Non-linear canonical correlation analysis in regional frequency analysis

详细信息    查看全文

• 作者：D. Ouali ; F. Chebana ; T. B. M. J. Ouarda
• 关键词：Non ; linear canonical correlation analysis ; Neural network ; Regional frequency analysis ; Homogeneous region ; Hydrological neighborhood ; Ungauged basins
• 刊名：Stochastic Environmental Research and Risk Assessment (SERRA)
• 出版年：2016
• 出版时间：February 2016
• 年：2016
• 卷：30
• 期：2
• 页码：449-462
• 全文大小：1,702 KB
• 参考文献：Akaho S (2001) A kernel method for canonical correlation analysis. In: Proceedings of the International Meeting of Psychometric Society (IMPS). University Convention Center-Osaka, Japan
  Aziz K, Rahman A, Fang G, Shrestha S (2014) Application of artificial neural networks in regional flood frequency analysis: a case study for Australia. Stoch Environ Res Risk Assess 28(3):541–554CrossRef
  Ballard DH (1981) Generalizing the Hough transform to detect arbitrary shapes. Pattern Recognit 13(2):111–122CrossRef
  Barnett TP, Preinsendorfer R (1987) Origins and levels of monthly and seasonal forecast skill for United States surface air temperatures determined by canonical correlation analysis. Mon Weather Rev 115:1825–1850CrossRef
  Bishop CM (1995) Neural networks for pattern recognition. Oxford University Press, Clarendon
  Bolton RJ, Hand DJ, Webb AR (2003) Projection techniques for nonlinear principal component analysis. Stat Comput 13(3):267–276CrossRef
  Botelho SSDC, Bem RAD, ÍLD Almeida, Mata MM (2005) C-nlpca: extracting non-linear principal components of image datasets. Anais XII Simposio Brasileiro de Sensoriamento Remoto, Goiania, Brasil, pp 3495–3502
  Campi C, Parkkonen L, Hari R, Hyvärinen A (2013) Non-linear canonical correlation for joint analysis of MEG signals from two subjects. Front Neurosci 7:107CrossRef
  Cannon AJ (2008) Multivariate statistical models for seasonal climate prediction and climate downscaling. Atmospheric Science, University of British Columbia. Doctor of philosophy, p 141
  Chebana F, Ouarda T (2007) Multivariate L‐moment homogeneity test. Water Resour Res 43(8)
  Chebana F, Ouarda TBMJ (2008) Depth and homogeneity in regional flood frequency analysis. Water Resour Res 44(11)
  Chebana F, C Charron, TBMJ Ouarda, B Martel (2014) Regional frequency analysis at ungauged sites with the generalized additive model. In: press J Hydrometeorol
  Chen C-S, Liu C-H, Su H-C (2008) A nonlinear time series analysis using two-stage genetic algorithms for streamflow forecasting. Hydrol Process 22:3697–3711CrossRef
  Chen L, Singh VP, Guo S, Zhou J, Ye L (2014) Copula entropy coupled with artificial neural network for rainfall–runoff simulation. Stoch Environ Res Risk Assess 28(7):1755–1767CrossRef
  Chokmani K, Ouarda TBMJ (2004) Physiographical space-based kriging for regional flood frequency estimation at ungauged sites. Water Resour Res 40(12)
  Dauxois J, Nkiet GM (1998) Nonlinear canonical analysis and independence tests. Ann Stat 26(4):1254–1278CrossRef
  Dawson C, Wilby R (2001) Hydrological modelling using artificial neural networks. Prog Phys Geogr 25(1):80–108CrossRef
  Duch W, Jankowski N (1999) Survey of neural transfer functions. Neural Comput Surv 2(1):163–212
  Frie KG, Janssen C (2009) Social inequality, lifestyles and health–a non-linear canonical correlation analysis based on the approach of Pierre Bourdieu. Int J Public Health 54(4):213–221CrossRef
  Gifi A (1990) Nonlinear multivariate analysis. Wiley, Chichester, p 579
  Guillemette N, St-Hilaire A, Ouarda TB, Bergeron N, Robichaud É, Bilodeau L (2009) Feasibility study of a geostatistical modelling of monthly maximum stream temperatures in a multivariate space. J Hydrol 364(1):1–12CrossRef
  Hardoon DR, Shawe-Taylor J (2009) Convergence analysis of kernel canonical correlation analysis: theory and practice. Mach Learn 74:23–38CrossRef
  Hsieh WW (2000) Nonlinear canonical correlation analysis by neural networks. Neural Netw 13:1095–1105CrossRef
  Hsieh WW (2001) Nonlinear canonical correlation analysis of the tropical Pacific climate variability using a neural network approach. J Clim 14:2528–2539CrossRef
  Khalil B, Ouarda T, St-Hilaire A (2011) Estimation of water quality characteristics at ungauged sites using artificial neural networks and canonical correlation analysis. J Hydrol 405(3):277–287CrossRef
  Kramer MA (1991) Nonlinear principal component analysis using autoassociative neural networks. Am Inst Chem Eng J 37(2):233–243CrossRef
