Computer-oriented method of optimizing hydrologic model parameters

作者: Donn G. Decoursey , Willard M. Snyder

DOI: 10.1016/0022-1694(69)90013-4

关键词: Reduction (complexity)Conceptual modelFunction (mathematics)ResidualConvergence (routing)Computer sciencePrincipal component analysisMathematical optimizationStatisticsHydrological modellingMathematical model

摘要: Abstract Much hydrologic information and significant advancement in understanding are based upon interpretive analysis of observations. The rational beginning point such an inductive process is the construction a conceptual model. Realistic models not normally simple forms. quantification verification realistic concepts requires specialized optimizing methods. Optimum values coefficients mathematical can be found by iterative reduction residual error associated with estimates coefficient values. Orthogonal corrections to insured using principal component regress term on group correction terms. A digital computer program capable performing optimization has been tested four varying complexity used several fieldscale models. It was that optimized function number roots solution, convergence observations, feedback, random noise, or equation complexity. While method discussed here developed primarily response needs modelling it should applicable any disciplines inputs data analysis.

参考文章(4)
Maurice G. Kendall, A course in multivariate analysis ,(1957)
Willard M. Snyder, Some possibilities for multivariate analysis in hydrologic studies Journal of Geophysical Research. ,vol. 67, pp. 721- 729 ,(1962) , 10.1029/JZ067I002P00721
K. B. Haley, Numerical Methods and Computers Journal of the Operational Research Society. ,vol. 17, pp. 203- 203 ,(1966) , 10.1057/JORS.1966.33
T. Lewis, K. L. Nielsen, Methods in Numerical Analysis. Biometrika. ,vol. 44, pp. 299- ,(1957) , 10.2307/2333287