An Angled Multivariate Analysis Method for Eliminating Spectral Multiplicative Random Errors
A multiplicative random error, multivariate analysis technology, applied in the measurement of color/spectral characteristics, etc., can solve the problems of accurate quantitative interference, multiplicative error, infeasible EMSC method, etc., to avoid interference and eliminate multiplicative errors. Effect
Inactive Publication Date: 2016-08-17
GUANGXI UNIVERSITY OF TECHNOLOGY
View PDF6 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Multiplicative errors lead to changes in signal amplitude. Under the premise of equal variance distribution, standard normal variable correction (SNV) and multiple scatter correction (MSC) methods are usually used to correct; in the case of different signal variances, extended multiple scatter can be used. Correction (EMSC) method correction, but requires prior knowledge support
[0006] In most cases, the signal variance varies with the intensity, which cannot meet the constraints of SNV and MSC, and often due to the lack of sufficient or accurate prior knowledge, the EMSC method is not feasible
In real-time and on-site measurements of spectra, intensity disturbances and particle scattering disturbances are common, resulting in multiplicative errors that directly interfere with quantitative accuracy
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0084] Select the Raman spectrometer to measure the carbon tetrachloride content in the mixture as an example.
[0085] 1) Spectral signal acquisition
[0086] Add carbon tetrachloride to the multi-component mixture according to the content of 2.5% to 25%, and add carbon tetrachloride at ten equal intervals to prepare 5 groups of 50 series samples, each of which contains 10 samples with a content of 2.5% to 25%. Concentrations of other components in the sample are not relevant.
[0087] By determining the measurement distance and integration time of each sample, the multiplicative disturbance in the measurement signal is formed, and the Raman spectrum signals of a series of samples are collected as Figure 4 shown.
[0088] 2) Direct PLS modeling
[0089] From 50 samples, a total of 15 samples with 3 concentrations were randomly selected from each group as the modeling calibration set, and partial least squares modeling was performed directly. The number of hidden variable...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses an angular multi-variable analysis method for eliminating a spectral multiplicative random error. The method comprises the steps of modeling and prediction. A direct processing method based on a spectral intensity signal is replaced with an angular multi-variable conversion method, so that a multiplicative error of the signal is eliminated, and the influence of disturbance on the signal intensity of a system is avoided.
Description
Technical field: [0001] The invention relates to an angled multivariate analysis method for eliminating spectral multiplicative random errors. Background technique: [0002] In spectral analysis, the acquired spectral signal is disturbed by various random errors, especially the reflection measurement method commonly used in Raman and near-infrared. Interference with multiplicative errors. [0003] In systems where there is a linear relationship between the measurand (x) and the response signal (y), additive (e a ) and multiplicative (e m ) The two types of errors can be expressed as: [0004] y=e m ·a·x+e a (1) [0005] Additive errors cause overall signal fluctuations and deflection, but do not affect the signal's amplitude, and are better removed by background subtraction or differential (difference) methods. Multiplicative errors lead to changes in signal amplitude. Under the premise of equal variance distribution, standard normal variable correction (SNV) and mul...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G01N21/31
Inventor 姚志湘粟晖
Owner GUANGXI UNIVERSITY OF TECHNOLOGY