Name: Puteri Shareena Ayu Binti Mohd Yusah
Metric No: BK10110243
Topic: PRDICTION OF INDOOR AIR TEMPERATURE IN NATURALLY VENTILATEDÂ BUILDINGS
Abstract: The indoor temperatures in naturally ventilated buildings vary according to the outdoor climatic conditions, building designs and building materials. Thereby, the prediction of the indoor temperatures is important for the assessment of occupants’ thermal satisfaction toward their indoor environments. Givoni presented formulas for predicting daily maximum average and minimum indoor temperatures. Givoni procedure in its simplest form is a multiple regression models. This method was chosen for the present study due to its simplicity. This is because it does not require building thermal design software to be generated and most importantly because it provides good results close to the predicted values by software. However, in this study only least linear regression was applied for understanding the thermal behaviour of the selected houses in Kota Kinabalu, Sabah. This is crucial step in the Givoni suggested procedure prior development of more robust models for indoor temperature predictions; three houses were selected based on their building types and materials to carry out this study. it was found that is possible to predict the indoor temperature of the houses by this methodology. Several regression models were developed for the predictions of daily maximum, minimum and average indoor temperatures in this study. The validity of the predicted models was addressed and comparisons between models were made according to the accuracy of the models. Further improvement might be possible by introducing other variables such as solar radiation and other parameters.