  Kruger U, Sharma SK, Irwin GW (2004) Improved nonlinear canonical correlation analysis using genetic strategies. UKACC control. University of Bath, Bath
  Li C, Tang H, Ge Y, Hu X, Wang L (2014) Application of back-propagation neural network on bank destruction forecasting for accumulative landslides in the three Gorges Reservoir Region, China. Stoch Environ Res Risk Assess 28(6):1465–1477CrossRef
  Malthouse EC (1998) Limitations of nonlinear PCA as performed with generic neural networks. IEEE Trans Neural Netw 9(1):165–173CrossRef
  Michael AG, Raymond CP (2003) Using traffic conviction correlates to identify high accident-risk drivers. Accid Anal Prev 35(6):903–912CrossRef
  Monahan AH (2000) Nonlinear principal component analysis by neural networks: theory and application to the Lorenz system. J Clim 13:821–835CrossRef
  Nagai I (2013) Optimization using cross-validation for penalized nonlinear canonical correlation analysis. Graduate School of Science and Technology, Kwansei Gakuin University 2-1 Gakuen, Sanda, pp 669–1337
  Ouarda TBMJ (2013) Hydrological frequency analysis, regional. Encycl Environ. doi:10.​1002/​9780470057339.​vnn9780470057043​
  Ouarda TBMJ, Shu C (2009) Regional low-flow frequency analysis using single and ensemble artificial neural networks. Water Resour Res 45(11)
  Ouarda TBMJ, Girard C, Cavadias GS, Bobée B (2001) Regional flood frequency estimation with canonical correlation analysis. J Hydrol 254(1–4):157–173CrossRef
  Ouarda T, Bâ K, Diaz-Delgado C, Cârsteanu A, Chokmani K, Gingras H, Quentin E, Trujillo E, Bobée B (2008a) Intercomparison of regional flood frequency estimation methods at ungauged sites for a Mexican case study. J Hydrol 348(1):40–58CrossRef
  Ouarda TBMJ, St-Hilaire A, Bobée B (2008b) Synthèse des développements récents en analyse régionale des extrêmes hydrologiques. Revue des sciences de l’eau 21(2):219–232CrossRef
  Pandey G, Nguyen V-T-V (1999) A comparative study of regression based methods in regional flood frequency analysis. J Hydrol 225(1):92–101CrossRef
  Riad S, Mania J (2004) Rainfall-runoff model using an artificial neural network approach. Math Comput Model 40:839–846CrossRef
  Rumelhart DE, GE Hinton, RJ Williams (1985) Learning internal representations by error propagation. In: Rumelhart DE, McClelland JL, P. R. Group, 1: 318–362
  Sengupta S, Boyle J (1995) Non-linear principal component analysis of climate data. PCMDI
  Shu C, Ouarda TBMJ (2007) Flood frequency analysis at ungauged sites using artificial neural networks in canonical correlation analysis physiographic space. Water Resour Res 43(07)
  Tasker GD, Hodge SA, Barks CS (1996) Region of influence regression for estimating the 50-year flood at ungauged sites. Water Resour Res 1(32):163–170
  Tishlert A, Lipovetsky S (1996) Canonical correlation analyses for three data sets: a unified framework with application to management. Comput Oper Res 23(7):667–679CrossRef
  Van Den Wollenberg AL (1977) Redundancy analysis an alternative for canonical correlation analysis. Psychometrika 42(2):207–219CrossRef
  Wang D, Shi L, Yeung DS, Tsang E (2005) Nonlinear canonical correlation analysis of fMRI signals using HDR models. 27th Annual international conference of the IEEE engineering in medicine and biology society, 2005. IEEE-EMBS 2005
  Wazneh H, Chebana F, Ouarda T (2013) Optimal depth-based regional frequency analysis. Hydrol Earth Syst Sci 17(6):2281–2296CrossRef
  Wu A, Hsieh WW (2002) Nonlinear canonical correlation analysis of the tropical Pacific wind stress and sea surface temperature. Clim Dyn 19:713–722CrossRef
  Xu J, Li W, Ji M, Lu F, Dong S (2010) A comprehensive approach to characterization of the nonlinearity of runoff in the headwaters of the Tarim River, Western China. Hydrol Process 24:136–146CrossRef
  Yin H (2007) Nonlinear dimensionality reduction and data visualization: a review. Int J Autom Comput 4(3):294–303CrossRef
  Yonaba H, Anctil F, Fortin V (2010) Comparing sigmoid transfer functions for neural network multistep ahead streamflow forecasting. J Hydrol Eng 15(4):275–283CrossRef
  Zaier I, Shu C, Ouarda T, Seidou O, Chebana F (2010) Estimation of ice thickness on lakes using artificial neural network ensembles. J Hydrol 383(3):330–340CrossRef
  Zhihua J, Zhen Y (2010) On using non-linear canonical correlation analysis for voice conversion based on Gaussian mixture model. J Electron 27(1):1–7
  
