Artificial Neural network and experimental work of a solar cavity collector

Abstract

A collector is a device that can convert the available sun’s radiations into some useful heat energy with better entrapment of sunlight and is economical too. To heat the water and aqueous fluids, flat plate collector (FPC) is widely used. The main objective of this study is to find out the alternate for FPC with improved performance. Solar cavity collector (SCC) is one such type of an enhanced model of FPC. It consists of five cavities that have an arrangement for inlet and also outlet tubes to flow over. The same has been constructed with copper pipes as a receiver material to carry out the trails in order to find the suitable parametric study. The performance of any collector affects mainly the physical modifications of the collector. Therefore, the performance of SCC includes the prime parameters such as the out-turn of aperture entrance gap and cavity structure outer cover materials. Water rate of flow and comparison of SCC and FPC are the other parameters considered for the current study. Experimental results have to be compared with any simulation tool in order to achieve better performance. Using the Artificial Neural Network (ANN) tool of the MATLAB software, the simulations have been validated. The successful creation of the progressed model in ANN has the set of variables that are fetched out. The experimental and simulated end results match with each other in an appropriate manner.