基于数据仓库技术的能效分析软件开发
|
摘要
随着全球能源日益紧张以及能源价格的不断上涨,人们对于节能降耗的实施力度给予了高度的重视。作为节能降耗的主要目标,建筑节能成为了当今普遍关注的热点问题。如何提高智能建筑中的能源利用效率,合理有效地利用能源,推动建筑节能显得十分重要。目前,在智能建筑中采取的比较广泛的节能手段就是利用楼宇自控系统(BAS)来控制中央空调系统等,虽然在整个智能建筑的控制管理、节能降耗方面取得了一定的成就,但由于BAS是分设备来进行状态的调节,往往各设备不是处于最佳工况,而且可能还会出现牺牲某些设备的运行状态来获取别的参数改善的情况,这样就缺乏一种立足于全局的管理控制措施。从系统的角度来看,节能降耗方面的效果不是很理想。
为此,本文在分析原有智能建筑控制方面不足的基础上,借鉴楼宇自控系统的成功经验,从全局提高节能措施的可操作性入手,针对智能建筑中的主要能耗设备中央空调系统和照明系统的节能降耗问题,从理论和实践两个方面,开发了一种以能量传输效率与能量利用效率为基础的能效分析系统。
本文主要研究了数据仓库技术,并将其应用在所开发的能效分析系统的数据处理模块中。对数据挖掘技术做了初步的研究,并对于将数据挖掘技术应用于系统中的可能性进行了探索,本文所作主要工作如下:
1、针对现有智能建筑介绍了一种基于能量传输效率和能量利用效率的能效理论。新能效理论指出:建筑物能量评价体系不仅包括传统的能量传输效率,还包括能量利用效率。能量利用效率与建筑物中HVAC系统、照明系统的管理和运行模式有关。阐述了能量传输效率的计算方法,介绍能量利用效率的概念及其检测方法,在能效理论的基础上提出能效分析系统的整体设计方案。
2、针对寿光商务小区能效分析系统做了详细分析,开发出较为通用的能效分析软件。从系统的体系结构和功能方面,较为详细地分析并解决了设计和软件实现过程中一些关键性技术问题。
3、在能效分析系统中央数据服务器中引入数据仓库技术,利用数据仓库技术进行数据的清洗、数据格式的转换、主题数据仓库的建立等,建立了多维数据集和转换后的关系数据库,为后续数据挖掘做好准备。
4、以具体挖掘工具SQL Server为例,简单介绍了数据挖掘的操作,同时对数据挖掘算法做了初步了解,为后续深入研究挖掘算法,并从中得出决策分析以支持系统进行更好的能量调度与管理做准备。
最后,在总结以前工作的基础上,对于所开发的能效分析系统存在的缺点和今后工作中需要进一步探索和研究的问题进行了分析。
Facing the increasingly tense of the global energy and the rising energy prices, people attaches great importance to the intensity of the implementation of saving energy. AS the main target of saving energy, building energy conservation becomes a current hot issue of common concern. How to improve the Intelligent Building energy efficiency, rationally and effectively use the energy, it is important to promote energy-saving construction. For example, building automation systems (BAS) are widely employed in public intelligent buildings to monitor and control the operation of HVAC, which make progress in control management and energy savings. But the BAS monitor the system according the equipments, which can lead the equipments not all operation in the best condition. But there may be some equipment expense of the operational status of other parameters to obtain the improvement of the situation, which lack of a foothold in the overall management and control measures. From the system's point of view, the energy saving effect is not very good.
Based on the efficiency of energy transmission and energy usage, a new method of measurement and evaluation of building energy efficiency is described in this thesis. A lot of building energy management strategies on HVAC and lighting system is investigated in order to reduce the energy consumption in intelligent buildings. The probability and the principle of building energy efficiency management systems are systematically introduced based on the energy efficiency theory.
This thesis studies the data warehouse technology and its application in the development of the energy management system of data processing module. A preliminary study is done on the Data Mining techniques and explore the possible application of data mining technology in the system.The thesis's main tasks are as follows:
In section I, a new concept of energy efficiency is introduced in detail by analyzing the existent energy management systems. The evaluation of energy efficiency is composed of energy transmission efficiency and energy usage efficiency. Building energy usage efficiency is connected with modes of operation and management. On the energy transfer efficiency calculation method, introduced the concept of energy efficiency. The concept of energy efficiency and its detection method is introduced. Proposed the design overall energy management system in the theory of energy efficiency.
