Multi-objective optimization for green retrofitting of existing school buildings considering life cycle carbon emissions, energy use intensity and indoor environment

The green retrofitting of existing buildings is critical for achieving energy savings and carbon emission reduction within the construction sector. Numerous school buildings have significant potential for environmental performance optimization, energy savings and carbon emission reduction. The current research disregards students' environmental needs, complex budget constraints and life cycle carbon (LCC) emissions. This study uses a typical school building in Hangzhou as a case. We establish a multi-objective optimization model for existing school buildings, which comprehensively considers LCC emissions, energy use intensity (EUI), initial investment cost (IC), and thermal and lighting environment. By integrating an artificial neural network (ANN) model for rapid building performance prediction and employing Non-dominated Sorting Genetic Algorithm II (NSGA-II), the model optimizes the building envelope, lighting, air-conditioning, and photovoltaic (PV) system. The results show that, without a PV system, the optimal solutions, which require an average investment of 280 CNY/m2 and an average payback period of 22.3 years, can achieve an average energy saving rate of 32 % and an average carbon emission reduction rate of 21.6 %. With a PV system, the average payback periods range from 13.4 to 21.2 years across different IC levels. An investment of 630 CNY/m2 can achieve net-zero energy consumption; for retrofit solutions exceeding 995 CNY/m2, however, LCC emissions begin to increase. This study reveals a nonlinear relationship between IC and LCC emissions and proposes differentiated retrofit strategies based on different PV installation scenarios and IC gradients. These findings provide technical pathways and decision-making support for the green retrofitting of school buildings.