您的位置:首页  > 论文页面

城市用水量曲线聚类算法的研究与实现

发表时间:2020-07-01  浏览量:155  下载量:14
全部作者: 刘春柳,张征
作者单位: 华中科技大学人工智能与自动化学院
摘 要: 准确预测城市用水量可以对智慧水务调度、报警提供支持,预测前对所有用水量曲线进行聚类可以提高预测的精度。为满足实时性和运行效率的要求,提出基于形态特征的分段聚合近似(shape-based piecewise aggregate approximation,SPAA)表示方法,同时为解决传统基于欧氏距离的聚类算法无法包含曲线的形状特征的问题,提出自适应聚类数的基于序列形态相似性的k-shape聚类算法。另外,采用一种基于质心的聚类中心计算方式,提取每类用水量曲线形态。最后,对某水务公司的用水量数据实例进行聚类分析。结果表明,本文所提的SPAA-k-shape算法可以有效降维,减少聚类计算时间,比传统仅考虑欧氏距离的算法更准确。
关 键 词: 市政工程;模式识别与智能系统;曲线聚类;k-shape算法;基于形态特征的分段聚合近似
Title: Research and implementation on curve clustering algorithm for urban water consumption
Author: LIU Chunliu, ZHANG Zheng
Organization: School of Artificial Intelligence and Automation, Huazhong University of Science and Technology
Abstract: Accurate prediction of urban water consumption can provide support for smart water dispatching and alarming. Clustering all the water consumption curves before forecasting can improve the accuracy of prediction. In order to meet the requirements of real-time performance and operational efficiency, a shape-based piecewise aggregate approximation (SPAA) method is proposed. At the same time, in order to solve the problem that the traditional clustering algorithm based on Euclidean distance cannot include the shape features of curves, a k-shape clustering algorithm based on sequence shape similarity of adaptive clustering numbers is proposed. In addition, a centroid-based clustering center calculation method is adopted to extract the curve shape of water consumption from each type of cluster. Finally, the clustering algorithm is applied to analyze the water consumption data from a water supply company. The result shows that the SPAA-k-shape algorithm proposed in this paper effectively reduces the dimension and shortens the clustering calculation time, which is more accurate than the traditional algorithms that only consider Euclidean distance.
Key words: municipal engineering; pattern recognition and intelligent system; curve clustering; k-shape algorithm; shape-based piecewise aggregate approximation (SPAA)
发表期数: 2020年6月第2期
引用格式: 刘春柳,张征. 城市用水量曲线聚类算法的研究与实现[J]. 中国科技论文在线精品论文,2020,13(2):212-220.
 
0 评论数 0
暂无评论
友情链接