Artificial Neural Network Modeling for Predicting and Evaluating the Mean Radiant Temperature around Buildings on Hot Summer Days

In recent years, the phenomenon of urban warming has become increasingly serious, and with number residents increasing, risk heatstroke in extreme weather higher than ever. order to mitigate adapt it, many researchers have been paying increasing attention outdoor thermal comfort. The mean radiant temperature (MRT) is one most important variables affecting human comfort spaces. purpose this paper predict distribution MRT around buildings based on a commonly used multilayer neural network (MLNN) that optimized by genetic algorithms (GA) backpropagation (BP) algorithms. Weather data from 2014 2018 together related indexes grid were selected as input parameters for training, 2019 was predicted. This study obtained very high prediction accuracy, which can be combined sensitivity analysis methods analyze hot summer days (the highest air over 30 °C). significant implications optimization strategies future building designers improve conditions buildings.