Development of an Adaptive Neuro Fuzzy Inference Model for Prediction of Heating Loads in Residential Buildings

Abstract

This paper introduces a new approach to predicting heating loads in homes using the adaptive neuro- fuzzy inference system (ANFIS). Predicting heating loads accurately is crucial for managing energy usage and ensuring that indoor temperatures are comfortable for occupants. Traditional models often struggle with the complexity and fluctuating nature of heating needs in different homes. To tackle this, this article proposes the an ANFIS framework, which blends the flexibility of fuzzy inference logic (FIS) with the learning abilities of artificial neural networks (ANN) for solving heating load prediction tasks. The data set used includes building parameters such as wall area, relative compactness, and surface area of the residential building. The performance results observed from the proposed model showed good predictive accuracy and generalization ability. This technique therefore stands as a viable approach for modeling and predicting heating loads in residential buildings.