Susan L. Rose-Pehrsson

A comparison study of chemical sensor array pattern recognition algorithms

Probabilistic neural networks (PNN), learning vector quantization (LVQ) neural networks, back-propagation artificial neural networks (BP-ANN), soft independent modeling of class analogy (SIMCA), Bayesian linear discriminant analysis (BLDA), Mahalanobis linear discriminant analysis (MLDA), and the nearest-neighbor (NN) pattern recognition algorithms are compared for their ability to classify chemical sensor array data. Comparisons are made based on five qualitative criteria (speed, training difficulty, memory requirements, robustness to outliers, and the ability to produce a measure of uncertainty) and one quantitative criterion (classification accuracy). Four sample data sets from our laboratory, involving simulated data and polymer-coated surface acoustic wave chemical sensor array data, are used to estimate classification accuracies for each method. Among the seven algorithms in this study and the four data sets, the neural network based algorithms (LVQ, PNN, and BP-ANN) have the highest classification accuracies. When considering the qualitative criteria, the LVQ and PNN approaches fare well compared to BP-ANN due to their simpler training methods. The PNN is recommended for applications where a confidence measure and fast training are critical, while speed and memory requirements are not. LVQ is suggested for all other applications of chemical sensor array pattern recognition.

A comparison of the performance of statistical and fuzzy algorithms for unexploded ordnance detection

We focus on the development of signal processing algorithms that incorporate the underlying physics characteristic of the sensor and of the anticipated unexploded ordnance (UXO) target, in order to address the false alarm issue. In this paper, we describe several algorithms for discriminating targets from clutter that have been applied to data obtained with the multisensor towed array detection system (MTADS). This sensor suite includes both electromagnetic induction (EMI) and magnetometer sensors. We describe four signal processing techniques: a generalized likelihood ratio technique, a maximum likelihood estimation-based clustering algorithm, a probabilistic neural network, and a subtractive fuzzy clustering technique. These algorithms have been applied to the data measured by MTADS in a magnetically clean test pit and at a field demonstration. The results indicate that the application of advanced signal processing algorithms could provide up to a factor of two reduction in false alarm probability for the UXO detection problem.

Multi-criteria fire detection systems using a probabilistic neural network

Abstract The Navy program, Damage Control Automation for Reduced Manning (DC-ARM), is focused on enhancing automation of ship functions and damage control systems. A key element to this objective is the improvement of current fire detection systems. As in many applications, it is desired to increase detection sensitivity and, more importantly, increase the reliability of the detection system through improved nuisance alarm immunity. Improved reliability is needed such that fire detection systems can automatically control fire suppression systems. The use of multi-criteria-based detection technology continues to offer the most promising means to achieve both improved sensitivity to real fires and reduced susceptibility to nuisance alarm sources. A multi-signature early warning fire detection system is being developed to provide reliable warning of actual fire conditions in less time with fewer nuisance alarms than can be achieved with commercially available smoke detection systems. In this study, a large database consisting of the responses of 20 sensors to several different types of fires and nuisance sources was generated and analyzed using a variety of multivariate methods. Three data matrices were developed at discrete times corresponding to the different alarm levels of a conventional photoelectric smoke detector. The alarm times represent 0.82%, 1.63% and 11% obscurations per meter. The datasets were organized into three classes representing the sensor responses for baseline (nonfire), fires and nuisance sources. A robust data analysis strategy for use with a sensor array consisting of four to five sensors for early fire detection and nuisance source rejection was developed using a probabilistic neural network (PNN) that was developed at the Naval Research Laboratory for chemical sensor arrays. The analysis algorithms described in this paper evaluate discrete samples and develop classification models that examine individual chemical signatures at discrete points.

