Development of a model for urban heat island prediction using neural network techniques

The urban heat island (UHI) phenomenon is mainly caused by the differences in the thermal behaviour between urban and rural settlements that are associated with the thermal properties of urban materials, urban geometry, air pollution, and the anthropogenic heat released by the urban activities. The UHI has a serious impact on the energy consumption of buildings, increases smog production, while contributing to an increasing emission of pollutants from power plants, including sulfur dioxide, carbon monoxide, nitrous oxides and suspended particulates.This study presents the applicability of artificial neural networks (ANNs) and learning paradigms for UHI intensity prediction in Athens, Greece. The proposed model is tested using Elman, Feed-Forward and Cascade neural network architecture. The data of time, ambient temperature and global solar radiation are used to train and test the different models. The prediction accuracy is analyzed and evaluated.