In section II, done a detailed analysis of Shouguang business district energy management systems, and developed the common energy analysis software. From the system architecture and functionality, more detailed analysis and solution design is done, and a number of key technical issues are solved in the software implementation process.
In section III, developed the data warehouse technology in the central data server of the energy analysis system. Using the data warehouse technology for data cleansing, data format conversion, the establishment of data warehouse theme, etc. The establishment of multi-dimensional data sets and the converted relational database are prepared for the follown data mining.
In section IV, take the mining tools specific to SQL Server as an example, and briefly introduce the operation of data mining. Do a preliminary understanding of the data mining algorithms. Prepare for an in-depth follow-up study on mining algorithms, and attach the system for better control and management measure.
Finally, in conclusion on the basis of previous work, the development of the energy management system and the future of the shortcomings of the work of the need for further exploration and study of the issue.
为此,本文在分析原有智能建筑控制方面不足的基础上,借鉴楼宇自控系统的成功经验,从全局提高节能措施的可操作性入手,针对智能建筑中的主要能耗设备中央空调系统和照明系统的节能降耗问题,从理论和实践两个方面,开发了一种以能量传输效率与能量利用效率为基础的能效分析系统。
本文主要研究了数据仓库技术,并将其应用在所开发的能效分析系统的数据处理模块中。对数据挖掘技术做了初步的研究,并对于将数据挖掘技术应用于系统中的可能性进行了探索,本文所作主要工作如下:
1、针对现有智能建筑介绍了一种基于能量传输效率和能量利用效率的能效理论。新能效理论指出:建筑物能量评价体系不仅包括传统的能量传输效率,还包括能量利用效率。能量利用效率与建筑物中HVAC系统、照明系统的管理和运行模式有关。阐述了能量传输效率的计算方法,介绍能量利用效率的概念及其检测方法,在能效理论的基础上提出能效分析系统的整体设计方案。
2、针对寿光商务小区能效分析系统做了详细分析,开发出较为通用的能效分析软件。从系统的体系结构和功能方面,较为详细地分析并解决了设计和软件实现过程中一些关键性技术问题。
3、在能效分析系统中央数据服务器中引入数据仓库技术,利用数据仓库技术进行数据的清洗、数据格式的转换、主题数据仓库的建立等,建立了多维数据集和转换后的关系数据库,为后续数据挖掘做好准备。
4、以具体挖掘工具SQL Server为例,简单介绍了数据挖掘的操作,同时对数据挖掘算法做了初步了解,为后续深入研究挖掘算法,并从中得出决策分析以支持系统进行更好的能量调度与管理做准备。
最后,在总结以前工作的基础上,对于所开发的能效分析系统存在的缺点和今后工作中需要进一步探索和研究的问题进行了分析。
Facing the increasingly tense of the global energy and the rising energy prices, people attaches great importance to the intensity of the implementation of saving energy. AS the main target of saving energy, building energy conservation becomes a current hot issue of common concern. How to improve the Intelligent Building energy efficiency, rationally and effectively use the energy, it is important to promote energy-saving construction. For example, building automation systems (BAS) are widely employed in public intelligent buildings to monitor and control the operation of HVAC, which make progress in control management and energy savings. But the BAS monitor the system according the equipments, which can lead the equipments not all operation in the best condition. But there may be some equipment expense of the operational status of other parameters to obtain the improvement of the situation, which lack of a foothold in the overall management and control measures. From the system's point of view, the energy saving effect is not very good.
Based on the efficiency of energy transmission and energy usage, a new method of measurement and evaluation of building energy efficiency is described in this thesis. A lot of building energy management strategies on HVAC and lighting system is investigated in order to reduce the energy consumption in intelligent buildings. The probability and the principle of building energy efficiency management systems are systematically introduced based on the energy efficiency theory.
This thesis studies the data warehouse technology and its application in the development of the energy management system of data processing module. A preliminary study is done on the Data Mining techniques and explore the possible application of data mining technology in the system.The thesis's main tasks are as follows:
In section I, a new concept of energy efficiency is introduced in detail by analyzing the existent energy management systems. The evaluation of energy efficiency is composed of energy transmission efficiency and energy usage efficiency. Building energy usage efficiency is connected with modes of operation and management. On the energy transfer efficiency calculation method, introduced the concept of energy efficiency. The concept of energy efficiency and its detection method is introduced. Proposed the design overall energy management system in the theory of energy efficiency.
In section II, done a detailed analysis of Shouguang business district energy management systems, and developed the common energy analysis software. From the system architecture and functionality, more detailed analysis and solution design is done, and a number of key technical issues are solved in the software implementation process.
