Using Machine Learning to Identify and Optimize Sensitive Parameters in Urban Flood Model Considering Subsurface Characteristics

Hengxu Jin; Yu Zhao; Pengcheng Lu; Shuliang Zhang; Yiwen Chen; Shanghua Zheng; Zhizhou Zhu

doi:10.1007/s13753-024-00540-2

Volume 15 Issue 1

Feb. 2024

Turn off MathJax

Article Contents

Article Navigation > International Journal of Disaster Risk Science > 2024 > 15(1): 116-133

Hengxu Jin, Yu Zhao, Pengcheng Lu, Shuliang Zhang, Yiwen Chen, Shanghua Zheng, Zhizhou Zhu. Using Machine Learning to Identify and Optimize Sensitive Parameters in Urban Flood Model Considering Subsurface Characteristics[J]. International Journal of Disaster Risk Science, 2024, 15(1): 116-133. doi: 10.1007/s13753-024-00540-2

Citation:

Hengxu Jin, Yu Zhao, Pengcheng Lu, Shuliang Zhang, Yiwen Chen, Shanghua Zheng, Zhizhou Zhu. Using Machine Learning to Identify and Optimize Sensitive Parameters in Urban Flood Model Considering Subsurface Characteristics[J]. International Journal of Disaster Risk Science, 2024, 15(1): 116-133. doi: 10.1007/s13753-024-00540-2

Citation:

Hengxu Jin, Yu Zhao, Pengcheng Lu, Shuliang Zhang, Yiwen Chen, Shanghua Zheng, Zhizhou Zhu. Using Machine Learning to Identify and Optimize Sensitive Parameters in Urban Flood Model Considering Subsurface Characteristics[J]. International Journal of Disaster Risk Science, 2024, 15(1): 116-133. doi: 10.1007/s13753-024-00540-2

PDF

Using Machine Learning to Identify and Optimize Sensitive Parameters in Urban Flood Model Considering Subsurface Characteristics

doi: 10.1007/s13753-024-00540-2

Hengxu Jin^1,2,
Yu Zhao^1,2,
Pengcheng Lu^1,2,
Shuliang Zhang^{1,2
,
,},
Yiwen Chen^1,2,
Shanghua Zheng^1,2,
Zhizhou Zhu^1,2

1. Key Laboratory of VGE of Ministry of Education, Nanjing Normal University, Nanjing, 210023, China;
2. Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application, Nanjing, 210023, China

Funds:

This study was supported by the National Natural Science Foundation of China (Grant Nos. 42271483, 42071364) and the Postgraduate Research &

amp

Practice Innovation Program of Jiangsu Province (Grant No. KYCX23_1696). We would like to thank Editage for English language editing.

Accepted Date: 2024-01-21

Available Online: 2024-03-14

Publish Date: 2024-02-21

Abstract

Abstract

This study presents a novel method for optimizing parameters in urban flood models, aiming to address the tedious and complex issues associated with parameter optimization. First, a coupled one-dimensional pipe network runoff model and a two-dimensional surface runoff model were integrated to construct an interpretable urban flood model. Next, a principle for dividing urban hydrological response units was introduced, incorporating surface attribute features. The K-means algorithm was used to explore the clustering patterns of the uncertain parameters in the model, and an artificial neural network (ANN) was employed to identify the sensitive parameters. Finally, a genetic algorithm (GA) was used to calibrate the parameter thresholds of the sub-catchment units in different urban land-use zones within the flood model. The results demonstrate that the parameter optimization method based on K-means-ANN-GA achieved an average Nash-Sutcliffe efficiency coefficient (NSE) of 0.81. Compared to the ANN-GA and K-means-deep neural networks (DNN) methods, the proposed method better characterizes the runoff generation and flow processes. This study demonstrates the significant potential of combining machine learning techniques with physical knowledge in parameter optimization research for flood models.
- Artificial neural network,
- Coupled urban flooding model,
- Genetic algorithm,
- K-means algorithm,
- Sub-catchment delineation,
- Uncertain parameters

FullText(HTML)

References(56)