Sensitive, fluorescent detection of hydrazine via derivatization with 2,3-naphthalene dicarboxaldehyde

Abstract The derivatization of hydrazine with 2,3-naphthalene dicarboxaldehyde (NDA) followed by fluorescence detection is shown to be an effective method for the selective determination of extremely low concentrations of hydrazine in solution. Kinetic studies of the reaction between hydrazine and NDA indicate that the reaction is first order in both hydrazine and NDA, and is dependent upon an acidic pH (optimum pH = 2.5) for the formation of the fluorescent derivative. The NDA-hydrazine derivative is readily formed (time response ex = 403 nm and λ em = 500 nm). A linear concentration dependence is observed over a dynamic range from 50 ng/l to 500 μg/l of hydrazine (correlation coefficient, r > 0.999), with a signal-to-noise ratio of 3:1 for 50 ng/l of hydrazine. The selectively of the NDA reagent is examined with respect to several possible interferents, including ammonia, and is shown to be a highly specific reagent for hydrazine.

Early warning fire detection system using a Probabilistic Neural Network

The Navy program, Damage Control-Automation for Reduced Manning is focused on enhancing automation of ship functions and damage control systems. A key element to this objective is the improvement of current fire detection systems. As in many applications, it is desired to increase detection sensitivity and,more importantly increase the reliability of the detection system through improved nuisance alarm immunity. Improved reliability is needed, such that fire detection systems can automatically control fire suppression systems. The use of multi-criteria based detection technology continues to offer the most promising means to achieve both improved sensitivity to real fires,and reduced susceptibility to nuisance alarm sources. A multi-criteria early warning fire detection system, has been developed to provide reliable warning of actual fire conditions, in less time, with fewer nuisance alarms,than can be achieved with commercially available smoke detection systems. In this study a four-sensor array and a Probabilistic Neural Network have been used to produce an early warning fire detection system. A prototype early warning fire detector was built and tested in a shipboard environment. The current alarm algorithm resulted in better overall performance than the commercial smoke detectors, by providing both improved nuisance source immunity with generally equivalent or faster response times.

Using physics-based modeler outputs to train probabilistic neural networks for unexploded ordnance (UXO) classification in magnetometry surveys

The outputs from a physics-based modeler of magnetometry data have been successfully used with a probabilistic neural network (PNN) to discriminate unexploded ordnance (UXO) from ordnance-related scrap. Cross-validation predictions were performed on three data sets to determine which modeler parameters were most valuable for UXO classification. The best performing parameter combination consisted of the modeler outputs depth, size, and inclination. The cross-validation results also indicated that good prediction performance could be expected. Model outputs from one location at a site were used to train a PNN model, which could correctly discriminate UXO from scrap at a different location of the same site. In addition, data from one site, the former Buckley Field, Arapahoe County, CO, was used to predict targets detected at an entirely different training range. The Badlands Bombing Range, Bulls Eye 2 (BBR 2), Cuny Table, SD. Through careful selection of the probability threshold cutoff, the UXO detection rate obtained was 95% with a false alarm rate of only 37%. The ability to distinguish individual UXO types has been demonstrated with correct classifications between 71% and 95%.

Detection of volatile organic compounds in the vapor phase using solvatochromic dye-doped polymers

Abstract The solvatochromic dye, Reichardts Betaine [2,6-diphenyl-4-(2,4,6-triphenylpyridinio)phenolate] (RDye), was investigated for use in chemical sensing applications. The polarity-sensitive dye was incorporated into several polymer films of differing polarity and sorbent characteristics, and exposed to the headspace above a variety of solvents. Reversible and reproducible changes in the absorption maximum (λmax) of RDye were observed for the series of vapors comprising a range of polarities. The polymers absorbed and concentrated the analyte vapors, thereby influencing the environment surrounding the dye embedded in the polymer. Comparison of the vapor/film results with those of RDye dissolved in the neat solvents were favorable. Methanol and ethanol vapors produced large blue shifts in λmax (avg ≈103 and 63 nm, respectively) for most of the polymer films tested, while isopropanol, acetone, and some chlorinated hydrocarbons produced varying responses in the different films. Water, the most polar vapor tested, did not have much influence on λmax in the films. This is attributed to RDyes low solubility in water and the hydrogen-bonded clustering of water in the films. Exposure of the films to vapors of binary methanol-water mixtures demonstrated that the dye molecules were preferentially solvated by methanol, indicating the detectability of polar vapors in humid environments.

