Prediction of the performance of a shower cooling tower based on projection pursuit regression

• Published in: Applied thermal engineering Vol. 28; no. 8; pp. 1031 - 1038
• Main Authors: Qi, Xiaoni, Liu, Zhenyan, Li, Dandan
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.06.2008; Elsevier
• Subjects: Applied sciences; Devices using thermal energy; Energy; Energy. Thermal use of fuels; Exact sciences and technology; Heat and mass transfer; Heat exchangers (included heat transformers, condensers, cooling towers); Heat transfer; Projection pursuit regression; Shower cooling tower; Theoretical studies. Data and constants. Metering; Shower cooling tower; Heat and mass transfer; Projection pursuit regression; Costs; Cooling; Error; Numerical method; Replacement; Experimental study; Modeling; Mass transfer; Sampling; Performance; Comparative study; Cooling tower
• ISSN: 1359-4311
• DOI: 10.1016/j.applthermaleng.2007.06.029

This study was prompted by the need to design towers with no tower packing which causes salt deposition on the packing and subsequent blockage. The cooling tower analyzed in this study, named shower cooling tower (SCT), is such kind of towers. In the present study of SCTs, however, no systematic numerical method is available, as the focus has been mostly put on experimental investigation on SCTs. It is desirable to develop new methods that can provide quick, easy, and cost-effective estimates of the performance of SCTs, for traditional approaches to yield information are often time-consuming, expensive, and limited in individual samplings. In this paper we developed a yield model based on the projection pursuit regression (PPR) method by analyzing the heat and mass transfer process of SCT in a one-dimensional model). The model can help the designer evaluate the outlet water temperature from a given set of experimentally obtained data. For comparison purposes and accurate evaluation of the predictions, part of the experimental data was used as training samples and the remainder to test the model. The results predicted by the PPR model were compared with those of heat and mass transfer (HMT) model and the experimental data. The PPR model predicted the cooling range (i.e. the difference between the inlet and the outlet water temperatures) with a MAE (mean absolute error) of 3.75%, whereas, the MHT model showed a MAE of 8.78%. However, a PPR model is not a replacement for the convention heat and mass transfer model as the former does not deal with the transport mechanisms and its applicability may be limited by the range and quality of input data. It is a new applicable approach to studying the performance of a SCT. It is our hope that this paper can draw more attention to the study on the PPR model and to its application in SCT study.