Abstract:Building Services Engineering Research and Technology, Ahead of Print. Building design optimization (BDO) provides an approach for decreasing global energy consumption and achieving the goal of carbon neutrality. However, energy efficiency and comfort performance are two conflicting objectives when making optimal building design schemes. This study proposes a surrogate model-based multiple-objective optimization framework to balance the conflicting objectives and obtain an optimal design scheme for buildings. Firstly, an energy simulation model for generating energy consumption and design parameters is constructed, and the obtained data are utilized to train the surrogate model with the random forest (RF) algorithm. Then, multi-objective optimization algorithms are employed to generate a set of alternative plans for building schemes and determine the optimal building design solutions that can equilibrate the requirements for both energy conservation and building comfort. To verify the proposed optimization method in this paper, a residential building in Suzhou was selected as a case study. The study considered 10 building design parameters that are related to energy efficiency and thermal comfort. The results indicate that the RF surrogate model accurately predicts energy consumption, with a predicted MSE of 0.00012 and R2 of 0.99. In evaluating the Pareto set size, Pareto solution diversity, Pareto front proximity, and best solution quality, NSGA-II proved to be the most effective optimization algorithm for BDO problems. The final optimal solution of design parameters obtained by NSGA-II obviously improves the building performance of comfort and energy efficiency, and the results of the performance evaluation for different optimization algorithms provide guidance to make decisions on suitable algorithms and hyperparameter settings based on the greatest preference of the performance criteria. This study will help determine the best design options for buildings to achieve better energy efficiency in sustainable development and provide reference for similar projects.Practical applicationsThis research makes valuable contributions in the following aspects:(a) It establishes a multi-objective optimization design model for a virtual building environment. This enables the visualization and digitization of the building model and further facilitates the transformation of the optimization model, thereby providing users with convenient decision-making tools; (b) The study provides designers and other stakeholders with comprehensive simulation-based analysis results and optimization techniques. These tools aid in making energy-saving multi-objective optimization decisions;(c) The research compares various optimization algorithms and presents their strengths and limitations, which will help designers select suitable algorithms based on practical requirements.

Performance Indices and Evaluation of Algorithms in Building Energy Efficient Design Optimization

Multi-Objective Optimization Technology for Building Energy-Saving Renovation Strategy Based on Genetic Algorithm

Multiobjective Optimization Method for Energy‐Saving Design of Green Buildings

Multi-objective Optimization of Building Energy Performance Using a Particle Swarm Optimizer with Less Control Parameters

Quantitative Analysis of Performance-Oriented Design Efficiency in Early Divergent Parametric Design of Office Building Façade

Based on ANN and many-objective optimization to improve the performance and economy of village houses in Chinese cold regions

Decomposition based multi-objective evolutionary algorithm for energy-saving design of homestay buildings

Research on Green Building Design Optimization Based on Building Information Modeling and Improved Genetic Algorithm

Research on multi-objective optimization of building energy efficiency based on energy consumption and thermal comfort

Optimal building energy management using intelligent optimization

An Intelligent Optimization for Building Design Based on BP Neural Network and SPEA-II Multiobjective Algorithm

Design of low-emission and energy-efficient residential buildings using a multi-objective optimization algorithm

Intelligent Optimization: A Novel Framework to Automatize Multi-Objective Optimization of Building Daylighting and Energy Performances

Application of Green Building Materials and Multi-objective Energy-Saving Optimization Design

Performance in generation: An automatic generalizable generative-design-based performance optimization framework for sustainable building design

Building information modelling-enabled multi-objective optimization for energy consumption parametric analysis in green buildings design using hybrid machine learning algorithms

Many-Objective Optimization Design of a Public Building for Energy, Daylighting and Cost Performance Improvement

An integrated framework for multi-objective optimization of building performance: Carbon emissions, thermal comfort, and global cost

Generative Design and Performance Optimization of Residential Buildings Based on Parametric Algorithm

Design Optimization of Renewable Energy Systems in Low / Zero Energy Buildings Using Single and Multi-Objective Optimization Methods

Optimization Approaches in Performance-Based Architectural Design - A Comparison Study