Enhancing stormwater network overflow prediction: investigation of ensemble learning models

• Autorzy: Boughandjioua Samira , Laouacheria Fares , Azizi Nabiha

Identyfikatory

DOI

10.1007/s11600-024-01407-2

• Języki publikacji: EN

Abstrakty

EN

This study addresses the critical issue of urban flooding caused by stormwater network overflow, necessitating unified and efficient management measures to handle increasing water volumes and the effects of climate change. The proposed approach aims to improve the precision and efficiency of overflow rate predictions by investigating advanced machine learning algorithms, specifically ensemble methods such as gradient boosting and random forest algorithms. The main contribution lies in introducing the SWN-ML approach, which integrates hydraulic simulations using MIKE + with machine learning to predict average overflow rates for various rainfall durations and return periods. Mike + model was calibrated for the only available observed data of water depth at the outlet point during the storm event of February 4, 2019. The datasets for model calibration used in ML models consisted of many input variables such as peak flow, max depth, length, slope, roughness, and diameter and average overflow rate as output variable. Experimental results show that these methods are effective under a variety of scenarios, with the ensemble methods consistently outperforming classical machine learning models. For example, the models exhibit similar performance metrics with an MSE of 0.023, RMSE of 0.15, and MAE of 0.101 for a 2-h rainfall duration and a 10-year return period. Correlation analysis further confirms the strong correlation between ensemble method predictions and MIKE + simulated models, with values ranging between 0.72 and 0.80, indicating their effectiveness in capturing stormwater network dynamics. These results validate the utility of ensemble learning models in predicting overflow rates in flood-prone urban areas. The study highlights the potential of ensemble learning models in forecasting overflow rates, offering valuable insights for the development of early warning systems and flood mitigation strategies.

Słowa kluczowe

EN

stormwater; overflow; flooding; machine learning; ensemble learning models; Mike + model

PL

woda deszczowa; przelew; powódź; uczenie maszynowe; modele uczenia zespołowego

• Wydawca: Instytut Geofizyki PAN ; Springer
• Czasopismo: Acta Geophysica
• Rocznik: 2025
• Tom: Vol. 73, no. 1
• Strony: 875--899
• Opis fizyczny: Bibliogr. 57 poz.

Twórcy

autor

Boughandjioua Samira

• Laboratory of Soils and Hydraulic, Faculty of Technology, Badji Mokhtar Annaba University, P.O. Box 12, 23000 Annaba, Algeria

• https://orcid.org/0000-0001-8559-8707

autor

Laouacheria Fares

• Laboratory of Soils and Hydraulic, Faculty of Technology, Badji Mokhtar Annaba University, P.O. Box 12, 23000 Annaba, Algeria

• https://orcid.org/0000-0003-4169-0864

autor

Azizi Nabiha

• LABGED Laboratory, Faculty of Technology, Badji Mokhtar Annaba University, P.O. Box 12, 23000 Annaba, Algeria

