A Decision Tree Approach to the Risk Evaluation of Urban Water Distribution Network Pipes
Abstract
:1. Introduction
2. Materials and Methods
2.1. Data
2.2. Preprocessing of Variables
2.3. Data Preprocessing
2.4. Model
3. Results
3.1. Output Model
3.2. Fifty-Fold Cross-Validation
3.3. Model Performance
4. Discussion
Author Contributions
Funding
Conflicts of Interest
References
- Liu, X.Y.; Shu, S.H. Policy on water loss control in China. J. Geosci. Environ. Prot. 2018, 6, 100–107. [Google Scholar] [CrossRef] [Green Version]
- China Urban Water Association. Urban Water Supply Statistical Yearbook; China Urban Water Association: Beijing, China, 2017. [Google Scholar]
- Majidi Khalilabad, N.; Mollazadeh, M.; Akbarpour, A.; Khorashadizadeh, S. Leak detection in water distribution system using non-linear kalman filter. Int. J. Optim. Civil Eng. 2018, 8, 169–180. [Google Scholar]
- Kang, J.; Zou, Z.-H. Time prediction model for pipeline leakage based on grey relational analysis. Phys. Procedia 2012, 25, 2019–2024. [Google Scholar]
- Yang, Y.-Y.; Han, Y.-H.; Zheng, J.-C.; Wang, J.-J.; Zhao, M.; Zhu, W. Risk Evaluation of urban water distribution network pipes using neural network. In Proceedings of the 4th ACM SICSPATIAL International Workshop on Safety and Resilience, Seattle, WA, USA, 6 November 2018. [Google Scholar] [CrossRef]
- Zhang, H.-W.; Wang, L.; Yue, L.; Lian, P. Study on time-prediction models for urban water supply network leakage. China Water Wastewater 2006, 5, 52–55. [Google Scholar]
- Wang, X.; Wang, Y.; Tian, W.-B. Time-prediction model for leakage in urban water supply network. J. Water Resour. Water Eng. 2012, 4, 151–154. [Google Scholar]
- Jin, S.; Tao, T. Comparison of the application of water supply network leakage forecast model. Water Technol. 2015, 6, 35–38. [Google Scholar]
- Wang, J.; Wang, Y.; Qin, Z.-F. Prediction of the leakage of urban water supply networks by wavelet neural network. Comput. Digit. Eng. 2016, 45, 1357–1360. [Google Scholar]
- Zhang, H.-W.; Niu, Z.-G.; Chen, C.; Hong, X. Study on the prediction model for water supply net leakage. China Water Wastewater 2001, 17, 7–10. [Google Scholar]
- He, F.; Liu, S.-Q. Analysis on pipe break in water distribution system and its countermeasures. Pipeline Tech. Equip. 2004, 5, 20–23. [Google Scholar]
- Shamir, U.; Charles, D.; Howard, D. An analytic approach to scheduling pipe replacement. J. Am. Water Works Assoc. 1979, 71, 248–258. [Google Scholar] [CrossRef]
- Ma, X.-W.; Xia, L.; Cheng, L. Urban water supply pipeline renewal time prediction model. J. Shenyang Jianzhu Univ. (Nat. Sci.) 2008, 1, 129–131. [Google Scholar]
- Zhou, C.; Xin, K.-L.; Tao, T.; Yin, Z.-L. Risk assessment and visualization of water supply pipe burst. City Town Water Supply 2015, 1, 68–70. [Google Scholar]
- Ke, Q.; Zhou, C.; Wang, L.-S.; Tao, T. Burst risk assessment model for water supply networks. Water Wastewater Eng. 2016, 7, 114–118. [Google Scholar]
- Zeng, H.; Ke, Q.; Zhou, C.; Tao, T. Dynamic risk assessment of water pipes burst in water supply distribution network. Water Purif. Technol. 2018, 2, 94–99. [Google Scholar]
- Chang, T.; Liu, S.-M.; Wang, M.; Li, M.-M.; Wu, X. Health condition assessment of urban water supply network based on BP neural network. Water Wastewater Eng. 2016, 6, 138–141. [Google Scholar]
- Kumar, A.; Asad Rizvi, A.A.; Brooks, B.; Vanderveld, R.A.; Wilson, K.H.; Kenney, C.; Edelstein, S.; Finch, A.; Maxwe, A.; Zuckerbraun, J.; et al. Using machine learning to assess the risk of and prevent water main breaks. In Proceedings of the ACM SIGKDD (SIGKDD’18), London, UK, 19–23 August 2018. [Google Scholar]
- Li, F.; Tao, T. Assessment and control of water leakage from water supply system. China Water Wastewater 2012, 18, 35–39. [Google Scholar]
- Hu, Y.-C.; Xing, D.-L.; Dong, S. A new digital city management based on the adaptive spatial information multi-grid technology. In Proceedings of the Asia-Pacific Computer Science and Application Conference, Shanghai, China, 27–28 December 2014. [Google Scholar]
- Ai, Z.-Y.; Yang, M. Extended mindlin solution of horizontal force at a point in the interior of a layered soil. J. Tongji Univ. (Nat. Sci.) 2000, 28, 272–276. [Google Scholar]
- Zhang, P.; Han, X. Study on control standards of deformation and damage of underground pipelines under metro construction. In Proceedings of the 2nd National Conference on Engineering Safety and Protection, Beijing, China, 10 August 2010. [Google Scholar]
- Yang, Y.-Y.; Han, Y.-H.; Zheng, J.-C.; Yu, F.-C. Study on the performance evaluation model of underground pipeline using big data. In Tech. Rep.; 2017; pp. 35–48. Available online: https://mis.kw.beijing.gov.cn/bstrs/ (accessed on 17 February 2020).