In section III, developed the data warehouse technology in the central data server of the energy analysis system. Using the data warehouse technology for data cleansing, data format conversion, the establishment of data warehouse theme, etc. The establishment of multi-dimensional data sets and the converted relational database are prepared for the follown data mining.
In section IV, take the mining tools specific to SQL Server as an example, and briefly introduce the operation of data mining. Do a preliminary understanding of the data mining algorithms. Prepare for an in-depth follow-up study on mining algorithms, and attach the system for better control and management measure.
Finally, in conclusion on the basis of previous work, the development of the energy management system and the future of the shortcomings of the work of the need for further exploration and study of the issue.
引文
[1]杨星虎.建筑节能现状与对策[J].能源研究与信息,1997,Vol.13 No.4:36-39.
[2]王娜等.智能建筑概论.北京:人民交通出版社.2002,8.
[3]黄绪镜.百货商场空调设计[M].北京:中国建筑工业出版社,1991.
[4]王天富.空调工程[M].北京:机械工业出版社,2006.
[5]F.C.麦奎斯顿,J.D.帕克,J.D.斯皮特勒.供暖、通风及空气调节[M].俞炳丰译.北京:化学工业出版社,2005.
[6]李先瑞.供热空调系统运行管理、节能、诊断[M].北京:中国电力出版社,2003.
[7]吴晓燕.公共建筑空调系统的节能设计与优化管理[D].长沙:湖南大学,2006.
[8]梁春生,智勇等.中央空调变流量控制节能技术[M].北京:电子工业出版社,2005.
[9]康晓东,黎小沛,饶友玲,张建平.基于数据仓库的数据挖掘技术.北京:机械工业出版社,2004.
[10]W.H.Irmon,Building the Data Warehouse,The third Edition.Copyright.(?)2002by John Wiley &Sons,Inc.21-24.
[11]马丽娜,刘弘,张希林.数据挖掘、OLAP在决策支持系统中的应用.计算机应用研究,2001,第11期:10-12.
[12]David Hand,Heikki Mannila,Padhraic Smyth.数据挖掘原理[M].张银奎,廖丽,宋俊译.北京:机械工业出版社:2003.
[13]Andrzej Czyzewski.Automatic identification of sound source position employing neural networks and rough sets[J].Pattem Recognition Letters.2003,24:921-933.
[14]陈京民等.数据仓库与数据挖掘技术.北京:电子工业出版社,2002.
[15](美)米哈尔斯基等.机器学习与数据挖掘:方法和应用.北京:电子工业出版社,2004.
[16]王美华.数据挖掘领域中的聚类方法[J].南华大学学报.2004,18(1).
[17]邹志文,朱金伟.数据挖掘算法研究与综述[J].计算机工程.2005,26(9).
[18]武森,高学东,[德]M.巴斯蒂安.数据仓库与数据挖掘技术.北京:冶金工业出版 社,2003.
[19]萨师煊,王珊.数据库系统概论[M].北京:高等教育出版社,2000.
[20]杜晓通,孙毅,袁亚平.基于IP技术的能效分析系统的设计与实现[A].智能建筑与绿色建筑文集3[C].北京:中国建筑工业出版社,2007,618-620.
[21](美)Lawrence A.klein著.多传感器数据融合理论及应用.北京:北京理工大学出版社,2004.
[22]孙毅.基于能效理论的能效分析系统软件开发.山东大学硕士论文,2007.
[23]杨万海编著.多传感器数据融合及其应用.西安:西安电子科技大学出版社,2004
[24][美]Pete Loshin著.刘谦等译.TCP/IP透彻理解.北京:电子工业出版社,2003
[25][美]Rob Scrimger著.赵刚等译.TCP/IP宝典.北京:电子工业出版,2002.
[26]王达.计算机网络远程控制制.北京:清华大学出版社,2003.
[27]陈坚等.Visual C++网络高级编程.北京:人民邮电出版社,2002.
[28]易君著.Visual C++.NET网络编程.北京:中国铁道出版社,2003.
[29]汪翔等.Visual C++实践与提高 网络编程篇.北京:中国铁道出版社,2001.
[30]姚春燕等.利用Winsock实现Palm与PC机的远程数据通信.计算机应用,Jan,2003.
[31]刘刀桂,孟繁晶.Visual C++实践与提高 数据库篇.北京:中国铁道出版社,2001.
[32]四维科技,沈炜,徐慧.Visual C++数据库编程技术与实例.北京:人民邮电出版社,2005.
