池玻璃表面的问题。三个不同水深的表面温度模拟结果与现有的实验研究进行了比较。淡水的水温和生产力与实验结果完全一致。太阳能蒸发器中的冷凝和蒸发过程采用三维域和两相流体模型。对性能评估参数,即对流、蒸发和辐射传热系数、不同温度值、蒸馏产量和系统效率进行了数值计算。对参数分析进行了扩展,并利用 MATLAB R2020a 中的 ANN 模型对年度性能进行了估算,降低了数值分析的高计算成本。太阳辐射、设计和运行参数的数据被输入 ANN 模型,并由此计算出不同深度的水温。利用 6 月份的数值结果对 ANN 模型进行了训练和测试,结果显示与 CFD 结果的准确率为 99.7%。使用 ANN 模型对一天中 9 个小时和不同边界条件下太阳能蒸发器的年度性能进行了评估。为了降低计算和实验成本,建议采用 CFD 和 ANN 模型相结合的技术来计算(SBSS)太阳能蒸发器的年度性能,这是最有效的选择。
CFD and artificial neural network-based modeling approach for the annual performance assessment of single slope single basin solar still
Thermal performance modeling and performance prediction of a solar still which is single basin single slope (SBSS) for the typical climatic condition of India at Jalandhar (31.3260° N, 75.5762° E) is analyzed in the present work. A numerical investigation of an SBSS solar still is conducted during the month of June 2022 using the ANSYS Fluent 2021 computational fluid dynamics (CFD) package and artificial neural network (ANN) prediction model. A user define function is written and used in fluent to formulate the problem with 9-h solar radiation flux, on solar still glass surface. The simulation outcomes for surface temperature at three different water depths were compared with the existing experimental study. Water temperature and productivity of freshwater were well aligned with experimental results. Three-dimensional domain is used with a two-phase volume of fluid model for the condensation and evaporation processes in a solar still. The performance evaluation parameters, that is, coefficients of convective, evaporative, and radiative heat transfer, different temperature values, distillation output, and system efficiency were calculated numerically. The parametric analysis is expanded, and an ANN model in MATLAB R2020a is utilized to estimate yearly performance and reduce the high computational cost of numerical analysis. The data for solar radiation, design, and operational parameters are fed into the ANN model, and as a result, the water temperature at various depths is computed. The ANN model was trained using numerical results computed for the month of June and tested, showing 99.7% accuracy with CFD results. The annual performance of the solar still was evaluated using ANN models for 9 h of the day and with different boundary conditions. To decrease computational and experimental costs, the recommended technique of combined CFD and ANN models for computing the annual performance of (SBSS) solar still is the most effective option.