- Biganzoli, E.; Boracchi, P.; Mariani, L.; Marubini, E. Feed forward neural networks for the analysis of censored survival data: A partial logistic regression approach. Stat. Med. 1998, 17, 1169–1186. [Google Scholar] [CrossRef]
- Binganzoli, E.; Borrachi, P.; Marubini, E. A general framework for neural network models on censored survival data. Neural Netw. 2002, 15, 209–218. [Google Scholar] [CrossRef]
- Hothorn, T.; Zeileis, A. partykit: A modular toolkit for recursive partytioning in R. J. Mach. Learn. Res. 2015, 16, 3905–3909. [Google Scholar]
No. | Factors | Description | Classification Boundary |
---|---|---|---|
1 | Number of buildings | Number of buildings within 240 m of the accident site | 0–104, 105–133, 134–168, 169–208, ≥ 209 |
2 | Total area of buildings | Total area of buildings within 240 m of the accident site, m2 | 0–53, 329, 53, 330–97, 121, 97, 122–155, 373, 155, 374–253, 347,≥ 253, 348 |
3 | Mean area of buildings | Mean area of buildings within 240 m of the accident site, m2 | 0–512, 513–753, 754–1100, 1101–1591, ≥1592 |
4 | Number of high-rise buildings | Number of high-rise buildings (buildings with ten or more floors, Code for Design of Civil Building) | 0–1, 2–5, 6–7, 8–17, ≥18 |
5 | Number of super-high-rise buildings | The number of super-high-rise buildings (higher than 100 m, Code for Design of Civil Building) surrounding the accident site | 0, 1, 2–3, ≥4 |
6 | Mean pressure of buildings | Mean pressure contributed by all surrounding buildings, kN | 0–25, 213, 25, 214–60, 763, 60, 764–93, 101, ≥ 93, 102 |
7 | Displacement | Displacement caused by surrounding buildings, estimated by the aforementioned deformation analysis model | 0–100, 101–323, ≥ 324 |
8 | Regional construction year | The construction year of the area where the accident occurred | ≤1987, 1988–1993, 1994–2005, ≥2006 |
9 | Pipeline lifetime | The interval between the regional construction year and the year of the accident, years | ≤20, 20–30, ≥30 |
10 | Month of accident | Month of the accident | 1–4, 5–6, 7–8, 9–10, 11–12 |
11 | Road width | The width of the road close to the accident site, m | 0–6, 7–9, 10–18, 19–21, 22–30, ≥31 |
12 | Category of accident | The category was defined by the gridding urban management (GUM) system, including 539 leakages and 320 bursts | Leakage/burst |
No. | Input | Target | ||
---|---|---|---|---|
X1 | X2 | Time Interval | ||
1 | 1 | 2 | 1 | 0 |
2 | 1 | 2 | 2 | 0 |
3 | 1 | 2 | 3 | 0 |
4 | 1 | 2 | 4 | 1 |
Cut-Off Value | Test Data | Training Data | ||
---|---|---|---|---|
ARR | Percentage of Positives | ARR | Percentage of Positives | |
0.019 | 19.2 | 0.41% | 24 | 0.23% |
© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Yang, Y.; Hu, Y.; Zheng, J. A Decision Tree Approach to the Risk Evaluation of Urban Water Distribution Network Pipes. Safety 2020, 6, 36. https://doi.org/10.3390/safety6030036
Yang Y, Hu Y, Zheng J. A Decision Tree Approach to the Risk Evaluation of Urban Water Distribution Network Pipes. Safety. 2020; 6(3):36. https://doi.org/10.3390/safety6030036
Chicago/Turabian StyleYang, Yanying, Yu Hu, and Jianchun Zheng. 2020. "A Decision Tree Approach to the Risk Evaluation of Urban Water Distribution Network Pipes" Safety 6, no. 3: 36. https://doi.org/10.3390/safety6030036
APA StyleYang, Y., Hu, Y., & Zheng, J. (2020). A Decision Tree Approach to the Risk Evaluation of Urban Water Distribution Network Pipes. Safety, 6(3), 36. https://doi.org/10.3390/safety6030036