[33]刘静执.数据仓库技术在智能建筑中的应用分析.北京:建筑工程学院学报,2001.12:45-47.
[34]C.Skalak,E.Rissland.Arguments and cases:An inevitable twining.Artificial Intelligence and Law,An International Journal,1992,1(1):3-48
[35][美]Claude Seidman.SQL Server 2000数据挖掘技术指南.北京:机械工业出版社,2002.
[37]梁循.数据挖掘算法与应用.北京:北京大学出版社,2006.4.
[38]QuinlanJ.R.Induction of DecisionTrees.Machine learning.1986(1):81-106
[39]Breiman.L,Friedman.J,Olshen.R.et al.Classification and Regression Trees. Monterey,CA.Wadsworth International Group.1994.
[40]丽华,吉根林.决策树分类技术研究.计算机工程.2004.5 30(9)94-97.
[41]Mehta M.Rissanen J.AgrawalcR.MDL-Based Decision Tree Pruning.In Proc eedings of KDD'98.
[42]Yu-Shan Shih.A Note on Split Selection Bias in Classification.Computational Statistics and Data Analysis 2004(45)457-466.
[43]Loh W.Y,Smith YS.Split Selection methods for Classification Trees.Statistics Sinica.1997(7):815-840.
[44]Kass GV.An Exploratory for Investigating Large Quantities of Categorical Data.Applied Staistics.1998(29):119-127.
[45]Paul E.Utgoff.Incremental Induction of Decision Trees.Machine Learning.1997(4):161-186.
[46]Utgoff P.E.An Incremental ID3.In Proceeding 5th International Conference Ma chine Learning.San Mateo,CA.1988:107-120.
[47]Murthy K.V.S.On Growing Better Decision Tree from Data.P.H.D.Thesis of John Hopkins University,1995.
[48]邵峰晶,于忠清.数据挖掘原理与算法.中国水利水电出版社.2003.8.
[49][美]Mehmed Kantardzic.数据挖掘.北京:清华大学出版社,2003.
[50]刘小石,郑淮,杨帆.精通Visual C++6.0.北京:清华大学出版社,2000.
[51]周生炳,万映辉,邸晓奕.面向对象程序设计.北京:清华大学出版社,2004.
[52]甘玲,丘劲.面向对象技术与Visual C++.北京:清华大学出版社,2004.
[53]Timothy C.Lethbridge,Robert Langaniere著.面向对象软件工程.北京:机械工业出版社,2003.
[54]http://www.itisedu.com/phrase/200604240956125.html.
[55]AmizicB.et al.Two Dimensional Wavelet Mapping Techniques for Damage Detection[R].Smart Structures and Materials.San Diego.2002,4693-32.
[2]王娜等.智能建筑概论.北京:人民交通出版社.2002,8.
[3]黄绪镜.百货商场空调设计[M].北京:中国建筑工业出版社,1991.
[4]王天富.空调工程[M].北京:机械工业出版社,2006.
[5]F.C.麦奎斯顿,J.D.帕克,J.D.斯皮特勒.供暖、通风及空气调节[M].俞炳丰译.北京:化学工业出版社,2005.
[6]李先瑞.供热空调系统运行管理、节能、诊断[M].北京:中国电力出版社,2003.
[7]吴晓燕.公共建筑空调系统的节能设计与优化管理[D].长沙:湖南大学,2006.
[8]梁春生,智勇等.中央空调变流量控制节能技术[M].北京:电子工业出版社,2005.
[9]康晓东,黎小沛,饶友玲,张建平.基于数据仓库的数据挖掘技术.北京:机械工业出版社,2004.
[10]W.H.Irmon,Building the Data Warehouse,The third Edition.Copyright.(?)2002by John Wiley &Sons,Inc.21-24.
[11]马丽娜,刘弘,张希林.数据挖掘、OLAP在决策支持系统中的应用.计算机应用研究,2001,第11期:10-12.
[12]David Hand,Heikki Mannila,Padhraic Smyth.数据挖掘原理[M].张银奎,廖丽,宋俊译.北京:机械工业出版社:2003.
[13]Andrzej Czyzewski.Automatic identification of sound source position employing neural networks and rough sets[J].Pattem Recognition Letters.2003,24:921-933.
[14]陈京民等.数据仓库与数据挖掘技术.北京:电子工业出版社,2002.
[15](美)米哈尔斯基等.机器学习与数据挖掘:方法和应用.北京:电子工业出版社,2004.
[16]王美华.数据挖掘领域中的聚类方法[J].南华大学学报.2004,18(1).
