Skip directly to search Skip directly to A to Z list Skip directly to page options Skip directly to site content

NIOSHTIC-2 Publications Search

Search Results

Expert system for interpretation of the infrared spectra of environmental mixtures.

Authors
Ying-L; Levine-SP; Tomellini-SA
Source
Computer-Enhanced Analytical Spectroscopy. Meuzelaar HLC, ed., NY: Plenum Press, 1990 Feb; 2:245-257
Link
NIOSHTIC No.
00228764
Abstract
An expert system for interpreting infrared spectra of environmental mixtures was discussed. The system (intIRpret) was a successor to a previously developed computer assisted technique for automatically identifying the components of hazardous waste mixtures based on their infrared spectra (PAWMI), and was designed to overcome the limitations of PAWMI. To use PAWMI, rules for deducing the components of a mixture had to be written manually and the PAWMI identification software provided only peak location information. IntIRpret consisted of five major subroutines: the interferogram processing and peak selection subroutine (PUSHSUB), the automated knowledge acquisition subroutine (AUTOGEN), the system optimization subroutine (STO), the interpretation subroutine (PAIRS), and the final processing subroutine to subtract spectral similarity (PAIRSPLUS). The subroutines were designed to make intIRpret automatic, self training, and self optimizing. The functions of PUSHSUB, AUTOGEN, STO, PAIRS, and PAIRSPLUS were discussed. STO was discussed in detail and illustrated using the infrared spectrum of benzene (71432) as an example. The ability of intIRpret and PAWMI to analyze 67 four component mixtures of 62 compounds commonly found at hazardous waste sites was compared. Both systems used the infrared spectra of 62 compounds as a training set. IntIRpret showed a 40% decrease in false positive results and a 24% decrease in false negative results compared to PAWMI. There were problems such as interferences and high noise levels when applying intIRpret to mixtures consisting of traces of vapors and gases in ambient air.
Keywords
NIOSH-Grant; Grants-other; Spectrographic-analysis; Infrared-spectrophotometry; Computer-models; Mathematical-models; Hazardous-materials; Waste-disposal; Organic-vapors
Contact
Environmental & Indust Health School of Public Health II 1420 Washington Heights Ann Arbor, MI 48109-2029
CAS No.
71-43-2
Publication Date
19900228
Document Type
Book or book chapter
Editors
Meuzelaar-HLC
Funding Amount
238744
Funding Type
Grant
Fiscal Year
1990
NTIS Accession No.
NTIS Price
ISBN No.
9780306432767
Identifying No.
Grant-Number-R01-OH-02404' Grant-Number-R01-OH-02066
Priority Area
Other Occupational Concerns; Grants-other
Source Name
Computer-Enhanced Analytical Spectroscopy
State
MI
Performing Organization
University of Michigan at Ann Arbor, Ann Arbor, Michigan
TOP