再生水与自来水管道错接事故探测方法
|
摘要
再生水与自来水管道错接会导致严重的用水安全事故。通过开展管道错接模拟试验,开发了基于多常规水质指标协同反馈和支持向量机的管道错接探测算法。结果表明,本试验所用城市杂用再生水混入自来水后,会导致后者余氯、pH、氧化还原电位等水质指标降低,浊度和电导率升高;管道错接探测算法经训练后对1%、5%和10%掺混比的管道错接探测准报率分别为91.7%、98.3%和98.3%(平均为96.1%),误报率为0。研究结果有助于实时探测再生水输配中的安全风险,保障用户用水安全。
Cross-connections between potable and reclaimed water distribution pipes can cause serious water security events. In this study, experiments simulating cross-connections were carried out and an algorithm detecting cross-connections based on coordinated responses of multiple conventional water quality sensors and a support vector machine was developed. Results showed that residual chlorine, pH, and oxidation-reduction potential decreased, and turbidity and conductivity increased when reclaimed water for municipal use was mixed with potable water in this study. The algorithm performed well after being trained in advance. True positive rates were 91.7%, 98.3%, and 98.3% when the percentage of reclaimed water in the mixed water was 1%, 5%, and 10%, respectively(the average true positive rate was 96.1%). And false positive rates were 0. The results are useful for detecting reclaimed water distribution risk and ensuring water supply security.
Cross-connections between potable and reclaimed water distribution pipes can cause serious water security events. In this study, experiments simulating cross-connections were carried out and an algorithm detecting cross-connections based on coordinated responses of multiple conventional water quality sensors and a support vector machine was developed. Results showed that residual chlorine, pH, and oxidation-reduction potential decreased, and turbidity and conductivity increased when reclaimed water for municipal use was mixed with potable water in this study. The algorithm performed well after being trained in advance. True positive rates were 91.7%, 98.3%, and 98.3% when the percentage of reclaimed water in the mixed water was 1%, 5%, and 10%, respectively(the average true positive rate was 96.1%). And false positive rates were 0. The results are useful for detecting reclaimed water distribution risk and ensuring water supply security.
引文
[1] 北京市统计局. 北京统计年鉴[M]. 北京: 中国统计出版社, 2016.
[2] 赵晗,张帆,罗瑞垚.北京回龙观局部自来水异常,或为突发性中水污染[EB/OL].http://info.water.hc360.com/2016/05/252059551946.shtml,2016-05-25.
[3] 京华时报.业主私接管线导致25户家庭自来水掺入中水[EB/OL]. http://society.people.com.cn/GB/136657/17238320.html,2012-02-28.
[4] 北京青年报. 中水污染自来水, 本市已发生多起[EB/OL]. http://news.sina.com.cn/c/2012-05-10/000024394665.shtml, 2012-05-10.
[5] Liu S, Che H, Smith K, et al. A real time method of contaminant classification using conventional water quality sensors[J]. Journal of Environmental Management, 2015,(154): 13-21.
[6] Hall J, Zaffiro A D, Marx R B, et al. On-line water quality parameters as indicators of distribution system contamination[J]. American Water Works Association, 2007, 99(1): 66-77.
[7] Jeffrey Yang Y, Haught 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.
[8] Friedler E, Alfiya Y, Shaviv A, et al. A continuous active monitoring approach to identify cross-connections between potable water and effluent distribution systems[J]. Environmental Monitoring and Assessment, 2015, 187(3): 131-148.
[9] Hambly A C, Henderson R K, Baker A, et al. Application of portable fluorescence spectrophotometry for integrity testing of recycled water dual distribution systems[J]. Applied Spectroscopy, 2015, 69(1): 124-129.
[10] Hambly A, Henderson R K, Baker A, et al. Probabilistic analysis of fluorescence signals for monitoring dual reticulation water recycling schemes[J]. Water Science & Technology, 2010, 62(9): 2059-2065.
[11] Liu S, Smith K, Che H. A multivariate based event detection method and performance comparison with two baseline methods[J]. Water Research, 2015,(80): 109-118.
[12] 周志华. 机器学习[M]. 北京: 清华大学出版社, 2016.
[2] 赵晗,张帆,罗瑞垚.北京回龙观局部自来水异常,或为突发性中水污染[EB/OL].http://info.water.hc360.com/2016/05/252059551946.shtml,2016-05-25.
[3] 京华时报.业主私接管线导致25户家庭自来水掺入中水[EB/OL]. http://society.people.com.cn/GB/136657/17238320.html,2012-02-28.
[4] 北京青年报. 中水污染自来水, 本市已发生多起[EB/OL]. http://news.sina.com.cn/c/2012-05-10/000024394665.shtml, 2012-05-10.
[5] Liu S, Che H, Smith K, et al. A real time method of contaminant classification using conventional water quality sensors[J]. Journal of Environmental Management, 2015,(154): 13-21.
[6] Hall J, Zaffiro A D, Marx R B, et al. On-line water quality parameters as indicators of distribution system contamination[J]. American Water Works Association, 2007, 99(1): 66-77.
[7] Jeffrey Yang Y, Haught 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.
[8] Friedler E, Alfiya Y, Shaviv A, et al. A continuous active monitoring approach to identify cross-connections between potable water and effluent distribution systems[J]. Environmental Monitoring and Assessment, 2015, 187(3): 131-148.
[9] Hambly A C, Henderson R K, Baker A, et al. Application of portable fluorescence spectrophotometry for integrity testing of recycled water dual distribution systems[J]. Applied Spectroscopy, 2015, 69(1): 124-129.
[10] Hambly A, Henderson R K, Baker A, et al. Probabilistic analysis of fluorescence signals for monitoring dual reticulation water recycling schemes[J]. Water Science & Technology, 2010, 62(9): 2059-2065.
[11] Liu S, Smith K, Che H. A multivariate based event detection method and performance comparison with two baseline methods[J]. Water Research, 2015,(80): 109-118.
[12] 周志华. 机器学习[M]. 北京: 清华大学出版社, 2016.