[17]邹志文,朱金伟.数据挖掘算法研究与综述[J].计算机工程.2005,26(9).
[18]武森,高学东,[德]M.巴斯蒂安.数据仓库与数据挖掘技术.北京:冶金工业出版 社,2003.
[19]萨师煊,王珊.数据库系统概论[M].北京:高等教育出版社,2000.
[20]杜晓通,孙毅,袁亚平.基于IP技术的能效分析系统的设计与实现[A].智能建筑与绿色建筑文集3[C].北京:中国建筑工业出版社,2007,618-620.
[21](美)Lawrence A.klein著.多传感器数据融合理论及应用.北京:北京理工大学出版社,2004.
[22]孙毅.基于能效理论的能效分析系统软件开发.山东大学硕士论文,2007.
[23]杨万海编著.多传感器数据融合及其应用.西安:西安电子科技大学出版社,2004
[24][美]Pete Loshin著.刘谦等译.TCP/IP透彻理解.北京:电子工业出版社,2003
[25][美]Rob Scrimger著.赵刚等译.TCP/IP宝典.北京:电子工业出版,2002.
[26]王达.计算机网络远程控制制.北京:清华大学出版社,2003.
[27]陈坚等.Visual C++网络高级编程.北京:人民邮电出版社,2002.
[28]易君著.Visual C++.NET网络编程.北京:中国铁道出版社,2003.
[29]汪翔等.Visual C++实践与提高 网络编程篇.北京:中国铁道出版社,2001.
[30]姚春燕等.利用Winsock实现Palm与PC机的远程数据通信.计算机应用,Jan,2003.
[31]刘刀桂,孟繁晶.Visual C++实践与提高 数据库篇.北京:中国铁道出版社,2001.
[32]四维科技,沈炜,徐慧.Visual C++数据库编程技术与实例.北京:人民邮电出版社,2005.
[33]刘静执.数据仓库技术在智能建筑中的应用分析.北京:建筑工程学院学报,2001.12:45-47.
[34]C.Skalak,E.Rissland.Arguments and cases:An inevitable twining.Artificial Intelligence and Law,An International Journal,1992,1(1):3-48
[35][美]Claude Seidman.SQL Server 2000数据挖掘技术指南.北京:机械工业出版社,2002.
[37]梁循.数据挖掘算法与应用.北京:北京大学出版社,2006.4.
[38]QuinlanJ.R.Induction of DecisionTrees.Machine learning.1986(1):81-106
[39]Breiman.L,Friedman.J,Olshen.R.et al.Classification and Regression Trees. Monterey,CA.Wadsworth International Group.1994.
[40]丽华,吉根林.决策树分类技术研究.计算机工程.2004.5 30(9)94-97.
[41]Mehta M.Rissanen J.AgrawalcR.MDL-Based Decision Tree Pruning.In Proc eedings of KDD'98.
[42]Yu-Shan Shih.A Note on Split Selection Bias in Classification.Computational Statistics and Data Analysis 2004(45)457-466.
[43]Loh W.Y,Smith YS.Split Selection methods for Classification Trees.Statistics Sinica.1997(7):815-840.
[44]Kass GV.An Exploratory for Investigating Large Quantities of Categorical Data.Applied Staistics.1998(29):119-127.
[45]Paul E.Utgoff.Incremental Induction of Decision Trees.Machine Learning.1997(4):161-186.
[46]Utgoff P.E.An Incremental ID3.In Proceeding 5th International Conference Ma chine Learning.San Mateo,CA.1988:107-120.
[47]Murthy K.V.S.On Growing Better Decision Tree from Data.P.H.D.Thesis of John Hopkins University,1995.
[48]邵峰晶,于忠清.数据挖掘原理与算法.中国水利水电出版社.2003.8.
[49][美]Mehmed Kantardzic.数据挖掘.北京:清华大学出版社,2003.
[50]刘小石,郑淮,杨帆.精通Visual C++6.0.北京:清华大学出版社,2000.
[51]周生炳,万映辉,邸晓奕.面向对象程序设计.北京:清华大学出版社,2004.
[52]甘玲,丘劲.面向对象技术与Visual C++.北京:清华大学出版社,2004.
[53]Timothy C.Lethbridge,Robert Langaniere著.面向对象软件工程.北京:机械工业出版社,2003.
[54]http://www.itisedu.com/phrase/200604240956125.html.
[55]AmizicB.et al.Two Dimensional Wavelet Mapping Techniques for Damage Detection[R].Smart Structures and Materials.San Diego.2002,4693-32.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |