Application of neural network-based corrected intelligent residual error gray models to load forecast;

Application of Residual Error Gray Forecast Model on Logistics Demand Forecast;

The basic principles and methods of conventional residual error Gray Model are presented.

Based on the residue theorem of Interferometric SAR,in depth analysis and detailed description of Branch Cut algorithm,one of the most popular InSAR phase unwrapping approaches,has been given in this paper.

Based on analyzing the characteristics of residue,the Letts criterion method was used to an identify an outlier.

The results show the built model has lower residual,higher accuracy and better feasibility and suitability,which can be used as decision reference for the management and planning.

A novel calibration algorithm, PCRRANN (principal component regression residual artificial neural network) method, was proposed based on the intrinsic non linearity of the prediction of gasoline octane number, and then applied to the calibration of the prediction model of the near infra red measurement of gasoline octane number.

The signal was divided into three parts: beat template, residual and position parameter.

First we define the residual estimates based on the nonparametric estimator gn(x) for g(x),then we construct the estimator f^n(x) for f(x) on the basis of the residuals.

Based on residual sensitivity m atrix and signs of residuals,this paper presents a new method to determine the possibility of existence of the worst residual contamination.

Based on the tracking residuals of object,the theory residuals self-covariance was gained,and the supposed testing value of residuals was gained,which utilized the statistical residuals self-covariance.