[1]林康强,林育松.基于模糊气候聚类和改进BP神经网络的建筑气候数据清洗方法[J].西安建筑科技大学学报(自然科学版),2021,(02):275-282.[doi:10.15986/j.1006-7930.2021.02.017]
 LIN Kangqiang,LIN Yusong.Building climate data cleaning method based on fuzzy climate clustering and improved BP neural network[J].J. Xi'an Univ. of Arch. & Tech.(Natural Science Edition),2021,(02):275-282.[doi:10.15986/j.1006-7930.2021.02.017]
点击复制

基于模糊气候聚类和改进BP神经网络的建筑气候数据清洗方法
分享到:

西安建筑科技大学学报(自然科学版)[ISSN:1006-7930/CN:61-1295/TU]

卷:
期数:
2021年02期
页码:
275-282
栏目:
出版日期:
2021-04-28

文章信息/Info

Title:
Building climate data cleaning method based on fuzzy climate clustering and improved BP neural network
文章编号:
1006-7930(2021)02-0275-08
作者:
林康强1林育松2
(1.广州美术学院 建筑艺术设计学院,广东 广州 510006; 2.香港中文大学(深圳)深圳高等金融研究院,广东 深圳 518000)
Author(s):
LIN Kangqiang1 LIN Yusong2
(1.School of Architecture and Applied Art, Guangzhou Academy of Fine Arts, Guangzhou 510006, China; 2.Shenzhen Finance Institute,The Chinese University of Hong Kong(Shenzhen),Shenzhen 518000, China)
关键词:
建筑节能 气候数据 数据清洗 自适应聚类 BP神经网络
Keywords:
building energy saving climate data data cleaning adaptive clustering BP neural network
分类号:
TU201.5;TP389.1
DOI:
10.15986/j.1006-7930.2021.02.017
文献标志码:
A
摘要:
针对建筑节能气候数据质量较差的问题,提出一种基于K-MEANS的模糊气候聚类和改进BP神经网络模型的建筑物气候数据清洗方法.首先利用K-MEANS算法根据数据相关性将其划分为不同子类,针对K-MEANS聚类个数和初始聚类中心的选取问题,将主分量分析(Principal Component Analysis,PCA)与K-MEANS结合,利用PCA的主分量作为初始聚类中心; 然后利用BP神经网络对每个子类分别构建数据清洗模型,降低运算复杂度,同时利用遗传模拟退火(Genetic Simulated Annealing, GSA)算法对BP神经网络的初值进行全局寻优,解决BP网络参数选择困难、易陷入局部极值问题的同时提升模型的数据清洗性能.采用某市实际气候数据开展试验,对所提方法的数据清洗性能进行验证,结果表明所提方法可以获得优于94%的清洗效率,并且在小样本情况下具备稳健性.
Abstract:
AbstractIn view of the poor quality of building energy-saving climate data, a method of building climate data cleaning based on K-means fuzzy climate clustering and improved BP neural network model is proposed. Firstly, the K-means algorithm is used to divide the data into different sub classes according to the data correlation. Aiming at the problem of selecting the number of K-means clusters and the initial clustering center, the principal component analysis(PCA)is used to analyze the cluster number and the initial cluster center. The principal component of PCA is used as the initial clustering center, and then BP neural network is used to construct data cleaning model for each subclass to reduce the computational complexity. Meanwhile, GSA algorithm optimizes the initial value of BP neural network globally, solves the difficulty of parameter selection of BP neural network, avoids the problem of local extremum, and improves the data cleaning performance of the model. The results show that the proposed method can achieve a cleaning efficiency higher than 94%, and is robust in the case of small samples.

参考文献/References:

