Non-dominated sorting genetic algorithm-II: A multi-objective optimization method for building renovations with half-life cycle and economic costs

Abstract

In this paper, we present a comprehensive optimization framework that identifies renovation plans to minimize half-life cycle carbon emissions, investment payback period, and indoor discomfort hours. The framework consists of four stages. First, relevant data were collected, building models were established, and the renovation scope and preliminary parameters were determined. Second, a sensitivity analysis of the initial parameter set was conducted, and important parameters were selected and input into a back-propagation neural network model for prediction. Finally, an optimal renovation plan was obtained through multi-objective optimization and the technique for order of preference by similarity to the ideal solution (TOPSIS) decision-making. To illustrate the framework's feasibility, it was applied to a building as an example. Remarkably, carbon emissions were reduced by 82.2 %, and zero carbon was achieved during the half-life cycle. Moreover, this achievement resulted in a relatively swift payback period of 3.9 years, coupled with a commendable 30 % decrease in indoor discomfort hours. Hence, the framework is effective in optimizing building renovation objectives, yielding a more harmonious and ideal building renovation strategy, and can be widely utilized to enhance building performance.