北京市大型办公建筑能耗比对与评价体系研究

发布时间：2018-02-24 17:16

本文关键词： 大型办公建筑能耗基准多重共线性回归方法评价体系　出处：《北京工业大学》2015年硕士论文　论文类型：学位论文

【摘要】：首都北京作为全国的政治中心、文化中心、国际交往中心及科技创新中心,拥有众多高密度、大容量的办公类建筑,此类建筑多属高能耗、低能效建筑,是北京市公共建筑节能工作最重要的部分之一。从国家宏观层面,为降低公共建筑运行能耗,有效提高其能源利用效率,建立公共建筑节能长效机制,住建部搭建了国家机关办公建筑和大型公共建筑节能监管体系。北京市虽建立了公共建筑能耗监测平台,但目前只能实现能耗的监测,无法实现能效的管理与分析功能。因此,在北京建立大型办公建筑能耗比对与评价体系具有重要的实际应用价值。多重共线性对能耗基准模型估计的准确性与稳定性有很大影响,本课题以北京市大型办公建筑能耗调研数据为基础,通过能耗及建筑运行数据分析与统计方法研究,提出了解决多重共线性问题的能耗基准模型开发的新方法,并建立了北京市大型办公建筑能耗比对与评价体系。主要工作包括:1)分析了北京市大型办公建筑能耗现状。通过对筛选出的88栋样本建筑进行能耗数据统计发现:不同建筑能耗密度(Energy Use Intensity,简称EUI)差异性很大,其分布服从正态分布,可代表北京市大型办公建筑能耗水平。2)对能耗影响因素进行了相关性分析。通过相关性分析,筛选出10个显著因素:建筑面积、用能人数、运行时间、建筑高度、楼层数、商业区面积、餐饮区面积、数据机房面积、地下车库面积以及建筑年代。3)建立了能耗基准模型。对影响因素进行多重共线性检验,研究发现各因素存在严重多重共线性,根据数据特点,通过对数学统计方法进行研究分析,选取适合能耗基准模型建立的统计方法。分别以逐步回归、主元回归、偏最小二乘回归三种方法建立了能耗基准模型,通过统计学检验,三个模型均显著,偏最小二乘回归模型效果最好。4)开发了能耗比对体系。以偏最小二乘回归模型为最终能耗基准模型,能效比率(实际EUI/预测基准EUI)为评价指标,根据能效比率的分布情况拟合出Gamma分布曲线,以此开发了百分制能效评分表,最终确定了偏最小二乘回归能耗基准模型与评分表构成的北京市大型办公建筑能耗比对与评价体系。
[Abstract]:As the national political center, cultural center, international communication center and science and technology innovation center, the capital Beijing has many high-density and high-capacity office buildings, most of which are energy-efficient and energy-efficient. It is one of the most important parts of the energy-saving work of public buildings in Beijing. In order to reduce the energy consumption of public buildings, improve the efficiency of energy utilization and establish a long-term energy-saving mechanism for public buildings at the national macro level, The Ministry of Housing and Construction has set up a state office building and a large public building energy conservation supervision system. Although Beijing has established a monitoring platform for energy consumption of public buildings, it can only monitor energy consumption at present, but it cannot achieve the function of energy efficiency management and analysis. It has important practical application value to establish the energy consumption comparison and evaluation system of large office buildings in Beijing. The accuracy and stability of energy consumption benchmark model estimation are greatly affected by multiple collinearity. Based on the investigation data of energy consumption of large office buildings in Beijing, this paper puts forward a new method to develop the energy consumption benchmark model based on the analysis of energy consumption and building operation data and statistical method. The energy consumption comparison and evaluation system of large office buildings in Beijing is established. The main work includes: 1) analyzing the present situation of energy consumption of Beijing large office buildings. Energy Use Intensity (EUI) is very different among different building energy consumption densities. From the normal distribution, which can represent the energy consumption level of large office buildings in Beijing, the influence factors of energy consumption are analyzed. Through the correlation analysis, 10 significant factors are selected: the building area, the number of people who use energy, the running time. The building height, the number of floors, the area of commercial district, the area of dining area, the area of data machine room, the area of underground garage and the building age. It is found that there are serious multiple collinearity among the factors. According to the characteristics of the data, the statistical methods suitable for the establishment of the energy consumption benchmark model are selected through the study and analysis of the mathematical statistical methods. The stepwise regression and principal component regression are used, respectively. Three methods of partial least squares regression are used to establish the benchmark model of energy consumption. The statistical tests show that the three models are significant. The energy consumption comparison system was developed by using partial least squares regression model as the final energy consumption benchmark model, and the energy efficiency ratio (EER) ratio (actual EUI- / predictive benchmark EUI-) was used as the evaluation index, and the energy consumption comparison system was developed by using the partial least squares regression model as the final energy consumption benchmark model. According to the distribution of energy efficiency ratio (EER), the distribution curve of Gamma was fitted, and the energy efficiency score table of percent system was developed. Finally, the energy consumption comparison and evaluation system of large office buildings in Beijing is established, which is composed of partial least square regression energy consumption benchmark model and scoring table.
【学位授予单位】：北京工业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TU243;TU111.195

【参考文献】