Research On Multi-objective Optimization And Decision-making Of Existing Building Energy Efficiency Renovation

In recent years,due to the contradiction between limited energy reserves and increasing energy demand,energy-saving and emission reduction have become a hot issue that has attracted the common attention of industry and academia,and energy-saving renovation of existing buildings is an important part.In response to the national call for energy-saving and emission reduction,this study focuses on the ambitious goals of improving energy efficiency,reducing emissions,and optimizing resource allocation in building energy systems.This thesis regards the existing building energy-saving retrofit planning as a complex system optimization problem,studies the corresponding optimization models and methods,and formulates a framework to help energy-saving retrofit planning decision-making.Focusing on the optimization of building energy efficiency renovation,this thesis has conducted the following research.After in-depth analysis of the facilities damage and performance attenuation phenomena that must exist in the operation of building energy systems,the loss functions of several common types of equipment are extracted,and the dynamic change model of equipment population is established in combination with different maintenance strategies.Based on this dynamic model,an old office building in a southern city of China is used as a prototype to design an energy-saving renovation plan with multiple retrofitting options and retrofitting measures,and a multi-objective optimization model of energy-saving building renova-tion including energy consumption and investment is constructed.An optimal power supply scheduling scheme is designed for the new energy system involved in the multi-objective optimization model.Since the scheduling algorithm is timeconsuming to simulate,a neural network proxy model is proposed to reduce the computational load caused by the simulation.The agent model is coupled with nondominated sorting genetic algorithm and is used to solve the multi-objective optimization model proposed above.In order to improve the evaluation of the impact of energy-saving renovation on users and make it quan-tifiable,further research is conducted on the relationship between energy-saving renovation projects and users of retrofitted buildings,and four categories of satisfaction related to user experience are summarized.These four satisfaction models are modeled based on the previous research and the analytic hier-archy process.Two multi-objective evolutionary algorithms based on reference vectors are used to solve the multi-objective optimization problem with six objectives,and the optimization results of the two are compared through multi-objective evaluation indicators to obtain better results.In addition,according to the analysis of the optimization process the deficiencies of poorly performing algorithms is pointed out.In order to solve the difficulty of decision makers facing high-dimensional Pareto solution sets in decision-making,a multi-objective decision-making framework based on hierarchical clustering is designed,and the framework is used to analyze six-dimensional Pareto solution sets and make decisions.As a result,the proposed framework is feasible and robust.