[1]李建成.建筑节能的基础工作一建筑气候基础数据建设[J].能源工程,2002(6):17-20.
LI Jiancheng. Improvement of basic climate data for architecture-A foundational task of energy efficiency in building[J].Energy Engineering,2002(6):17-20.
[2]李红莲,杨柳.不同古典型气象年生成方法对建筑能耗的影响[J].暖通空调,2015,45(9):59-63
LI Honglian, YANG Liu. Effect of several methods for generating typical meteorological year on building energy consumption[J].HV&AC, 2015,45(9):59-63.
[3]张燕.基于聚类算法的数据清洗的研究与实现[D].保定:华北电力大学,2007.
ZHANG Yan. Research and implementation of data cleansing Based on clustering algorithm[D]. Baoding:North China Electric Power University, 2007.
[4]李亚坤.基于网络的数据清洗技术研究[D].哈尔滨:哈尔滨工业大学,2013.
LI Yakun. Research on data cleaning using web information[D].Harbin:Harbin Institute of Technology, 2013.
[5]WANG Xuyang,ZHANG Pengyuan,NA Xingyu,et al.Handling 00V.words in mandarin-spoken term detectio with hierarchical n-Gram language model[J].Chinese Journal of Electronics,2017,26(6):1239-1244.
[6]BOHANNON-P,FAN-W,GEERTS F,et al. Conditional functional-dependencies-for-data cleaning[C]//ICDE2007:Proceedings of the 2007 IEEE 234.International Conference on Data Engineering.Piscatawax:IEEE,2007:746-755.
[7]陈晋音,何辉. 基于密度的聚类中心自动确定的混合属性数据聚类算法研究[J].自动化学报,2015,41(10):1798-1831.
CHEN Jinyin, HE Hui. Research on density-based clustering algorithm for mixed data with determine cluster centers automatically[J].ACTA AUTOMATICA SINICA, 2015,41(10):1798-1831.
[8]李昌华,卜亮亮,刘欣.基于聚类和神经网络对建筑节能气候数据清洗的算法[J].计算机应用,2018,38(S1):83-86.
LI Changhua, BU Liangliang, LIU Xin. Building energy saving climate data cleaning algorithm based on clustering and neural network[J].Journal of Computer Applications, 2018,38(S1):83-86.
[9]和诺, 马苗苗. 一种改进的K均值微博热点话题发现方法[J].数据通信, 2019(1):31-35.
HE Nuo, MA Miaomiao. An improved K-means microblog hot topic discovery method[J].Data Communication, 2019(1):31-35.
[10]陈宝楼.K-Means算法研究及在本文聚类中的应用[D].合肥:安徽大学,2013:9-22
CHEN Baolou. The Research and Application in Text Clustering of K-Means Algorithm[D].Hefei:Anhui University, 2013:9-22.
[11]JOLLIFFE, I, T, Principal Component Analysis[M]. New York: Springer-Verlag. 1986.
[12]裴瑞, 白尚旺, 党伟超, 等. 自适应遗传退火算法优化BP神经网络及其应用[J].计算机系统应用, 2019, 28(7):109-113.
PEI Rui, BAI Shangwang, DANG Weichao, et al. Adaptive Genetic Annealing Algorithm for Optimizing BP Neural Network and its application[J].Computer Systems & Applications, 2019, 28(7):109-113.
[13]叶林, 陈政, 赵永宁, 等. 基于遗传算法—模糊径向基神经网络的光伏发电功率预测模型[J].电力系统自动化, 2015, 39(16):16-22.
YE Lin, CHEN Zheng, ZHAO Yongning, et al. Photovoltaic power forecasting model based on genetic algorithm and fuzzy radial basis function neural network[J].Automation of Electric Power Systems, 2015, 39(16):16-22.
[14]张子成, 韩伟, 毛波. 基于模拟退火的自适应离散型布谷鸟算法求解旅行商问题[J].电子学报, 2019, 46(8):1850-1857.
ZHANG Zicheng, HAN Wei, MAO Bo. Adaptive discrete cuckoo algorithm based on simulated annealing for sovling TSP[J].Acta Electronica Sinica,2019, 46(8):1850-1857.
[15]杨志军, 陈超然, 黄观新. 面向机器人优化设计的GA-非均匀 Kriging-梯度投影混合全局优化算法[J].机械工程学报, 2019, 55(11):61-68.
YANG Zhijun, CHEN Chaoran, HUANG Guanxin. GA non-uniform Kriging gradient projection hybrid global optimization algorithm for robot optimization design[J].Journal of Mechanical Engineering, 2019, 55(11):61-68.
(编辑 桂智刚)

相似文献/References:

[1]刘艳峰,陈迎亚,王登甲,等.垂直绿化对室内热环境影响测试研究[J].西安建筑科技大学学报(自然科学版),2015,(03):423.[doi:10.15986/j.1006-7930.2015.03.021]
 LIU Yanfeng,CHEN Yingya,WANG Dengjia,et al.Research of vertical green impact on the indoor thermal environment[J].J. Xi'an Univ. of Arch. & Tech.(Natural Science Edition),2015,(02):423.[doi:10.15986/j.1006-7930.2015.03.021]
[2]张 明,杨 柳,冯旭明.基于EnergyPlus的几种典型平面形状的塔式办公建筑的能耗差异分析[J].西安建筑科技大学学报(自然科学版),2013,(04):565.[doi:10.15986/j.1006-7930.2013.04.017]
 ZHANG Ming,YANG Liu,FENG Xu-ming.Energy consumption difference analysis of some typical geometry office tower based on EnergyPlus[J].J. Xi'an Univ. of Arch. & Tech.(Natural Science Edition),2013,(02):565.[doi:10.15986/j.1006-7930.2013.04.017]
[3]赵 亮,张吉礼,梁若冰.建筑能耗数据特性分析及优化采样方法研究[J].西安建筑科技大学学报(自然科学版),2013,(05):688.[doi:10.15986/j.1006-7930.2013.05.014]
 ZHAO Liang,ZHANG Ji-li,LIANG Ruo-bing.Analysis on the characteristics of building energy consumption data and research of the optimized sampling method[J].J. Xi'an Univ. of Arch. & Tech.(Natural Science Edition),2013,(02):688.[doi:10.15986/j.1006-7930.2013.05.014]
[4]徐 鹏1,赵嘉靖2,李俊明2.一种新型蒸发冷却式通风外墙的热工性能研究[J].西安建筑科技大学学报(自然科学版),2011,(04):546.[doi:DOI:10.15986/j.1006-7930.2011.04.020]
 ,Thermal performance study on a new modular evaporativecooling ventilated wall[J].J. Xi'an Univ. of Arch. & Tech.(Natural Science Edition),2011,(02):546.[doi:DOI:10.15986/j.1006-7930.2011.04.020]
[5]张 群1,2,梁 锐1,等.宁夏地区乡村民居节能设计研究[J].西安建筑科技大学学报(自然科学版),2011,(04):570.[doi:DOI:10.15986/j.1006-7930.2011.04.023]
 ,,et al.Energy efficiency of rural dwellings in Ningxia Hui Autonomous Region[J].J. Xi'an Univ. of Arch. & Tech.(Natural Science Edition),2011,(02):570.[doi:DOI:10.15986/j.1006-7930.2011.04.023]

备注/Memo

备注/Memo:
收稿日期:2020-11-11 修改稿日期:2021-03-20
基金项目:国家自然科学基金资助项目(51308218)
第一作者:林康强(1989-)男,博士,讲师,主要研究方向为数字建筑、参数化建筑等.E-mail:Linkangqiang577@163.com
通信作者:林育松(1990-)男,硕士,高级审计师,主要研究方向为金融统计学和数理统计学.E-mail:sendlys@126.com
更新日期/Last Update: 2021-04-28