Compression of chemical process data by functional approximation and feature extraction

Bhavik R. Bakshi, George Stephanopoulos

Research output: Contribution to journalArticle

44 Citations (Scopus)

Abstract

Effective utilization of measured process data requires efficient techniques for their compact storage and retrieval, as well as for extracting information on the process operation. Techniques for the on-line compression of process data were developed based on their contribution in time and in frequency using the theory of wavelets. Existing techniques for compression via wavelets and wavelet packets are inconvenient for on-line compression and are best suited for stationary signals. These methods were extended to the on-line decomposition and compression of nonstationary signals via time-varying wavelet packets. Various criteria for the selection of the best time-varying wavelet packet coefficients are derived. Explicit relationships among the compression ratio, local and global errors of approximation, and features in the signal were derived and used for efficient compression. Extensive case studies on industrial data demonstrate the superior performance of wavelet-based techniques as compared to existing piecewise linear techniques.

Original languageEnglish (US)
Pages (from-to)477-492
Number of pages16
JournalAICHE Journal
Volume42
Issue number2
DOIs
StatePublished - Jan 1 1996
Externally publishedYes

Fingerprint

Piecewise linear techniques
Chemical Phenomena
Feature extraction
Compaction
Decomposition
Data Compression
Patient Selection

ASJC Scopus subject areas

  • Biotechnology
  • Environmental Engineering
  • Chemical Engineering(all)

Cite this

Compression of chemical process data by functional approximation and feature extraction. / Bakshi, Bhavik R.; Stephanopoulos, George.

In: AICHE Journal, Vol. 42, No. 2, 01.01.1996, p. 477-492.

Research output: Contribution to journalArticle

@article{a802a7c2868e4c208146ebf2fca1f9e8,
title = "Compression of chemical process data by functional approximation and feature extraction",
abstract = "Effective utilization of measured process data requires efficient techniques for their compact storage and retrieval, as well as for extracting information on the process operation. Techniques for the on-line compression of process data were developed based on their contribution in time and in frequency using the theory of wavelets. Existing techniques for compression via wavelets and wavelet packets are inconvenient for on-line compression and are best suited for stationary signals. These methods were extended to the on-line decomposition and compression of nonstationary signals via time-varying wavelet packets. Various criteria for the selection of the best time-varying wavelet packet coefficients are derived. Explicit relationships among the compression ratio, local and global errors of approximation, and features in the signal were derived and used for efficient compression. Extensive case studies on industrial data demonstrate the superior performance of wavelet-based techniques as compared to existing piecewise linear techniques.",
author = "Bakshi, {Bhavik R.} and George Stephanopoulos",
year = "1996",
month = "1",
day = "1",
doi = "10.1002/aic.690420217",
language = "English (US)",
volume = "42",
pages = "477--492",
journal = "AICHE Journal",
issn = "0001-1541",
publisher = "American Institute of Chemical Engineers",
number = "2",

}

TY - JOUR

T1 - Compression of chemical process data by functional approximation and feature extraction

AU - Bakshi, Bhavik R.

AU - Stephanopoulos, George

PY - 1996/1/1

Y1 - 1996/1/1

N2 - Effective utilization of measured process data requires efficient techniques for their compact storage and retrieval, as well as for extracting information on the process operation. Techniques for the on-line compression of process data were developed based on their contribution in time and in frequency using the theory of wavelets. Existing techniques for compression via wavelets and wavelet packets are inconvenient for on-line compression and are best suited for stationary signals. These methods were extended to the on-line decomposition and compression of nonstationary signals via time-varying wavelet packets. Various criteria for the selection of the best time-varying wavelet packet coefficients are derived. Explicit relationships among the compression ratio, local and global errors of approximation, and features in the signal were derived and used for efficient compression. Extensive case studies on industrial data demonstrate the superior performance of wavelet-based techniques as compared to existing piecewise linear techniques.

AB - Effective utilization of measured process data requires efficient techniques for their compact storage and retrieval, as well as for extracting information on the process operation. Techniques for the on-line compression of process data were developed based on their contribution in time and in frequency using the theory of wavelets. Existing techniques for compression via wavelets and wavelet packets are inconvenient for on-line compression and are best suited for stationary signals. These methods were extended to the on-line decomposition and compression of nonstationary signals via time-varying wavelet packets. Various criteria for the selection of the best time-varying wavelet packet coefficients are derived. Explicit relationships among the compression ratio, local and global errors of approximation, and features in the signal were derived and used for efficient compression. Extensive case studies on industrial data demonstrate the superior performance of wavelet-based techniques as compared to existing piecewise linear techniques.

UR - http://www.scopus.com/inward/record.url?scp=0030085768&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030085768&partnerID=8YFLogxK

U2 - 10.1002/aic.690420217

DO - 10.1002/aic.690420217

M3 - Article

VL - 42

SP - 477

EP - 492

JO - AICHE Journal

JF - AICHE Journal

SN - 0001-1541

IS - 2

ER -