Font Size: a A A

Application Of Wavelet Transform And Modern Optimization Algorithm In The Modeling Of Near Infrared Spectroscopy

Posted on:2010-01-08Degree:MasterType:Thesis
Country:ChinaCandidate:F Y HuFull Text:PDF
GTID:2178360278950682Subject:Pulp and paper engineering
Abstract/Summary:PDF Full Text Request
Near infrared spectroscopy is an efficient, rapid, low-cost, non-invasive and not destroying environment. It can be used for rapid detection and analysis of real-time online,and it has a very wide application prospect in the production of pulp and paper industry. However, due to near infrared absorption spectra of the mechanism of their own reasons, the absorption signal to noise ratio is relatively low, resulting in a lower sensitivity of the analysis. A number of conventional approaches have been used, but to establish the accuracy of the model still needs to be improved.As a new signal processing tool, wavelet transform have a good nature of time-frequency localization, especially in signal de-noising and compression. Genetic algorithm is one of the most widely used and the most successful algorithm in intelligent optimization algorithm, which has been successfully applied in many fields.In this thesis, wavelet transform and genetic algorithm are applied to near infrared spectroscopy pre-processing. Models were used the wavelet transform soft threshold and the several wavelet coefficients. The modeling results show that wavelet transform is simple and the effect of modeling is superior to conventional methods. Wavelet transform and genetic algorithm are combined for modeling, and as a result, used the GA, the region are selected, and reduces of the modeling spectral data saves modeling time and improves the accuracy of the model.
Keywords/Search Tags:near infrared spectroscopy, wavelet transform, wavelet de-noising, genetic algorithm, region selection, partial least square
PDF Full Text Request
Related items