Multiscale Latent Variable Regression
Multiscale wavelet-based representation of data has been shown to be a powerful tool in feature extraction from practical process data. In this paper, this characteristic of multiscale representation is utilized to improve the prediction accuracy of some of the latent variable regression models, suc...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2010-01-01
|
| Series: | International Journal of Chemical Engineering |
| Online Access: | http://dx.doi.org/10.1155/2010/935315 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832556299650334720 |
|---|---|
| author | Mohamed N. Nounou Hazem N. Nounou |
| author_facet | Mohamed N. Nounou Hazem N. Nounou |
| author_sort | Mohamed N. Nounou |
| collection | DOAJ |
| description | Multiscale wavelet-based representation of data has been shown to be a powerful tool in feature extraction from practical process data. In this paper, this characteristic of multiscale representation is utilized to improve the prediction accuracy of some of the latent variable regression models, such as Principal Component Regression (PCR) and Partial Least Squares (PLS), by developing a multiscale latent variable regression (MSLVR) modeling algorithm. The idea is to decompose the input-output data at multiple scales using wavelet and scaling functions, construct multiple latent variable regression models at multiple scales using the scaled signal approximations of the data and then using cross-validation, and select among all MSLVR models the model which best describes the process. The main advantage of the MSLVR modeling algorithm is that it inherently accounts for the presence of measurement noise in the data by the application of the low-pass filters used in multiscale decomposition, which in turn improves the model robustness to measurement noise and enhances its prediction accuracy. The advantages of the developed MSLVR modeling algorithm are demonstrated using a simulated inferential model which predicts the distillate composition from measurements of some of the trays' temperatures. |
| format | Article |
| id | doaj-art-d19f97f0aa33491fb31a842c6a6ff2d4 |
| institution | Kabale University |
| issn | 1687-806X 1687-8078 |
| language | English |
| publishDate | 2010-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | International Journal of Chemical Engineering |
| spelling | doaj-art-d19f97f0aa33491fb31a842c6a6ff2d42025-02-03T05:45:49ZengWileyInternational Journal of Chemical Engineering1687-806X1687-80782010-01-01201010.1155/2010/935315935315Multiscale Latent Variable RegressionMohamed N. Nounou0Hazem N. Nounou1Chemical Engineering Program, Texas A&M University at Qatar, P.O. Box 23874, Doha, QatarElectrical and Computer Engineering Program, Texas A&M University at Qatar, P.O. Box 23874, Doha, QatarMultiscale wavelet-based representation of data has been shown to be a powerful tool in feature extraction from practical process data. In this paper, this characteristic of multiscale representation is utilized to improve the prediction accuracy of some of the latent variable regression models, such as Principal Component Regression (PCR) and Partial Least Squares (PLS), by developing a multiscale latent variable regression (MSLVR) modeling algorithm. The idea is to decompose the input-output data at multiple scales using wavelet and scaling functions, construct multiple latent variable regression models at multiple scales using the scaled signal approximations of the data and then using cross-validation, and select among all MSLVR models the model which best describes the process. The main advantage of the MSLVR modeling algorithm is that it inherently accounts for the presence of measurement noise in the data by the application of the low-pass filters used in multiscale decomposition, which in turn improves the model robustness to measurement noise and enhances its prediction accuracy. The advantages of the developed MSLVR modeling algorithm are demonstrated using a simulated inferential model which predicts the distillate composition from measurements of some of the trays' temperatures.http://dx.doi.org/10.1155/2010/935315 |
| spellingShingle | Mohamed N. Nounou Hazem N. Nounou Multiscale Latent Variable Regression International Journal of Chemical Engineering |
| title | Multiscale Latent Variable Regression |
| title_full | Multiscale Latent Variable Regression |
| title_fullStr | Multiscale Latent Variable Regression |
| title_full_unstemmed | Multiscale Latent Variable Regression |
| title_short | Multiscale Latent Variable Regression |
| title_sort | multiscale latent variable regression |
| url | http://dx.doi.org/10.1155/2010/935315 |
| work_keys_str_mv | AT mohamednnounou multiscalelatentvariableregression AT hazemnnounou multiscalelatentvariableregression |