校园建筑能耗统计分析与预测优化的智控技术研究

发布时间：2018-03-11 09:15

本文选题：高校校园建筑能耗　切入点：能耗数据处理　出处：《福建师范大学》2016年硕士论文　论文类型：学位论文

【摘要】：随着国家经济的迅猛发展,不断增加的国家大型公共建筑数量及其高耗能现象与日益严重的能源短缺问题逐渐被人们所关注。高校作为社会的重要组成成员之一,拥有的大型建筑数量之多、面积之广,使其能耗不断攀升,远超我国的国民人均能耗消费水平。建设节约型校园是近年来国内众多高校积极响应国家建设环境友好与资源节约社会的号召而开展的工作。目前,众多的高校都开始着手甚至已建设完成能耗管理系统,实现能源实施监管。本文的主要研究工作也将在此方面进行展开。本文第一章首先介绍节约型校园建设的背景与意义,分析总结国内外节约型校园建设的近况,从而引出本文的研究意义与内容,并说明本文的组织结构。然后第二章中,在借鉴浙江大学、北京理工大学等国内高校节能监管系统建设的经验基础上,提出一套适合于某大学的校园建筑能耗智控平台解决方案,并利用物联网技术、云计算技术等相关成熟技术进行平台构建。文中第三章阐述校园建筑能耗统计分析的指标,说明校园建筑总能耗的评价指标组成,重点介绍校园建筑电耗与水耗方面的统计指标计算方法,并根据实际应用提出校园建筑节能指标。第四章将研究如何对传感设备实时采集上传的校园能耗数据进行分析处理与转存,并利用统计方法进行建筑能耗统计、比较、排名等应用。为了让平台更加的智能化,本文第五章将充分利用平台的用能统计数据,讨论分析建筑能耗的影响因子,建立高校建筑能耗预测模型,利用智能算法预测其建筑能耗,并详细说明该模型。第六章主要完成上述章节的测试与应用。首先总结应用第三、四章内容实现用能统计。接着对第五章的模型进行预测的实验测试,对比基于遗传算法改进的BP神经网络与普通的BP神经网络的预测结果。由分析结果可知,基于遗传算法改进的BP神经网络预测效果优于未优化的BP神经网络。实验测试成功后,选取某一座校园建筑对模型进行实际应用,用于采集该座建筑的报警数据,实际应用结果表明,该模型预测数据可相对准确地应用于平台中。综上,本文提出的预测模型可以正确地辅助相关规则、措施的制订,从而为节约型校园的建设提供技术手段。
[Abstract]:With the rapid development of national economy, the increasing number of large national public buildings, the phenomenon of high energy consumption and the increasingly serious problem of energy shortage have been paid more and more attention. As one of the important members of the society, colleges and universities are becoming one of the most important members of the society. The number of large buildings and the size of their buildings have increased their energy consumption. The construction of energy-saving campus is the work of many colleges and universities in China in recent years in response to the call of the country to build an environmentally friendly and resource-conserving society. At present, Many colleges and universities have begun to build and even completed energy management systems to implement energy management. The main research work of this paper will also be carried out in this respect. Chapter one of this paper first introduces the background and significance of energy-saving campus construction. This paper analyzes and summarizes the recent situation of energy-saving campus construction at home and abroad, which leads to the significance and content of this paper, and explains the organizational structure of this paper. Then, in the second chapter, we draw lessons from Zhejiang University. Based on the experience of energy conservation supervision system construction in universities such as Beijing University of Technology, this paper puts forward a solution to the intelligent control platform of energy consumption in campus buildings suitable for a university, and utilizes the technology of the Internet of things. Cloud computing technology and other relevant mature technologies to build the platform. The third chapter describes the statistical analysis of campus building energy consumption indicators, explains the overall energy consumption of campus buildings evaluation index composition, This paper mainly introduces the calculation methods of the statistical indexes of power consumption and water consumption of campus buildings. According to the practical application, the paper puts forward the energy saving index of campus building. Chapter 4th will study how to analyze and store the campus energy consumption data collected and uploaded in real time by sensor equipment, and make use of the statistical method to calculate and compare the building energy consumption. In order to make the platform more intelligent, the 5th chapter of this paper will make full use of the platform's energy use statistics data, discuss and analyze the influence factors of building energy consumption, and establish the model of building energy consumption prediction in colleges and universities. The intelligent algorithm is used to predict the building energy consumption, and the model is described in detail. Chapter 6th mainly completes the testing and application of the above chapters. Four chapters implement energy use statistics. Then the model of chapter 5th is tested and compared with the results of the improved BP neural network based on genetic algorithm and the common BP neural network. The prediction effect of improved BP neural network based on genetic algorithm is better than that of unoptimized BP neural network. After the experiment is successful, a campus building is selected for practical application, which is used to collect the alarm data of the building. The practical application results show that the prediction data of the model can be applied to the platform relatively accurately. In summary, the prediction model presented in this paper can correctly assist the formulation of relevant rules and measures, thus providing a technical means for the construction of a conservation-oriented campus.
【学位授予单位】：福建师范大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TU111.195;TP18

【参考文献】