Please wait a minute...
浙江大学学报(工学版)
J4  2013, Vol. 47 Issue (4): 735-740    DOI: 10.3785/j.issn.1008-973X.2013.04.026
化学与环境工程     
基于多因子融合的水质异常检测算法
何慧梅, 侯迪波, 赵海峰, 黄平捷, 张光新
浙江大学 工业控制技术国家重点实验室,浙江 杭州 310027
Multi-parameters fusion algorithm for detecting anomalous water quality
HE Hui-mei, HOU Di-bo, ZHAO Hai-feng, HUANG Ping-jie, ZHANG Guang-xin
State Key Laboratory of Industrial Control Technology, Zhejiang University, Hangzhou 310027, China
 全文: PDF  HTML
摘要：

为了及时有效地发现偶发或蓄意的水质异常,保障饮用水供水安全,在总结现有异常检测算法的基础上,提出基于自回归（AR）模型和模糊C-均值聚类（FCM）的多因子融合水质异常检测算法.通过AR模型实现水质背景信号的高精度跟踪,采用FCM算法融合多种水质指标的AR预测残差,与设定阈值作比较并判断异常.实验结果表明,与单因子算法、直接融合水质指标监测值的算法、利用多维欧氏距离融合AR预测残差的算法相比,提出的多因子融合水质异常检测算法具有更高的异常检出率和较低的误报率.
Abstract:

 Rapid detection of anomalous water quality within a water quality early warning system is desirable for the protection of drinking water against both accidental and malevolent contamination events. Using existing data from in-situ water quality sensors, a multi-parameters fusion algorithm was presented based on AR model and fuzzy C-means clustering algorithm (FCM). AR model uses previous observations of the water quality to predict future water quality values. Then prediction deviations can be caculated and classificated into two clusters within FCM. To discriminate between normal and anomalous water quality, the distance between a new prediction deviation and the anomalous cluster center was calculated and compared to a constant threshold. Results show that this fusion algorithm produces the lowest false alarm rate (FAR) and highest positive detection (PD) for all cases of simulated event strength, which has better performance than single-parameter algorithm, multi-dimension algorithms fused water quality data directly and multi-dimension Euclidean distance algorithm fused predictive residuals.
出版日期: 2013-04-01
:  X 832  
基金资助:

国家自然科学基金资助项目(41101508)；国家重大科技专项资助项目（2008ZX07420-004）.
通讯作者: 侯迪波,男,副教授.     E-mail: houdb@zju.edu.cn
作者简介: 何慧梅（1987—）,女,硕士生,从事水质事件检测技术研究. E-mail: hhmrabbit@163.com
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章　　

引用本文:

何慧梅, 侯迪波, 赵海峰, 黄平捷, 张光新. 基于多因子融合的水质异常检测算法[J]. J4, 2013, 47(4): 735-740.

HE Hui-mei, HOU Di-bo, ZHAO Hai-feng, HUANG Ping-jie, ZHANG Guang-xin. Multi-parameters fusion algorithm for detecting anomalous water quality. J4, 2013, 47(4): 735-740.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2013.04.026    或    http://www.zjujournals.com/eng/CN/Y2013/V47/I4/735

［1］ MCKENNA S A, HART D B, KLISE K A, et al. Water quality event detection systems for drinking water contamination warning systems ［R］. New York: U.S. Environmental Protection Agency, 2010.

［2］ MAHONEY M V. A machine learning approach to detecting attacks by identifying anomalies in network traffic ［D］. Melbourne: Florida Institute of Technology, 2003.

［3］ 蒋盛益. 基于聚类的入侵检测算法研究［M］. 武汉: 科学出版社, 2008: 10-26.

［4］ CHEU R L, RITCHIE S G. Automated detection of lane-blocking freeway incidents using artificial neural networks ［J］. Transportation Research Part C: Emerging Technologies, 1995, 3(6): 371-388.

［5］ SMID R, DOCEKAL A, KREIDL M. Automated classification of eddy current signatures during manual inspection ［J］. NDT &amp|E International, 2005, 38(6): 462-470.

［6］ KLISE K A, MCKENNA S A. Water quality change detection: multivariate algorithms ［C］∥ Optics and Photonics in Global Homeland Security II. United States: SPIE, 2006.

［7］ MCKENNA S A, KLISE K A, WILSON M P. Testing water quality change detection algorithms ［C］∥8th Annual Water Distribution Systems Analysis Symposium. United States: ASCE, 2006: 132-145.

［8］ ALLGEIER S, MURRAY R, MCKENNA S A, et al. Overview of event detection systems for water sentinel ［R］. New York: U.S. Environmental Protection Agency, 2005.

［9］ KLISE K A, MCKENNA S A. Multivariate applications for detecting anomalous water quality ［C］∥ 8th Annual Water Distribution Systems Analysis Symposium. United States: \
[s.n.\], 2006: 120-130.

［10］ 吴富梅,杨元喜. 基于高阶AR模型的陀螺随机漂移模型［J］. 测绘学报, 2007(4): 389-394.

WU Fu-mei, YANG Yuan-xi. Gyroscope random drift model based on the higher-order AR model ［J］. ACTA Geodaetica et Carto Graphica Sinica, 2007(4): 389-394.

［11］ 杨叔子,吴雅,轩建平. 时间序列分析的工程应用(上)［M］. 武汉: 华中科技大学出版社, 2007: 176-189.

［12］ 高新波. 模糊聚类分析及其应用［M］. 西安: 西安电子科技大学出版社, 2004: 133-136.

［13］ 齐淼,张化祥. 改进的模糊C-均值聚类算法研究［J］. 计算机工程与应用, 2009, 45(20): 133-135.

QI Miao, ZHANG Hua-xiang. Research on modified fuzzy C-means clustering algorithm ［J］. Computer Engineering and Applications, 2009, 45(20): 133-135.

［14］ HALL J, SZABO J. WaterSentinel online water quality monitoring as an indicator of drinking water contamination ［R］. New York: U.S. Environmental Protection Agency, 2005.

［15］ MCKENNA S A, HART D B, MURRAY R, et al. Testing and evaluation of water quality event detection algorithms ［R］. U.S.: Sandia National Laboratories, 2011.

［16］ JEFFREY Y Y, HAUGHY R C, GOODRICH J A. Real-time contaminant detection and classification in a drinking water pipe using conventional water quality sensors: techniques and experimental results ［J］. Journal of Environmental Management, 2009, 90(8): 2494-2506.
No related articles found!