Spectral data analysis method and spectral data analysis apparatus

Abstract

The embodiments of the invention disclose a spectral data analysis method and a spectral data analysis apparatus. The spectral data analysis method comprises: establishing a sample database, screeningthe neighboring spectral groups of a to-be-detected sample, carrying out dimension reduction on the neighboring spectral groups through cumulative contributing rate of principal component, determining target neighboring spectral groups according a Mahalanobis distance ratio, predicting the to-be-detected sample by selecting the target neighboring spectral group with the smallest relative prediction error as a training set sample, determining whether the representativeness of the sample database is sufficient, and adding the to-be-detected sample to the sample database if the representativeness of the sample database is insufficient. According to the present invention, the sample database is established by using the molecular spectroscopy technology, the optimal target neighboring spectralgroup is automatically selected from the sample database according to the spectral characteristics of the to-be-detected sample, the physical and chemical data of the to-be-detected sample is furtherpredicted, and the representativeness of the sample database is further determined, such that the automation of the spectral data analysis process and database update is achieved, the prediction accuracy is improved, the complicated work of manual establishing and model maintaining is eliminated, and the molecular spectroscopy technology is easily promoted and applied.

Description

technical field [0001] The embodiments of the present invention relate to the technical field of data analysis, and in particular to a spectral data analysis method and device. Background technique [0002] Molecular spectroscopic analysis technology is a technology that uses the interaction between material molecules and electromagnetic waves to conduct qualitative or quantitative analysis of substances. Molecular spectroscopic analysis technology has the advantages of convenience, rapidity, high efficiency, accuracy, environmental friendliness and low cost, and has been widely used in agriculture, petrochemical industry, clinical diagnosis, environmental detection and other fields. [0003] The application of this technology also faces some serious challenges. Since the detection results of molecular spectroscopy are predicted based on a multivariate calibration model established with a large amount of data, the accuracy of the prediction can only be guaranteed if the cha...

AI Technical Summary

Problems solved by technology

With the change of factors such as time, place, raw materials, environment, etc., the number of samples beyond the scope of application of the model will continue to increase, and the original model needs to be continuously expanded and updated. However, the work of spectral data analysis requires quite professional knowledge background and experience. It is difficult for most users to achieve this condition, which has become a major bottleneck in the promotion and application of molecular spectroscopy technology

Images