Performance analysis of a parallel-counterflow vortex tube using machine learning methods.
Saved in:
| Title: | Performance analysis of a parallel-counterflow vortex tube using machine learning methods. |
|---|---|
| Authors: | Korkmaz, Murat1 (AUTHOR) mkorkmaz@hacettepe.edu.tr, Doğan, Ayhan2 (AUTHOR) aydogan@hacettepe.edu.tr, Kırmacı, Volkan3 (AUTHOR) volkankirmaci@bartin.edu.tr |
| Source: | Journal of Thermal Analysis & Calorimetry. May2025, Vol. 150 Issue 10, p7901-7919. 19p. |
| Subjects: | Machine learning, Vortex tubes, Fluid flow, Fluids, Sensitivity analysis, Cooling, Heat transfer, Experimental design |
| Abstract: | In this study, these machine learning methods (MLMs) were used for the first time in the literature to compare the temperature performance of two counterflow Ranque–Hilsch vortex tubes (RHVT) connected in parallel. As input parameters, two different pressurized fluids, three distinct materials, and six types of nozzles were selected, while the temperature difference was obtained from the hot and cold fluid outlets was designated as the output parameter. For the first time under these experimental conditions, a sensitivity analysis was conducted to determine the impact of the input parameters on the output. For each pressurized fluid and material, six models were developed using MLMs such as elastic net (EN), Bayesian ridge (BR), category boosting (CB), and light gradient boosting machine (LightGBM). A total of 30 different experimental setups were established by creating five separate setups for each model. The performance of these models was evaluated and compared using the K-fold cross-validation method. Upon analyzing the results, the BR method demonstrated the best performance for Model 1 (air, aluminum), Model 3 (air, polyamide), Model 4 (oxygen, aluminum), Model 5 (oxygen, brass), and Model 6 (oxygen, polyamide), with R2 values of 94%, 68%, 96%, 94%, and 95%, respectively. For Model 2 (air, brass), the highest performance was achieved using the CB method, with an R2 value of 93%. In this study, the optimal cooling performance results from the two parallel-connected counterflow RHVT tubes were achieved with Model 5. This model, operating at a pressure of 700 kPa and utilizing a brass material with six nozzles, produced a cooling performance of −242.55 K. The primary contribution of this research lies in the first-time application of the specified ML methods collectively for predicting the performance of the PCRHVT system, resulting in highly accurate prediction outcomes. [ABSTRACT FROM AUTHOR] |
| Copyright of Journal of Thermal Analysis & Calorimetry is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| FullText | Links: – Type: pdflink Text: Availability: 1 |
|---|---|
| Header | DbId: egs DbLabel: Engineering Source An: 186711517 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Performance analysis of a parallel-counterflow vortex tube using machine learning methods. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Korkmaz%2C+Murat%22">Korkmaz, Murat</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> mkorkmaz@hacettepe.edu.tr</i><br /><searchLink fieldCode="AR" term="%22Doğan%2C+Ayhan%22">Doğan, Ayhan</searchLink><relatesTo>2</relatesTo> (AUTHOR)<i> aydogan@hacettepe.edu.tr</i><br /><searchLink fieldCode="AR" term="%22Kırmacı%2C+Volkan%22">Kırmacı, Volkan</searchLink><relatesTo>3</relatesTo> (AUTHOR)<i> volkankirmaci@bartin.edu.tr</i> – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22Journal+of+Thermal+Analysis+%26+Calorimetry%22">Journal of Thermal Analysis & Calorimetry</searchLink>. May2025, Vol. 150 Issue 10, p7901-7919. 19p. – Name: Subject Label: Subjects Group: Su Data: <searchLink fieldCode="DE" term="%22Machine+learning%22">Machine learning</searchLink><br /><searchLink fieldCode="DE" term="%22Vortex+tubes%22">Vortex tubes</searchLink><br /><searchLink fieldCode="DE" term="%22Fluid+flow%22">Fluid flow</searchLink><br /><searchLink fieldCode="DE" term="%22Fluids%22">Fluids</searchLink><br /><searchLink fieldCode="DE" term="%22Sensitivity+analysis%22">Sensitivity analysis</searchLink><br /><searchLink fieldCode="DE" term="%22Cooling%22">Cooling</searchLink><br /><searchLink fieldCode="DE" term="%22Heat+transfer%22">Heat transfer</searchLink><br /><searchLink fieldCode="DE" term="%22Experimental+design%22">Experimental design</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: In this study, these machine learning methods (MLMs) were used for the first time in the literature to compare the temperature performance of two counterflow Ranque–Hilsch vortex tubes (RHVT) connected in parallel. As input parameters, two different pressurized fluids, three distinct materials, and six types of nozzles were selected, while the temperature difference was obtained from the hot and cold fluid outlets was designated as the output parameter. For the first time under these experimental conditions, a sensitivity analysis was conducted to determine the impact of the input parameters on the output. For each pressurized fluid and material, six models were developed using MLMs such as elastic net (EN), Bayesian ridge (BR), category boosting (CB), and light gradient boosting machine (LightGBM). A total of 30 different experimental setups were established by creating five separate setups for each model. The performance of these models was evaluated and compared using the K-fold cross-validation method. Upon analyzing the results, the BR method demonstrated the best performance for Model 1 (air, aluminum), Model 3 (air, polyamide), Model 4 (oxygen, aluminum), Model 5 (oxygen, brass), and Model 6 (oxygen, polyamide), with R2 values of 94%, 68%, 96%, 94%, and 95%, respectively. For Model 2 (air, brass), the highest performance was achieved using the CB method, with an R2 value of 93%. In this study, the optimal cooling performance results from the two parallel-connected counterflow RHVT tubes were achieved with Model 5. This model, operating at a pressure of 700 kPa and utilizing a brass material with six nozzles, produced a cooling performance of −242.55 K. The primary contribution of this research lies in the first-time application of the specified ML methods collectively for predicting the performance of the PCRHVT system, resulting in highly accurate prediction outcomes. [ABSTRACT FROM AUTHOR] – Name: AbstractSuppliedCopyright Label: Group: Ab Data: <i>Copyright of Journal of Thermal Analysis & Calorimetry is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.) |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=186711517 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1007/s10973-025-14245-1 Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 19 StartPage: 7901 Subjects: – SubjectFull: Machine learning Type: general – SubjectFull: Vortex tubes Type: general – SubjectFull: Fluid flow Type: general – SubjectFull: Fluids Type: general – SubjectFull: Sensitivity analysis Type: general – SubjectFull: Cooling Type: general – SubjectFull: Heat transfer Type: general – SubjectFull: Experimental design Type: general Titles: – TitleFull: Performance analysis of a parallel-counterflow vortex tube using machine learning methods. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Korkmaz, Murat – PersonEntity: Name: NameFull: Doğan, Ayhan – PersonEntity: Name: NameFull: Kırmacı, Volkan IsPartOfRelationships: – BibEntity: Dates: – D: 15 M: 05 Text: May2025 Type: published Y: 2025 Identifiers: – Type: issn-print Value: 13886150 Numbering: – Type: volume Value: 150 – Type: issue Value: 10 Titles: – TitleFull: Journal of Thermal Analysis & Calorimetry Type: main |
| ResultId | 1 |
