Quantification of butanol and ethanol in aqueous phases by reflectometric interference spectroscopy - different approaches to multivariate calibration

Dieterle, Frank; Nopper, Dirk; Gauglitz, Günter

doi:10.1007/s002160100908

Cited by 9 publications

(9 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Even optimisations such as the input range of the value of the sigmoid transfer function and the reduction of nodes in the input and hidden layers in the OSN and in first stage of the TSN should be tested. For the latter, pruning algorithms should be helpful to reduce overtraining effects and redundant information [24,40]. It should be pointed out that all measurements were done with nitrogen as carrier gas and at 70°C.…”

Section: Resultsmentioning

confidence: 99%

“…Therefore, techniques such as principle compound analyse (PCA) and artificial neural network (ANN) are implemented to analyse gas mixtures. Electronic tongues and noses [7,8,9] use these methods for pattern recognition and classifications in a wide range of applications, such as the discrimination of volatile organic compounds (VOCs) [10], nitroaromatic compounds (NACs) in VOC background [11], tastes of fruits, tomatoes [12], fruit juices, milk [9] and teas [13], odours of automotive textiles [14,15], perfume oils [16], alcohols, coffees [17], wines [18,19], beers [19], grains [20] and olive oils [21], the characterisation of the rancidity of oils [14], the evaluation of freshness of cod-fish fillets [22], the discrimination of odours of animal farms [23] and of breaths of healthy and ketotic cows [17], the monitoring of water quality [9,24], the discrimination of K. Henkel · D. Schmeißer human breaths from normal subjects and patients with renal diseases [25] and the analysis of binary and tertiary gas mixtures [26,27,28,29,30,31,32]. Mainly arrays of sensors are installed using different sensor elements based on resistive (semiconducting metal oxide thin films (MOX) [10,14,15,17,19,20,26,27,…”

Section: Introductionmentioning

confidence: 99%

“…ANNs are already successfully installed in several domains of medicine [33,34,35], biology [33,34,36], physics [33,34,36], mathematics [33,34], information engineering [33,34,36], traffic engineering [35], aerospace [33,34], financial service [33,34], meteorology [33], corrosion science [37] and sensors [7,10,16,17,20,23,24,30,31,32,34,35,36,37,38,39,40].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Back-propagation-based neural network with a two sensor system for monitoring carbon dioxide and relative humidity

Henkel

Schmeißer

2002

Anal Bioanal Chem

View full text Add to dashboard Cite

The delicate problem of independently determining the concentration of carbon dioxide and relative humidity in a gas mixture is solved by recording the signals of two quartz microbalances coated with a functionalized sensitive polymer layer. The data were analysed by back-propagation-based neural networks. We tested two different architectures, a one-stage net and a two-stage net (which consist of two network structures in series) with respect to their generalisation ability. The one-stage network calculated the concentration of carbon dioxide with an error less than 12% and that of relative humidity with an error less than 5%. The two-stage network recognised in the first step seven categories of relative humidity with a success rate of 100%, while the maximal error of carbon dioxide concentration calculated in the second level of this architecture is reduced to 9%. In addition, the generalisation ability of the two-stage network is improved compared to the one-stage network.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Back-propagation-based neural network with a two sensor system for monitoring carbon dioxide and relative humidity

Henkel

Schmeißer

2002

Anal Bioanal Chem

View full text Add to dashboard Cite

show abstract

“…Combined Effects of All Types of Noise. The effect of combining all three types of simulated noise was explored for all three combinations of fluorophores, namely our chosen combination of five fluorophores (Combination A, simulations 4-6), an alternative, less spectrally distinct combination of five (Combination B, simulations 7-9), and the combination of three fluorophores we are currently using for MSPs during initial in vivo experiments (Combination C, simulations [13][14][15]. Each type of simulated noise is independently varied, with the other two types of noise held at anticipated worst-case levels.…”