• 作者单位：D. Ouali (1)
  F. Chebana (1)
  T. B. M. J. Ouarda (1) (2)
  
  1. Institut National de la Recherche Scientifique, Centre Eau Terre et Environnement, 490, rue de la Couronne, Quebec, QC, G1K 9A9, Canada
  2. Institute Centre for Water Advanced Technology and Environmental Research, P.O. Box 54224, Abu Dhabi, UAE
  
• 刊物类别：Earth and Environmental Science
• 刊物主题：Environment
  Mathematical Applications in Environmental Science
  Mathematical Applications in Geosciences
  Probability Theory and Stochastic Processes
  Statistics for Engineering, Physics, Computer Science, Chemistry and Geosciences
  Numerical and Computational Methods in Engineering
  Waste Water Technology, Water Pollution Control, Water Management and Aquatic Pollution
  
• 出版者：Springer Berlin / Heidelberg
• ISSN：1436-3259

文摘

Hydrological processes are complex non-linear phenomena. Canonical correlation analysis (CCA) is frequently used in regional frequency analysis (RFA) to delineate hydrological neighborhoods. Although non-linear CCA (NL-CCA) is widely used in several fields, it has not been used in hydrology, particularly in RFA. This paper presents an overview of techniques used to reproduce non-linear relationships between two sets of variables. The approaches considered in this work are based on NL-CCA using neural networks (CCA-NN), coupled to a log-linear regression model for flood quantile estimation. In order to demonstrate the usefulness of these approaches in RFA, a comparative study between the latter and linear CCA is performed using three different databases from North America. Results show that CCA-NN is more robust and can better reproduce the non-linear relationship structures between physiographical and hydrological variables. This reflects the high flexibility of this approach. Results indicate that for all three databases, it is more advantageous to proceed with the non-linear CCA approach. Keywords Non-linear canonical correlation analysis Neural network Regional frequency analysis Homogeneous region Hydrological neighborhood Ungauged basins