Estimating cooling loads of Indian residences using building geometry data and multiple linear regression

Abstract

International Energy Agency (IEA) predicts India's AC stock will reach 1144 million units by 2050, making it the second largest ACs holder globally. Studies on the effect of building geometry on cooling load reduction are primarily focused on material and envelope specifications. However, studies on building morphological parameters in the Indian context are scarce. Therefore, this research quantifies the effect of four morphology predictors, namely, FL (floor number), ESA (exposed surface area), CZB (conditioned zones per building), and CZF (conditioned zones per floor) on cooling load in 75 dominant residential built forms of Navi Mumbai. The selected buildings are simulated using the Rhinoceros 6 tool with the energy plus plugin. Despite having the same simulation inputs, envelope parameters, and conditioned volume, the results indicated a 90 % variation between the compact and loosely designed forms. Multiple Linear Regression shows that the four predictors explain 78% (R2 = 0.78) of variation in the cooling load. It is observed that tall buildings show greater efficiency in cooling load reduction due to lesser CZF values. Also, an increase in CZB and a decrease in ESA significantly reduce the mean cooling load due to compactness and wall sharing, respectively.

Keywords

• Building Geometry
• Navi Mumbai
• Cooling loads
• Simulation
• Multiple Linear Regression