基于贝叶斯理论的区域建筑冷热负荷预测模型的应用研究

发布时间：2018-01-10 15:38

本文关键词：基于贝叶斯理论的区域建筑冷热负荷预测模型的应用研究　出处：《湖南大学》2015年硕士论文　论文类型：学位论文

【摘要】：近年来,区域供能技术的研究和应用逐步兴起。其中,对区域建筑的冷热负荷准确预测是区域能源技术开发及推广的重要基础。本课题组在分析现有的建筑负荷预测方法的基础上,提出一种基于贝叶斯理论的区域建筑冷热负荷预测模型,是专门针对在区域能源规划阶段缺乏详细建筑信息的一种区域建筑冷热负荷预测方法。由于该方法提出后缺少工程实例验证和分析,所以本文将通过对实际案例的分析对此方法的准确性和适用性进行验证与分析。本文首先对基于贝叶斯理论的区域建筑冷热负荷预测模型进行了系统介绍,为了验证该方法在不同气候区的适用性、在不同功能类型的区域建筑的适用性以及在冷负荷和热负荷的预测上是否都一样适用。选择3个案例:案例1——寒冷地区某城市的某住宅小区,案例2——夏热冬冷地区某城市的某学校,案例3——严寒地区某城市的大型综合生活区作为验证案例,采用Design Builder能耗模拟软件作为工程模拟软件以及Matlab作为数值计算软件。通过对三个案例基本信息的调研与统计,分别建立各自区域的标准单体建筑,并进行动态负荷的模拟获取逐时/日动态负荷。其次,将得到的各个案例的标准建筑的动态负荷分别作为先验信息,将调查和统计得到的各案例对应地区的其他区域对应逐时/日负荷值作为样本信息,分别将每个案例的先验信息和样本信息代入贝叶斯回归模型,得出各自的后验信息,以此作为各案例的负荷预测因子。将最终得到的修正结果与对应的简单面积叠加法得到的预测结果以及实测负荷值进行对比,并根据建筑负荷预测方法评价指标(逐时/日相对误差、均方根相对误差以及最大误差比)分析各案例的预测精度。通过对三个案例的误差结果的分析和比较,可以证明所提出的预测方法较简单面积叠加法的准确性更高。再由横向比较三个案例的相对精准度,分析得出该方法的适用性:1)对区域建筑的夏季冷负荷和冬季热负荷都是适用的;2)适用于我国各类建筑热工区域的区域建筑负荷预测;3)大多数类型和规模下的区域建筑的负荷预测,都是适用的;4)在样本信息质量越好数量越多的情况下,预测效果越好。
[Abstract]:In recent years, the research and application of regional energy technology gradually rise. Among them, the cold and heat load on the regional building accurate prediction is an important basis for regional energy technology development and promotion. This paper based on the analysis of the existing building load forecasting methods, put forward a forecasting model of regional building cooling load based on the theory of Juliu bay that is specific to the lack of detailed information on the construction of regional energy planning stage of a region and cold load forecasting method. Due to the lack of engineering examples and analysis method, so the accuracy of the analysis of the actual case by this method and applicability are verified and analyzed. This paper firstly introduces the the building area of the cold and heat load prediction model based on Bayesian theory, in order to verify the applicability of this method in different climatic zones, in the different type of functional area The applicability domain architecture as well as in the prediction of cooling load and thermal load are the same if applicable. 3 cases: a case of 1 - City in cold area of a residential area, 2 case - a city in hot summer and cold winter area of a school, a large 3 - a case of cold region city comprehensive living area as to verify the case, using Design Builder simulation software as engineering simulation software and Matlab as the numerical calculation software. Through the investigation and statistics of the three cases of basic information, establish their respective regional standard single building, simulation and dynamic load for hourly dynamic load / day. Secondly, the dynamic load of each standard building the cases are used as prior information, corresponding to other regions corresponding to the case area investigation and statistics of the hourly / daily load value as the sample information, respectively, each The case of the prior information and the sample information into the Bayesian regression model, we get their posterior information, as the case of the load forecasting factor. The final results will be corrected with the corresponding area of simple superposition of the predicted result and the measured value of the load ratio, and the evaluation index according to the hourly building load (on the relative error of mean square error and the maximum error ratio) analysis and prediction accuracy for each case. By analyzing and comparing the error results of three cases, the accuracy of prediction can prove that this method is simple area superposition method is higher. By the comparison of the three cases of the relative accuracy. Analysis of applicability of the method is derived: 1) on the regional building cooling load in summer and winter heat load are available; 2) regional construction suitable for China's various types of pre load thermal building area (3) most of the load prediction of regional buildings under most types and scales is applicable; 4) the better the quality of samples is, the better the prediction effect is.

【学位授予单位】：湖南大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TU831.2

【参考文献】