• https://orcid.org/0000-0002-2713-4792

Bibliografia

• 1. Abedi R, Costache R, Shafizadeh-Moghadam H et al (2022) Flash- flood susceptibility mapping based on XGBoost, random forest and boosted regression trees. GeoIn 37:5479-5496. https://doi.org/10.1080/10106049.2021.1920636
• 2. Afrin S, Islam MM, Rahman MM (2021) Adequacy assessment of an urban drainage system considering future land use and climate change scenario. J Water Clim Change 12:1944-1957. https://doi.org/10.2166/WCC.2020.369
• 3. Arabameri A, Saha S, Chen W et al (2020) Flash flood susceptibility modelling using functional tree and hybrid ensemble techniques. J Hydrol (amst) 587:125007. https://doi.org/10.1016/J.JHYDR OL.2020.125007
• 4. Arfiani A, Rustam Z (2019) Ovarian cancer data classification using bagging and random forest. In: AIP Conference Proceedings. American Institute of Physics Inc.
• 5. Aydin HE, Iban MC (2023) Predicting and analyzing flood susceptibility using boosting-based ensemble machine learning algorithms with SHapley Additive exPlanations. Nat Hazards 116:2957- 2991. https://doi.org/10.1007/S11069-022-05793-Y/METRICS
• 6. Band SS, Janizadeh S, Pal SC et al (2020) Flash flood susceptibility modeling using new approaches of hybrid and ensemble tree- based machine learning algorithms. Remote Sens 12:3568. https://doi.org/10.3390/RS12213568
• 7. Berkhahn S, Fuchs L, Neuweiler I (2019) An ensemble neural network model for real-time prediction of urban floods. J Hydrol (amst) 575:743-754. https://doi.org/10.1016/J.JHYDROL.2019.05.066
• 8. Berndtsson R, Becker P, Persson A et al (2019) Drivers of changing urban flood risk: a framework for action. J Environ Manage 240:47-56. https://doi.org/10.1016/J.JENVMAN.2019.03.094
• 9. Bin Z (2015) The application of SWMM model in the urban planning study on Sponge City. Earth Sci 4:205. https://doi.org/10.11648/J. EARTH.20150405.17
• 10. Bisht DS, Chatterjee C, Kalakoti S et al (2016) Modeling urban floods and drainage using SWMM and MIKE URBAN: a case study. Nat Hazards 84:749-776. https://doi.org/10.1007/s11069-016-2455-1
• 11. Breiman L (2001) Random forests. Mach Learn 45:5-32. https://doi. org/10.1023/A:1010933404324/METRICS
• 12. Brown G, Wyatt J, Harris R, Yao X (2005) Diversity creation methods: a survey and categorisation. Inf Fus 6:5-20. https://doi.org/10. 1016/J.INFFUS.2004.04.004
• 13. Chang YH, Tseng CW, Hsu HC (2023) Predicting the overflowing of urban personholes based on machine learning techniques. Water 15:4100. https://doi.org/10.3390/W15234100
• 14. Chaudhary P, Leitao JP, Schindler K, Wegner JD (2024) Flood water depth prediction with convolutional temporal attention networks. Water (switzerland). https://doi.org/10.3390/w16091286
• 15. Chen W, Zhang S, Li R, Shahabi H (2018) Performance evaluation of the GIS-based data mining techniques of best-first decision tree, random forest, and naive Bayes tree for landslide susceptibility modeling. Sci Total Environ 644:1006-1018. https://doi.org/10. 1016/J.SCITOTENV.2018.06.389
• 16. Chen X, Zhang H, Chen W, Huang G (2021) Urbanization and climate change impacts on future flood risk in the Pearl River Delta under shared socioeconomic pathways. Sci Total Environ. https://doi.org/10.1016/J.SCITOTENV.2020.143144
• 17. Cheriguene S, Azizi N, Dey N et al (2019) A new hybrid classifier selection model based on mRMR method and diversity measures. Int J Mach Learn Cybern 10:1189-1204. https://doi.org/10.1007/S13042-018-0797-6/METRICS
• 18. Chicco D, Warrens MJ, Jurman G (2021) The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation. PeerJ Comput Sci 7:1-24. https://doi.org/10.7717/PEERJ-CS.623
• 19. Dang TQ, Tran BH, Le QN et al (2024) Application of machine learning-based surrogate models for urban flood depth modeling in Ho Chi Minh City Vietnam. Appl Soft Comput 150:111031. https:// doi.org/10.1016/J.ASOC.2023.111031
• 20. DHI (2012). MIKE FLOOD user manual. https://manuals.mikepoweredbydhi.help/2021/Water_Resources/MIKE_FLOOD_UserM anual.pdf. Accessed 27 May 2024.
• 21. DHI (2016). MIKE URBAN CS - MOUSE, User guide. Danish Hydraulic Institute. https://manuals.mikepoweredbydhi.help/2017/Cities/CollectionSystem.pdf. Accessed on 27 May 2024.
• 22. DHI (2022) MIKE+ collection system. https://manuals.mikepoweredbydhi.help/2022/Cities/MIKE_Plus_Collection_System.pdf. Accessed on 27 May 2024.
• 23. DHI (2024). MIKE+ 2D Overland. https://manuals.mikepoweredby-dhi.help/latest/Cities/MIKE_Plus_2DOverland.pdf. Accessed 27 May 2024.
• 24. Dietterich TG (2000) Ensemble methods in machine learning. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). https://doi.org/10.1007/3-540-45014-9_1
• 25. Diogo AF, do Carmo JA (2019) Peak flows and stormwater networks design-current and future management of urban surface watersheds. Water (switzerland). https://doi.org/10.3390/w11040759
• 26. Friedman JH (2001) Greedy function approximation: a gradient boosting machine. Ann Stat 29:1189-1232. https://doi.org/10.1214/ AOS/1013203451
• 27. Game P, Wang M, Audra P, Gourbesville P (2023) Flood modelling for a real-time decision support system of the covered Lower Paillons River, Nice, France. J Hydroinf 25:1884-1908. https://doi.org/10.2166/HYDRO.2023.181/1303201/JH2023181.PDF
