基于多源数据的大型公共建筑中用能行为模式、机理与节能政策研究

Energy conservation and emission reduction is an important commitment of China to cope with global climate change, and it is also a major demand for strengthening ecological civilization construction and sustainable development in China. The building sector is the main source of global energy consumption and carbon emissions, especially improving the energy efficiency of large public buildings is crucial for China's energy conservation and emission reduction. This project explores the design and optimization of large-scale public building energy-saving policies from the perspective of residents' energy use behavior, and innovatively utilizes multi-source big data in buildings (including building energy consumption, resident behavior, environmental parameters, etc.) and uses data mining methods to identify The model characteristics and driving factors of residents' energy use behaviors, deeply understand the mechanism of behavior generation; and then combines the cognitive behavior theory to propose the theoretical model and logical framework of energy use behavior; finally, uses this algorithm framework to build multi-agent based model for energy behavior simulation. The intelligent model is used to simulate and analyze the energy use behavior of residents in large public buildings and to verify the evidence with the empirical data. Based on the differentiated energy-use behavior, the incentive mechanism of energy-saving behavior and the optimization path of building energy consumption are designed. The results of this project can enrich the research method of energy-use behavior, expand the research perspective of refined energy consumption management and improve the incentive mechanism of behavioral energy conservation. The achievement of this study can improve of building energy efficiency, and provide theoretical basis for sustainable development.

节能减排是我国应对全球气候变化的重要承诺，也是我国加强生态文明建设与可持续发展的重大需求。建筑业是全球能耗与碳排放的主要来源，尤其是大型公共建筑的能效提升对我国节能减排至关重要。本项目从居民用能行为视角出发探索大型公共建筑节能政策的设计与优化方案，创新性地利用建筑中多源大数据（包括建筑能耗、居民行为、环境参数等），采用数据挖掘方法识别居民用能行为的模式特征和驱动因素，深入理解行为产生的机理；进而结合认知行为理论构建用能行为的理论模型与逻辑框架；最后以此为算法框架构建基于多智能体的用能行为智能仿真模型，开展大型公共建筑中居民用能行为的仿真分析并与实证数据进行对比验证，基于差异化的用能行为模式设计节能行为的激励机制与建筑能耗的优化路径。本项目成果可以丰富居民用能行为的研究方法体系，拓展能耗精细化管理的研究视角，改善行为节能的激励机制，促进节能减排，为生态文明建设与可持续发展提供理论基础。

节能减排是我国应对全球气候变化的重要承诺，也是我国加强生态文明建设与可持续发展的重大需求。建筑业是全球能耗与碳排放的主要来源，尤其是大型公共建筑的能效提升对我国节能减排至关重要。本项目从居民用能行为视角出发探索大型公共建筑节能政策的设计与优化方案，主要内容包括：1）创新性地利用建筑中多源大数据（包括建筑能耗、居民行为、环境参数等），采用数据挖掘方法识别居民用能行为的模式特征和驱动因素，深入理解行为产生的机理；2）结合认知行为理论构建用能行为的理论模型与逻辑框架；3）以此为算法框架构建基于多智能体的用能行为智能仿真模型，开展大型公共建筑中居民用能行为的仿真分析并与实证数据进行对比验证，基于差异化的用能行为模式设计节能行为的激励机制与建筑能耗的优化路径。本项目主要成果包括：1）对居民用能行为的影响因素、居民用能行为与建筑能耗的关系、以及多智能体建模在决策行为中的应用等多个角度对文献进行了系统的梳理综述并进行发表，是本研究领域重要的研究基础；2）识别用能行为的影响因素，通过大量的调研对居民用能行为影响因素进行识别和分析，对影响因素进行了系统的整理；3）构建用能行为模型，本研究基于HOTCO理论，构建了有限理性的行为模型，同时包含理性影响机制和情感影响机制，是对用能行为的一个较大创新；4）实现了居民用能行为与建筑能耗模拟仿真系统，基于多智能体建模的方法对居民的行为进行建模仿真，并完成了仿真平台的构建；5）完成了政策仿真实验，在模拟仿真系统基础上，对不同的政策情景进行仿真实验，获得了不同政策情景影响下居民行为及能耗变化。本项目成果可以丰富居民用能行为的研究方法体系，拓展能耗精细化管理的研究视角，改善行为节能的激励机制，促进节能减排，为生态文明建设与可持续发展提供理论基础。