Section: Effects Of Individual Simulated Sources Of Noisementioning

confidence: 99%

“…Because PLS combines regression and decomposition into one step, it identifies spectral vectors that are directly related to the physical constituents of interest. 14,15 However, a large calibration or training set is necessary for accurate results. In our case, a spectra set required for training cannot be easily obtained in situ or in optical phantoms.…”

Section: Introductionmentioning

confidence: 99%

Partial Least Squares Based Decomposition of Five Spectrally Overlapping Factors

2005

View full text Add to dashboard Cite

The development of a multi-sensory fiber-optic based fluence rate probe (MSP) for light monitoring and dosimetry during photodynamic therapy (PDT) created the need for a robust multivariate signal analysis algorithm capable of quantifying the intensity of five component spectra, representing the sensors, which display a large degree of spectral overlap. Partial least squares (PLS) analysis, as an option for such an analysis algorithm, was evaluated through simulations in the presence of three types of noise, which experimentally may limit the accuracy of PLS quantification of component spectra contributions. Random, or white noise, background was varied over a range of 0-15% relative intensity. A non-random (Gaussian) background vector was simulated, as an experimentally relevant spectral interferent, over a range of 0-7% relative intensity. Spectral variation was modeled by a spectral shift of +/-1-5 nm. Procedures for selecting preferred combinations of fluorophores, with minimum possible spectral overlap, were developed. To illustrate the importance of this selection process, PLS performance results were compared for two possible combinations of five fluorophores, as well as for the combination of three fluorophores currently in experimental use with MSPs. Experimentally anticipated worst-case quantifications were identified for all three types of simulated noise (5% random background, 0.5% Gaussian background, and +/-2 nm spectral shift). The effects of single and combined types of noise were evaluated by independently varying each type of simulated noise over the experimentally relevant range. The mean percentage error in determining the fluorophore contributions and hence quantification of the fluence rate were compared with the required performance standard of better than 10% error for optical power meters in medical use. The PLS algorithm provided an accuracy of 7 +/- 2% for five fluorophores and 3.3 +/- 0.8% for three fluorophores, indicating that PLS is an appropriate choice for this application.

show abstract

Different approaches to multivariate calibration of nonlinear sensor data

2004

View full text Add to dashboard Cite

In this study, different approaches to the multivariate calibration of the vapors of two refrigerants are reported. As the relationships between the time-resolved sensor signals and the concentrations of the analytes are nonlinear, the widely used partial least-squares regression (PLS) fails. Therefore, different methods are used, which are known to be able to deal with nonlinearities present in data. First, the Box-Cox transformation, which transforms the dependent variables nonlinearly, was applied. The second approach, the implicit nonlinear PLS regression, tries to account for nonlinearities by introducing squared terms of the independent variables to the original independent variables. The third approach, quadratic PLS (QPLS), uses a nonlinear quadratic inner relationship for the model instead of a linear relationship such as PLS. Tree algorithms are also used, which split a nonlinear problem into smaller subproblems, which are modeled using linear methods or discrete values. Finally, neural networks are applied, which are able to model any relationship. Different special implementations, like genetic algorithms with neural networks and growing neural networks, are also used to prevent an overfitting. Among the fast and simpler algorithms, QPLS shows good results. Different implementations of neural networks show excellent results. Among the different implementations, the most sophisticated and computing-intensive algorithms (growing neural networks) show the best results. Thus, the optimal method for the data set presented is a compromise between quality of calibration and complexity of the algorithm.

show abstract

Quantification of butanol and ethanol in aqueous phases by reflectometric interference spectroscopy - different approaches to multivariate calibration

Cited by 9 publications

References 2 publications

Back-propagation-based neural network with a two sensor system for monitoring carbon dioxide and relative humidity

Back-propagation-based neural network with a two sensor system for monitoring carbon dioxide and relative humidity

Partial Least Squares Based Decomposition of Five Spectrally Overlapping Factors

Different approaches to multivariate calibration of nonlinear sensor data

Contact Info

Product

Resources

About