Prediction of Thermal Aspects for Brass Material-Based Natural Convection Heat Transfer System by Using Adaptive Neuro-fuzzy Inference System

In this paper, a model based on adaptive neuro-fuzzy inference system (ANFIS) is developed to predict the behavior of a natural thermal convection system. ANFIS model is able to successfully imitate the effects of variation in input parameters such as current and voltage on the response parameters such as temperature at different locations of the thermal system. The results obtained with the help of the developed ANFIS model are compared with the findings of the experiments in the form of graphs and also numerically by determining the error norms. Comparison of the ANFIS model-based predictions with the actual experimental results shows that the proposed model is able to identify the behavioral characteristics of the natural convection thermal system very accurately. The outcome of the proposed model is found out to be the best for the prediction of temperature at the location of fourth temperature sensor where the level of temperature is low. The proposed ANFIS model can be used further to develop a control system in order to control the temperature at different locations of the thermal system by varying the current and voltage parameters.