Determination of Effective Parameters on Flat Plate Collector Performance Using Machine Learning Method | ||
| مدل سازی در مهندسی | ||
| Articles in Press, Accepted Manuscript, Available Online from 22 December 2025 | ||
| Document Type: Research Paper | ||
| DOI: 10.22075/jme.2025.38847.2897 | ||
| Authors | ||
| Saadat zirak* 1; Mahtab Salimi2 | ||
| 1Faculty of Mechanical Engineering, Semnan University, Semnan, Iran | ||
| 2Mechanical Engineering Department/semnan University | ||
| Receive Date: 06 September 2025, Revise Date: 23 November 2025, Accept Date: 22 December 2025 | ||
| Abstract | ||
| In this paper, first, analytical relationships of flat plate collector absorption rate and solar-to-thermal energy efficiency are presented. For the considered collector, without the use of cooling water, the collector stagnation temperature is 132.5 degrees Celsius (absorber plate temperature) and by entering the cooling water of one liter per minute, the absorber plate temperature decreases to 33 degrees Celsius and the collector efficiency reaches to 77%. To predict the collector efficiency, three machine learning models were used: linear, random forest, and decision tree. Seven parameters of solar radiation intensity, collector tilt angle, wind speed, pipe diameter, number of pipes, ambient temperature, and cooling water flow rate, were selected as input parameters. Comparison of the predicted efficiency with actual values showed that the linear model has a weaker evaluation than the other two models. The random forest and decision tree models perform prediction with almost equal ability and high accuracy (the random forest model predicts negligibly better than the decision tree model). In addition, among the input parameters, changes in collector tilt angle, solar radiation and wind speed insignificantly affects the efficiency. The cooling water flow rate has the greatest effect. The pipe diameter, ambient temperature and the number of the tubes, have a moderate effect. | ||
| Keywords | ||
| Flat plate collector; Performance; Machine learning; Efficiency; Random Forest; Decision tree | ||
|
Statistics Article View: 66 |
||