Open AccessProportional Learning of Semi Indirect Evaporative Cooling System
Author(s) -
Meetu Bharti,
Abhishek Singh
Publication year - 2019
Publication title -
international journal of recent technology and engineering
Language(s) - English
Resource type - Journals
ISSN - 2277-3878
DOI - 10.35940/ijrte.d8851.118419
Subject(s) - evaporative cooler , parametric statistics , artificial neural network , channel (broadcasting) , computer science , approximation error , air temperature , machine learning , artificial intelligence , environmental science , simulation , meteorology , statistics , mathematics , engineering , algorithm , mechanical engineering , computer network , physics
In this study, the semi evaporative cooling framework is assessed which is rely upon the presentation of a massexchanger (RHMX) and regenerative warmth. Prior specialists find the general numerical modeling techniques to overcome the cumbersome computational burden, and built up the information driven artificial neural network (ANN) model. A 1-D numerical model was used to deliver test information for structure the ANN model. Both the ANN models and numerical and the ANN models were affirmed against test results open in the composition with a satisfactory ordinary error level of around 4% reliant on the air temperature change over the dry channel. The assessment between examination information and ANN prediction exhibited incredible prediction accuracy. The typical prediction error between the envisioned and attempted information was around 4% reliant on the air temperature change over the dry channel. With the information driven model, parametric assessments were made to examine the introduction of the RHMX under different working conditions. Finally, a structure advancement of semi meandering evaporative cooling plan of extraction air proportion was coordinated under different surrounding conditions. It was found that the perfect extraction air proportion reduced with the encompassing temperature and furthermore relative wetness which stretched out from 0.3 to 0.36.