论文部分内容阅读
针对黄河下游复杂水位过程的有效拟合,将多项式回归、逐步回归、参数的岭估计等有机集成,并引进累进变换,系统形成了一个新方法.新方法与同类方法的核心区别,在于考虑了复杂水位过程中常见的弱影响因素间的强耦合作用;在于综合采用了多种方法最大限度地降低拟合模型误差;在于引进了必要的累进变换.新方法有机集成了多个理论和方法的长处,且使用方便.新方法有着充分的理论基础,有着同类复杂拟合问题下的普遍适用性.进一步用该方法拟合多组黄河下游复杂水位过程,拟合精度都较现有其他方法提高显著,拟合模型各项都有明确的物理意义.
Aiming at the effective fitting process of complex water level in the lower Yellow River, a new method is formed by the organic integration of polynomial regression, stepwise regression, ridge parameter estimation, etc., and the introduction of progressive transformation.The core difference between the new method and the similar method is that The strong coupling between the weak influence factors commonly encountered in complex water level lies in the comprehensive use of a variety of methods to minimize the error of the fitting model and the introduction of the necessary progressive transformation.The new method integrates multiple theories and methods Advantages and easy to use.The new method has a sufficient theoretical basis and has universal applicability under the complex fitting problem of the same kind.Furthermore, this method fits multiple sets of complicated water level processes in the lower Yellow River so that the fitting accuracy is improved compared with other existing methods Significantly, each of the fitted models has a clear physical meaning.