Hourly Flood Forecasting Using Hybrid Wavelet-SVM

Document Type : Regular Article

Authors

1 Post Graduate Student, National Institute of Technology, Calicut, India

2 Assistant Professor, Department of Civil Engineering, National Institute of Technology Calicut, India

3 Professor, Department of Civil Engineering, National Institute of Technology Calicut, India

Abstract

The floods of 2018 and 2019 have underlined the urgent need for development and implementation of efficient and robust flood forecasting models for the major rivers in the State of Kerala, India. In this paper, the development and application of two hourly flood forecasting models are presented – one using Support Vector Machine (SVM) and the other based on hybrid wavelet-support vector machine (WSVM). The study was performed on the Achankovil River in Kerala. Wavelet technique was used to denoise the input signal (rainfall and water level) and the effective components of the input signal obtained after denoising were input to the SVM/ WSVM models for forecasting. These models' performance was assessed using standard performance rating criteria. Further, the performance of these models was compared with that of a flood forecasting model based on hybrid wavelet-artificial neural network (WANN) developed for this river in a previous study. Results of this study demonstrated the ability of the WSVM model to predict floods reasonably well. It was observed that the WSVM model performed better when compared to the WANN model. The WSVM model was able to accurately estimate peak discharge magnitude and time to peak, both of which are critical inputs in many water resource design and management applications.

Keywords

Main Subjects

References

[1]     Jain SK, Mani P, Jain SK, Prakash P, Singh VP, Tullos D, et al. A Brief review of flood forecasting techniques and their applications. Int J River Basin Manag 2018;16:329–44. https://doi.org/10.1080/15715124.2017.1411920.
[2]     Adnan RM, Petroselli A, Heddam S, Santos CAG, Kisi O. Comparison of different methodologies for rainfall–runoff modeling: machine learning vs conceptual approach. Nat Hazards 2021;105:2987–3011. https://doi.org/10.1007/s11069-020-04438-2.
[3]     Amini A, Abdollahi A, Hariri-Ardebili MA, Lall U. Copula-based reliability and sensitivity analysis of aging dams: Adaptive Kriging and polynomial chaos Kriging methods. Appl Soft Comput 2021;109:107524. https://doi.org/10.1016/j.asoc.2021.107524.
[4]     Behzad M, Asghari K, Eazi M, Palhang M. Generalization performance of support vector machines and neural networks in runoff modeling. Expert Syst Appl 2009;36:7624–9. https://doi.org/10.1016/j.eswa.2008.09.053.
[5]     Komasi M, Sharghi S. Hybrid wavelet-support vector machine approach for modelling rainfall-runoff process. Water Sci Technol 2016;73:1937–53. https://doi.org/10.2166/wst.2016.048.
[6]     Zaker Esteghamati M, Flint MM. Developing data-driven surrogate models for holistic performance-based assessment of mid-rise RC frame buildings at early design. Eng Struct 2021;245:112971. https://doi.org/10.1016/j.engstruct.2021.112971.
[7]     Rhif M, Abbes A Ben, Farah IR, Martínez B, Sang Y. Wavelet transform application for/in non-stationary time-series analysis: A review. Appl Sci 2019;9:1–22. https://doi.org/10.3390/app9071345.
[8]     Wei CC. Wavelet support vector machines for forecasting precipitation in tropical cyclones: Comparisons with GSVM, regression, and MM5. Weather Forecast 2012;27:438–50. https://doi.org/10.1175/WAF-D-11-00004.1.
[9]     Adamowski JF. Development of a short-term river flood forecasting method for snowmelt driven floods based on wavelet and cross-wavelet analysis. J Hydrol 2008;353:247–66. https://doi.org/10.1016/j.jhydrol.2008.02.013.
[10]   Alexander AA, Thampi SG, Chithra NR. Development of hybrid wavelet-ANN model for hourly flood stage forecasting. ISH J Hydraul Eng 2018;24:266–74. https://doi.org/10.1080/09715010.2017.1422192.
[11]   Han D, Chan L, Zhu N. Flood forecasting using support vector machines. J Hydroinformatics 2007;9:267–76. https://doi.org/10.2166/hydro.2007.027.
[12]   Vapnik VN. The nature of statistical learning theory. vol. 37. 1st ed. 1995.
[13]   Liu Z, Zuo MJ, Zhao X, Xu H. An analytical approach to fast parameter selection of gaussian RBF kernel for support vector machine. J Inf Sci Eng 2015;31:691–710.
[14]   Seo Y, Kim S, Singh VP. Multistep-ahead flood forecasting using wavelet and data-driven methods. KSCE J Civ Eng 2015;19:401–17. https://doi.org/10.1007/s12205-015-1483-9.
[15]   Sang Y-F, Singh VP, Sun F, Chen Y, Liu Y, Yang M. Wavelet-Based Hydrological Time Series Forecasting. J Hydrol Eng 2016;21:06016001. https://doi.org/10.1061/(asce)he.1943-5584.0001347.
[16]   Oommen T, Coffman R, Sajinkumar KS, Vishnu CL. GEOTECHNICAL IMPACTS OF AUGUST 2018 FLOODS OF KERALA, INDIA Event: August 2018 Geotechnical Extreme Events Reconnaissance Turning Disaster into Knowledge Sponsored by the National Science Foundation GEER Association Report NO-058 2018:10–7. https://doi.org/10.18118/G6ZH3K.
[17]   Central Water Commission Government of India 2019. National Register of Large Dams -2019 2019:300.
[18]   Maier HR, Dandy GC. Neural networks for the prediction and forecasting of water resources variables: A review of modelling issues and applications. Environ Model Softw 2000;15:101–24. https://doi.org/10.1016/S1364-8152(99)00007-9.
[19]   Sarkar A, Kumar R. Artificial Neural Networks for Event Based Rainfall-Runoff Modeling. J Water Resour Prot 2012;04:891–7. https://doi.org/10.4236/jwarp.2012.410105.
[20]   Maheswaran R, Khosa R. Comparative study of different wavelets for hydrologic forecasting. Comput Geosci 2012;46:284–95. https://doi.org/10.1016/j.cageo.2011.12.015.
[21]   Nourani V, Komasi M, Mano A. A multivariate ANN-wavelet approach for rainfall-runoff modeling. Water Resour Manag 2009;23:2877–94. https://doi.org/10.1007/s11269-009-9414-5.
[22]   Wang W, Ding J. Wavelet Network Model and Its Application to the Prediction of Hydrology. Nat Sci 2003;1:67–71.
[23]   Lei L, Wang C, Liu X. Discrete Wavelet Transform Decomposition Level Determination Exploiting Sparseness Measurement. Int J Electr Comput Energ Electron Commun Eng 2013;7:691–4.
[24]   He C, Xing J, Li J, Yang Q, Wang R. A New Wavelet Threshold Determination Method Considering Interscale Correlation in Signal Denoising. Math Probl Eng 2015;2015. https://doi.org/10.1155/2015/280251.
[25]   Zhou T, Wang F, Yang Z. Comparative analysis of ANN and SVM models combined with wavelet preprocess for groundwater depth prediction. Water (Switzerland) 2017;9. https://doi.org/10.3390/w9100781.

Files

Cited by

History

  • Receive Date: 01 December 2021
  • Revise Date: 02 March 2022
  • Accept Date: 04 April 2022

Share

How to cite

Statistics