AI-DRIVEN PASSIVE COOLING ASSESSMENT: PREDICTIVE MODELING OF MULTI-LAYERED BUILDING ENVELOPES

Abstract

Rising energy consumption for building climate control drives the need for alternative cooling approaches that minimize power usage. This investigation examines how wall systems with multiple layers separated by air spaces perform thermally as a strategy for natural temperature regulation. An experimental structure measuring 4 feet by 6 feet was built using locally sourced laterite stone, with an additional outer wall positioned 1.5 feet away from the primary structure on each side. Eight temperature sensors were positioned systematically on interior and exterior wall surfaces to track thermal variations across all orientations. The research sought to measure the insulating capacity of the air space configuration and create forecasting tools for design enhancement. The working assumption suggested that dual-wall construction with intermediate air spacing would substantially limit thermal transfer relative to standard single-layer walls. Advanced computational analysis methods were applied to interpret the intricate thermal dynamics and forecast performance characteristics. Findings revealed considerable temperature differences between external and internal wall surfaces, confirming the air cavity's role as an effective thermal barrier. The application of intelligent data processing facilitated the determination of peak performance scenarios and enabled temperature forecasting based on ambient conditions. This work advances environmentally responsible construction practices by delivering measurable proof of how layered wall assemblies achieve natural cooling, validated through sophisticated analytical methods. The outcomes provide actionable guidance for deploying low-energy building exterior designs, especially applicable to regions experiencing elevated temperatures.