References

[1]	Adnan, R.M., A. Petroselli, S. Heddam, C.A.G. Santos, and O. Kisi. 2021. Short term rainfall-runoff modelling using several machine learning methods and a conceptual event-based model. Stochastic Environmental Research and Risk Assessment 35(3): 597–616.
[2]	Cai, T., X. Li, X. Ding, J. Wang, and J. Zhan. 2019. Flood risk assessment based on hydrodynamic model and fuzzy comprehensive evaluation with GIS technique. International Journal of Disaster Risk Reduction 35: Article 101077.
[3]	Chang, T.-J., H.-L. Yu, C.-H. Wang, and A.S. Chen. 2021. Overland-gully-sewer (2D-1D-1D) urban inundation modeling based on cellular automata framework. Journal of Hydrology 603(B): Article 127001.
[4]	Chen, W., H. Wu, J.S. Kimball, L. Alfieri, N. Nanding, X. Li, L. Jiang, W. Wu, et al. 2022. A coupled river basin-urban hydrological model (DRIVE-Urban) for real-time urban flood modeling. Water Resources Research 58(11): Article e2021WR031709.
[5]	Dao, D.A., D. Kim, and D.H.H. Tran. 2022. Estimation of rainfall threshold for flood warning for small urban watersheds based on the 1D–2D drainage model simulation. Stochastic Environmental Research and Risk Assessment 36(3): 735–752.
[6]	Feigl, M., S. Thober, R. Schweppe, M. Herrnegger, L. Samaniego, and K. Schulz. 2022a. Automatic regionalization of model parameters for hydrological models. Water Resources Research 58(12): Article e2022WR031966.
[7]	Feigl, M., S. Thober, R. Schweppe, M. Herrnegger, L. Samaniego, and K. Schulz. 2022b. Automatic regionalization of model parameters for hydrological models. Water Resources Research 58(12): Article e2022WR031966.
[8]	Guo, J., and X. Su. 2019. Parameter sensitivity analysis of SWAT model for streamflow simulation with multisource precipitation datasets. Hydrology Research 50(3): 861–877.
[9]	Hu, C., X. Zhang, C. Li, C. Liu, J. Wang, and S. Jian. 2022. Real-time flood classification forecasting based on K-means++ clustering and neural network. Water Resources Management 36(1): 103–117.
[10]	Hu, J., Q. Zhou, A. McKeand, T. Xie, and S.-K. Choi. 2020. A model validation framework based on parameter calibration under aleatory and epistemic uncertainty. Structural and Multidisciplinary Optimization 63: 645–660.
[11]	Huo, J., and L. Liu. 2019. Application research of multi-objective artificial bee colony optimization algorithm for parameters calibration of hydrological model. Neural Computing and Applications 31(9): 4715–4732.
[12]	Ichiba, A., A. Gires, I. Tchiguirinskaia, D. Schertzer, P. Bompard, and M.-C.T. Veldhuis. 2018. Scale effect challenges in urban hydrology highlighted with a distributed hydrological model. Hydrology and Earth System Sciences 22(1): 331–350.
[13]	Jung, D., Y. Choi, and J. Kim. 2017. Multiobjective automatic parameter calibration of a hydrological model. Water 9(3): Article 187.
[14]	Katipoğlu, O.M., and M. Sarıgöl. 2023. Prediction of flood routing results in the Central Anatolian region of Türkiye with various machine learning models. Stochastic Environmental Research and Risk Assessment 37(6): 2205–2224.
[15]	Kim, S.W., S.H. Kwon, and D. Jung. 2022. Development of a multiobjective automatic parameter-calibration framework for urban drainage systems. Sustainability 14(14): Article 8350.
[16]	Li, J. 2020. A data-driven improved fuzzy logic control optimization-simulation tool for reducing flooding volume at downstream urban drainage systems. Science of The Total Environment 732: Article 138931.
[17]	Li, D., H. Li, Y. Zhang, M. Guo, and D. Zhang. 2022. Influence of time step synchronization on urban rainfall-runoff simulation in a hybrid CPU/GPU 1D–2D coupled model. Water Resources Management 36(10): 3417–3433.
[18]	Li, Y., J. Zhou, R. Jiang, and M. Zhou. 2016. Classified correction of real-time flood forecasting based on K-Cluster analysis. China Rural Water and Hydropower 12(160–162): 167 (in Chinese).
