Multi-objective building envelope optimization for thermal comfort and energy efficiency of educational buildings in India

The global energy crisis and climate change have intensified concerns about sustainability, particularly since buildings are among the largest energy consumers. The indoor environmental quality of buildings significantly impacts occupants’ health, well-being, and productivity. In India, cooling loads are a major contributor to increased energy usage and CO2 emissions, driven by hot and humid climates that lead to uncomfortable indoor conditions. Early design decisions about the building envelope, such as building geometry, window-to-wall ratio, and material properties, are critical for optimizing energy efficiency and thermal comfort.

However, there have been limited efforts to develop simulation-based approaches to understand and optimize the impact of passive design solutions on energy needs and thermal comfort in buildings located in India’s tropical climates. This study examines a university building in a hot and humid region of India. The case study building was modelled in Openstudio and calibrated using collected energy and environmental data. Ten parametric design variables, including wall and roof construction assembly, insulation types, and orientation, were identified.

The objective was to optimize the building envelope to maximize thermal comfort and minimize energy consumption. JEPlus, a parametric tool designed for energy simulations using software like EnergyPlus/TRNSYS, was employed. Parametric simulations of different envelope parameter combinations were conducted, and the results were used to train a surrogate model with an artificial neural network (ANN) to predict thermal comfort and energy consumption.

The large dataset generated from parametric simulations was essential for training the ANN-based surrogate model, which provides a fast yet high-resolution way to predict the building’s performance. Multi-Objective Evolutionary Algorithms (MOEA) were applied to the outputs of the ANN-based surrogate model to find Pareto-optimal solutions, representing the best trade-offs between thermal comfort and energy efficiency. The optimized results of the ANN-MOEA model showed a reduction of 20% in building energy consumption and about 7% in discomfort hours compared to the base case scenario.

Key passive design solutions included improved insulation, optimized window-to-wall ratios, and suitable building orientation. The study’s findings offer practical and useful guidelines for building designers to select the optimal combination of building design parameters during the early stages of design and for building retrofits. These guidelines aim to improve occupants’ thermal comfort and reduce energy consumption in educational buildings within the Indian context.