• 28. Garzón A, Kapelan Z, Langeveld J, Taormina R (2022) Machine learning-based surrogate modeling for urban water networks: review and future research directions. Water Resour Res 58:e2021WR031808. https://doi.org/10.1029/2021WR031808
• 29. GeeksforGeeks (2023). Cross validation in machine learning. https://www.geeksforgeeks.org/cross-validation-machine-learning/. Accessed on 27 May 2024.
• 30. Hafnaoui MA, Madi M, Hachemi A, Farhi Y (2020) El Bayadh city against flash floods: case study. Urb Water J 17:390-395. https://doi.org/10.1080/1573062X.2020.1714671
• 31. Hukkeri GS, Naganna SR, Pruthviraja D et al (2023) Drought forecasting: application of ensemble and advanced machine learning approaches. IEEE Access 11:141375-141393. https://doi.org/10.1109/ACCESS.2023.3341587
• 32. Kabir S, Patidar S, Xia X et al (2020) A deep convolutional neural network model for rapid prediction of fluvial flood inundation. J
• 33. Hydrol (amst) 590:125481. https://doi.org/10.1016/J.JHYDROL.2020.125481
• 34. Karim F, Armin MA, Ahmedt-Aristizabal D et al (2023) A review of hydrodynamic and machine learning approaches for flood inundation modeling. Water. https://doi.org/10.3390/w15030566
• 35. Kotsiantis S, Pintelas PE (2014) Combining bagging and boosting
• 36. Kuncheva LI, Whitaker CJ (2003) Measures of diversity in classifier ensembles and their relationship with the ensemble accuracy. Mach Learn 51:181-207. https://doi.Org/10.1023/A:1022859003006/METRICS
• 37. Laouacheria F, Mansouri R (2015) Comparison of WBNM and HEC- HMS for runoff hydrograph prediction in a small urban catchment. Water Resour Manage 29:2485-2501. https://doi.org/10.1007/s11269-015-0953-7
• 38. Laouacheria F, Kechida S, Chabi M (2019) Modelling the impact of design rainfall on the urban drainage system by storm water management model. J Water Land Dev 40:119-125. https://doi.org/10.2478/jwld-2019-0013
• 39. Liu X, Zhang X, Kong X, Shen YJ (2022) Random forest model has the potential for runoff simulation and attribution. Water 14:2053. https://doi.org/10.3390/W14132053
• 40. Lowe R, Bohm J, Jensen DG et al (2021) U-FLOOD—topographic deep learning for predicting urban pluvial flood water depth. J Hydrol (amst) 603:126898. https://doi.org/10.1016/J.JHYDROL.2021.126898
• 41. Lu M, Hou Q, Qin S et al (2023) A stacking ensemble model of various machine learning models for daily runoff forecasting. Water 15:1265. https://doi.org/10.3390/W15071265
• 42. Mukaka MM (2012) A guide to appropriate use of correlation coefficient in medical research. Malawi Med J 24:69
• 43. Natekin A, Knoll A (2013) Gradient boosting machines, a tutorial. Front Neurorobot 7:63623. https://doi.org/10.3389/FNBOT.2013.00021/BIBTEX
• 44. Rokach L (2010) Ensemble-based classifiers. Artif Intell Rev 33:1- 39. https://doi.org/10.1007/S10462-009-9124-7/METRICS
• 45. Salim F, Bhattacharyya S (2023) Ensemble learning bagging and boosting
• 46. Salvati A, Nia AM, Salajegheh A et al (2023) Flood susceptibility mapping using support vector regression and hyper-parameter optimization. J Flood Risk Manag. https://doi.org/10.1111/JFR3.12920
• 47. Snider B, Mcbean E (2018) Improving time-to-failure predictions for water distribution systems using gradient boosting algorithm
• 48. Wood-Ponce R, Diab G, Liu Z et al (2024) Developing data-driven learning models to predict urban stormwater runoff volume. Urban Water J. https://doi.org/10.1080/1573062X.2024.2312514
• 49. Xiong J, Li J, Cheng W et al (2019) A GIS-based support vector machine model for flash flood vulnerability assessment and mapping in China. ISPRS Int Geo-Inf 297(8):297. https://doi.org/10.3390/IJGI8070297
• 50. Ying Z, Tian L, Reuse SKL of PC and R, University T (2016) Simulation on lid measures for control of combined sewer overflows in existing urban areas. China Water & Wastewater 127-131
• 51. Zhang T, Lin W, Vogelmann AM et al (2021) Improving convection trigger functions in deep convective parameterization schemes using machine learning. J Adv Model Earth Syst. https://doi.org/10.1029/2020MS002365
• 52. Zhao G, Pang B, Xu Z et al (2019) Assessment of urban flood susceptibility using semi-supervised machine learning model. Sci Total Environ 659:940-949. https://doi.org/10.1016/J.SCITOTENV.2018.12.217
• 53. Zhao G, Pang B, Xu Z et al (2021) Improving urban flood susceptibility mapping using transfer learning. J Hydrol (amst) 602:126777. https://doi.org/10.1016/J.JHYDROL.2021.126777
• 54. Zhou ZH (2012) Ensemble methods: foundations and algorithms. https://doi.org/10.1201/b12207
• 55. Ziadi SR, Keraghel MA (2024) Flooding vulnerability in Algiers (Algeria): an analytic hierarchy process. Nat Hazards 120:6199- 6221. https://doi.org/10.1007/S11069-024-06471-X/METRICS
• 56. Zoppou C (2001) Review of urban storm water models. Environ Model Softw 16:195-231. https://doi.org/10.1016/S1364-8152(00)00084-0
• 57. Zounemat-Kermani M, Batelaan O, Fadaee M, Hinkelmann R (2021) Ensemble machine learning paradigms in hydrology: a review. J Hydrol (amst) 598:126266. https://doi.org/10.1016/J.JHYDROL.2021.126266

Uwagi

Opracowanie rekordu ze środków MNiSW, umowa nr POPUL/SP/0154/2024/02 w ramach programu "Społeczna odpowiedzialność nauki II" - moduł: Popularyzacja nauki (2025).