[19]	Liao, R., Z. Xu, C. Ye, X. Shu, and D. Yang. 2023. Simulation of rainstorm waterlogging in Dahongmen drainage area based on SWMM and InfoWorks ICM models. Water Resources Protection 39(3): 109–117.
[20]	Liao, R., Z. Xu, C. Ye, B. Zuo, D. Xiang, and X. Shu. 2022. Parameter sensitivity analysis methods of storm water management model. Journal of Hydroelectric Engineering 41(6): 11–21.
[21]	Liu, K., W. Bao, J. Que, J. Li, and H. Shu. 2015. Application of K-mean cluster method to flood forecasting based on principal component analysis. Engineering Journal of Wuhan University 48(4): 447–450; 458 (in Chinese).
[22]	Liu, C., Z. Han, X. Li, Y. Sun, Y. Tang, D. Hou, and C. Hu. 2023. Urban stormwater rapid simulation based on BIC-Kmeans and SWMM. Water Resources Protection 39(5): 79–87.
[23]	Lu, P., and Y. Sun. 2021. Scenario-based hydrodynamic simulation of adaptive strategies for urban design to improve flood resilience: A case study of the Mingzhu Bay Region, Guangzhou, Greater Bay Area. River Research and Applications 39(7): 1425–1436.
[24]	Nandi, S., and M.J. Reddy. 2022. An integrated approach to streamflow estimation and flood inundation mapping using VIC, RAPID and LISFLOOD-FP. Journal of Hydrology 610: Article 127842.
[25]	Nguyen, D.H., and D.-H. Bae. 2020. Correcting mean areal precipitation forecasts to improve urban flooding predictions by using long short-term memory network. Journal of Hydrology 584: Article 124710.
[26]	O'Loughlin, F.E., J. Neal, G.J.P. Schumann, E. Beighley, and P.D. Bates. 2020. A LISFLOOD-FP hydraulic model of the middle reach of the Congo. Journal of Hydrology 580: Article 124203.
[27]	Padiyedath Gopalan, S., A. Kawamura, H. Amaguchi, T. Takasaki, and G. Azhikodan. 2019. A bootstrap approach for the parameter uncertainty of an urban-specific rainfall-runoff model. Journal of Hydrology 579: Article 124195.
[28]	Peng, L., Z. Xu, G. Zhao, B. Zuo, J. Wang, and S. Song. 2021. Simulation of urban rainstorm waterlogging processes based on SWMM and LISFLOOD-FP models: Case study in Jinan City. South-to-North Water Transfers and Water Science & Technology 19(6): 1083–1092.
[29]	Rai, P.K., B.R. Chahar, and C.T. Dhanya. 2016. GIS-based SWMM model for simulating the catchment response to flood events. Hydrology Research 48(2): 384–394.
[30]	Schilling, J., and J. Tränckner. 2022. Generate_SWMM_inp: An open-source QGIS plugin to import and export model input files for SWMM. Water 14(14): 2262.
[31]	Shahed Behrouz, M., Z. Zhu, L.S. Matott, and A.J. Rabideau. 2020. A new tool for automatic calibration of the storm water management model (SWMM). Journal of Hydrology 581: Article 124436.
[32]	Shen, H., Z. Xu, S. Zhang, and Q. Li. 2019. Analysis and test on rainfall control modes of LID facilities with different confluence types. Journal of Hydraulic Engineering 50(5): 578–588.
[33]	Shustikova, I., J.C. Neal, A. Domeneghetti, P.D. Bates, S. Vorogushyn, and A. Castellarin. 2020. Levee breaching: A new extension to the LISFLOOD-FP model. Water 12(4): Article 942.
[34]	Sinnathamby, S., K.R. Douglas-Mankin, and C. Craige. 2017. Field-scale calibration of crop-yield parameters in the soil and water assessment tool (SWAT). Agricultural Water Management 180(Part A): 61–69.
[35]	Snieder, E., and U.T. Khan. 2023. A novel ensemble algorithm based on hydrological event diversity for urban rainfall–runoff model calibration and validation. Journal of Hydrology 619: Article 129193.
[36]	Sosa, J., C. Sampson, A. Smith, J. Neal, and P. Bates. 2019. A toolbox to quickly prepare flood inundation models for LISFLOOD-FP simulations. Environmental Modelling & Software 123: Article 104561.
[37]	Song, X., Q. Shu, H. Wang, and W. Liao. 2009. Comparison and application of SCE-UA, genetic algorithm and simplex method. Engineering Journal of Wuhan University 42(1): 6–9; 15 (in Chinese).