Multiway analysis of preconcentrator-sampled surface acoustic wave chemical sensor array data

New data processing methods for preconcentrator-sampled surface acoustic wave (SAW) sensor arrays are described. The preconcentrator-sampling procedure is used to collect and concentrate analyte vapors on a porous solid sorbent. Subsequent thermal desorption provides a crude chromatographic separation of the collected vapors prior to exposure to the SAW array. This article describes experiments to test the effects of incorporating retention information into the pattern-recognition procedures and to explore the feasibility of multiway classification methods. Linear discriminant analysis (LDA) and nearest-neighbor (NN) pattern-recognition models are built to discriminate between SAW sensor array data for four toxic organophosphorus chemical agent vapors and one agent simulant collected under a wide variety of conditions. Classification results are obtained for three types of patterns: (a) first-order patterns; (b) first-order patterns augmented with the time of the largest peak; and (c) second-order patterns with the use of the SAW frequency for each sensor over a broad time window. Classification models for the second-order patterns are also developed with the use of unfolded and multiway partial least-squares discriminants (uPLSD and mPLSD) and NN and LDA of the scores from unfolded and multiway principal-component analysis (uPCA and mPCA). It is determined that classification performance improves when information about the desorption time is included. Treating the preconcentrator-sampled SAW sensor array as a second-order analytical instrument and using a classification model based upon either uPLSD, uPCA-LDA, or NN results in the correct identification of 100% of the patterns in the prediction set. With the second-order patterns, the other pattern-recognition algorithms only do slightly worse.

COULOMETRIC METHOD FOR THE QUANTIFICATION OF LOW-LEVEL CONCENTRATIONS OF HYDRAZINE AND MONOMETHYLHYDRAZINE

A sensitive and simple coulometric method has been developed for the determination of hydrazine and monomethylhydrazine (MMH) in solution. This coulometric method can readily be used in conjunction with impingers to measure the amount of hydrazine and MMH in air. The method has a limit of detection of less than 25 ng of hydrazine or MMH in 40 mL of solution, which corresponds to less than 2 ppb in 10 L of air. External standards are not necessary as calibration can be performed electronically. The dynamic range of the coulometric method extends from 25 ng to greater than 2500 ng in 40 mL of solution. At the higher concentrations the relative standard deviation was about 2%. The method gave excellent agreement with the two current National Institute for Occupational Safety and Health approved methods at the higher end of this concentration range.

Chemiluminescent chemical sensors for inorganic and organic vapors

Abstract A novel class of chemiluminescent, chemical sensors are described which are based upon the immobilization of a chemiluminescent reagent, luminol or tris(2,2′-bipyridyl)ruthenium(III) (Ru(bpy) 3 3+ ), between a miniature photomultiplier tube and a teflon diffusion membrane. The membrane serves to separate the chemiluminescent matrix from a sampled stream of air. Luminol was immobilized within a hydrogel support matrix and utilized for the detection of ppm levels of chlorinated hydrocarbons, including CCl 4 (g), CHCl 3 (g) and CH 2 Cl 2 (g). This was achieved via the incorporation of a heated Pt filament into the inlet as a pre-oxidative step prior to passage of the gas stream across the teflon membrane. In an alternate approach, the organometallic complex, Ru(bpy) 3 3+ ), an electrochemically regenerable chemiluminescent reagent, was investigated as a reagent for monitoring trace levels of strongly reducing vapors such as hydrazine or ammonia vapor. This approach has been developed into a prototype, real-time instrument. We report the initial testing of a hand-held instrument which is similar to the cell utilized for the luminol chemiluminescence, with the exception that a three electrode electrochemical cell has been incorporated into the system for continuously regenerating the chemiluminescent reagent, Ru(bpy) 3 3+ .