Automated infrared detection of organophosphorus compounds in multicomponent solutions. Master`s thesis
In-situ monitoring of the environment in peacetime and wartime is required of today`s technology with particular concern toward organophosphorus pesticides and chemical warfare agents. Of the many tools available for monitoring, infrared spectrometry combined with fiber optic sensors is one method of remote sensing which shows promise. This thesis models the problem of automatically detecting organophosphorus chemicals in water and automatically specifying what chemical is present. Computer generated infrared spectra are modeled as well as sensor characteristics and Computer generated infrared spectra are modeled as well as sensor characteristics and instrument noise over various binary mixture concentrations. A standard pattern recognition data reduction technique known as Principle Components Analysis is employed to compute meaningful features for detection of target chemicals in water. Methods of detection of the target chemical include the k-nearest neighbor classifier and Parzen window/maximum a-posteriori probability decision rule. By systematically varying chemical amounts, the order of magnitude of the limit of detection is determined. To determine the detection error bound beyond a simple point estimate of the error, Bayes error bounding is implemented. Once the lower range of detection is determined, k-nearest neighbor tests are done to determine classification of specific target chemicals.
- Research Organization:
- Air Force Inst. of Tech., Wright-Patterson AFB, OH (United States). School of Engineering
- OSTI ID:
- 251180
- Report Number(s):
- AD-A-303295/0/XAB; TRN: 61490445
- Resource Relation:
- Other Information: TH: Master`s thesis; PBD: Dec 1995
- Country of Publication:
- United States
- Language:
- English
Similar Records
Interrogating Proton Affinities of Organophosphonate Species Via Atmospheric Flow Tube Mass Spectrometry and Computational Methods
Localized keyhole pore prediction during laser powder bed fusion via multimodal process monitoring and X-ray radiography