Module transfer method based on multiscale modeling

Abstract

The invention discloses a module transfer method based on multiscale modeling. The method comprises the following steps: collecting original spectra of a host instrument and a slave instrument; by combining the characteristic of wavelet basis and the characteristic of the original spectrum of a sample, selecting the optimal wavelet basis to carry out wavelet multiscale decomposition on the spectrum to obtain wavelet coefficient; reconstructing the wavelet coefficient; respectively performing multivariate calibration on each layer of reconstructed spectrum; setting up a prediction model based on partial least squares method and leaving one cross-validation method for the spectrum after performing multivariate calibration to obtain a cross-validation root-mean-square error of the prediction model; performing model fusion on the prediction model by using weight number, and calculating the prediction root-mean-square error and related coefficient to evaluate the model transfer effect. Compared with a conventional model transfer method, the method greatly improves the model transfer efficiency and performance, and can be widely applied to fields of near-infrared and Raman spectrum.

Description

technical field [0001] The invention relates to the field of chemometrics, in particular to a model transfer method based on multi-scale modeling. Background technique [0002] The multivariate calibration model is the basis of quantitative analysis of near-infrared spectroscopy. Establishing a long-term stable, reliable and accurate mathematical model is a very complicated process that consumes a lot of time, manpower, material and financial resources. Since the model contains information such as the physical and chemical properties of the modeling sample, sample loading conditions, instrument status, and environmental factors, it is necessary to evaluate and study whether the model can be applied to new samples or spectra measured on new instruments. Not applicable, that is to say, the sample information encountered by the model during the prediction process is not the variation included in the calibration process, resulting in large deviations in the prediction results, s...

AI Technical Summary

Problems solved by technology

However, the conventional multivariate correction method is to correct the entire spectrum without dividing the spectrum in the frequency domain. Such a model transfer is very general, not fine enough, the accuracy is not enough, and the performance is naturally not good.

Images