[38]	Sun, Y., C. Liu, X. Du, F. Yang, Y. Yao, S. Soomro, and C. Hu. 2022. Urban storm flood simulation using improved SWMM based on K-means clustering of parameter samples. Journal of Flood Risk Management 15(4): Article e12826.
[39]	Wang, F., X. Qing, S. Yang, and Z. Cui. 2022. Automatic calibration of SWMM parameters based on PySWMM. China Water & Wastewater 38(21): 124–130.
[40]	Wang, Y., S. Zhu, L. Yuan, and R. Deng. 2020. An automatic parameter calibration method for the SWAT model in runoff simulation. River Research and Applications 36(7): 1321–1333.
[41]	Willis, T., N. Wright, and A. Sleigh. 2019. Systematic analysis of uncertainty in 2D flood inundation models. Environmental Modelling & Software 122: Article 104520.
[42]	Wood, M., R. Hostache, J. Neal, T. Wagener, L. Giustarini, M. Chini, G. Corato, P. Matgen, and P. Bates. 2016. Calibration of channel depth and friction parameters in the LISFLOOD-FP hydraulic model using medium-resolution SAR data and identifiability techniques. Hydrology and Earth System Sciences 20(12): Article 4983–4997.
[43]	Wu, Z., B. Ma, H. Wang, C. Hu, H. Lv, and X. Zhang. 2021. Identification of sensitive parameters of urban flood model based on artificial neural network. Water Resources Management 35(7): 2115–2128.
[44]	Xu, W., and Y. Peng. 2015. Research on classified real-time flood forecasting framework based on K-means cluster and rough set. Water Science and Technology 71(10): 1507–1515.
[45]	Yan, X., K. Xu, W. Feng, and J. Chen. 2021. A rapid prediction model of urban flood inundation in a high-risk area coupling machine learning and numerical simulation approaches. International Journal of Disaster Risk Science 12(6): 903–918.
[46]	Yang, Q., Z. Ma, and S. Zhang. 2022. Urban pluvial flood modeling by coupling raster-based two-dimensional hydrodynamic model and SWMM. Water 14(11): Article 1760.
[47]	Yang, Y., L. Sun, R. Li, J. Yin, and D. Yu. 2020. Linking a storm water management model to a novel two-dimensional model for urban pluvial flood modeling. International Journal of Disaster Risk Science 11(4): 508–518.
[48]	Ye, C., Z. Xu, X. Lei, Y. Chen, C. Ban, and H. Su. 2021. Rapid simulations of storm water runoff in urban community scale. Case study of a community compound in Fuzhou City. Journal of Hydroelectric Engineering 40(10): 81–94.
[49]	Ye, C., Z. Xu, X. Lei, W. Liao, X. Ding, and Y. Liang. 2022. Assessment of urban flood risk based on data-driven models: A case study in Fuzhou City, China. International Journal of Disaster Risk Reduction 82: Article 103318.
[50]	Yuan, S., D. Li, Y. Chen, Z. He, Q. Cheng, and F. Liu. 2021. Automatic calibration procedure of storm water management model parameters based on back propagation neural network algorithm. China Water & Wastewater 37(21): 125–130.
[51]	Zang, Y., Y. Meng, X. Guan, H. Lv, and D. Yan. 2022. Study on urban flood early warning system considering flood loss. International Journal of Disaster Risk Reduction 77: Article 103042.
[52]	Zeng, J., Y. Mai, Z. Li, X. Ren, J. Pan, and G. Huang. 2020. Sensitivity analysis of SWMM parameters in Guangzhou Tianhe wisdom city. Water Resources Protection 36(3): 15–21.
[53]	Zeng, Z., Z. Wang, and C. Lai. 2022. Simulation performance evaluation and uncertainty analysis on a coupled inundation model combining SWMM and WCA2D. International Journal of Disaster Risk Science 13(3): 448–464.
[54]	Zeng, Z., Z. Wang, X. Wu, C. Lai, and X. Chen. 2017. Rainstorm waterlogging simulations based on SWMM and LISFLOOD models. Journal of Hydroelectric Engineering 36(5): 68–77.
[55]	Zhang, M., X. Zhang, T. Wang, Y. Xu, and Q. Zhang. 2022. Method of determining hydrological response units for discretization of microtopography. Research of Soil and Water Conservation 29(1): 86–91; 99 (in Chinese).
[56]	Zhu, M., F. Wang, T. Zhang, X. Yin, W. Wu, and H. Zhu. 2018. Research and practice of standard system of city surveying and mapping geographic information—taking Nanjing city as an example. Science of Surveying and Mapping 43(12): 151–157.