Solar air heaters performance prediction using multi-layer perceptron neural network– A systematic review

Abstract

ABSTRACT In solar thermal systems, solar air heater (SAH) is an important device for heating air using energy of the sun. The solar collector is the most important part of SAH which collects solar radiations as thermal energy (heat) and transmits it to the air. For efficient utilization of solar energy, the system performance needs to be optimized. There are many techniques that are used for optimization but Artificial Intelligence (AI) technique is very effective. Artificial neural networks (ANN) are coming up as valuable options in place of traditional statistical modeling techniques in many scientific disciplines. ANN technique is the most used one for modeling, prediction, and control. This technique is computes faster results and it can solve the complex and nonlinear problems which cannot be solved by other conventional approaches. Also this technique requires large data sets but not separate programming for solving the problems. In this paper, all the published research works from year 2009 to 2021 related to MLP model for performance prediction has been reviewed, which are applied on solar air heating systems. The main aim of current study is to review research work related to multi-layer perceptron (MLP) model applied in SAH and to find a gap for future research. This paper encompasses many research articles related to neural networks which focus particularly on MLP model. Review of some research papers has been included in detail for the proposed work. Also the present review article reported that MLP structure of neural model is the most commonly applicable model used for prediction of SAHs performances with satisfactory results.