On reliability of neural network sensitivity analysis applied for sensor array optimization

Abstract Sensor arrays are nowadays first choice solution for low-cost and portable gas mixtures analysis systems. The key issue in construction of such systems is selection of sensors. The authors try to apply neural network sensitivity analysis for this task. The algorithm starts from huge set of sensors, which provides satisfying operation of the system, and then detects the most redundant elements, which may be removed without significant decrease of system accuracy. Eventually the small but efficient array of sensors is obtained. Authors present the method, propose some modifications and discuss problems with its application. Results of sensitivity analysis approach are compared with exhaustive search for the best set of sensors. The case study is quantitative analysis of volatile organic compounds mixtures by means of commercial, tin dioxide, TGS 800 series sensors characterized in in-house developed gas chamber.

[1]  P. Corcoran,et al.  The application of genetic algorithms to sensor parameter selection for multisensor array configuration , 1999 .

[2]  Véronique Bellon-Maurel,et al.  Optimisation of electronic nose measurements. Part I: Methodology of output feature selection , 1998 .

[3]  Stefano Tarantola,et al.  Global Sensitivity Analysis: An Introduction , 2005 .

[4]  L. G. Kwong,et al.  Data analysis for a hybrid sensor array , 2005 .

[5]  T. Kijima,et al.  A new type of host compound consisting of α-zirconium phosphate and an animated cyclodextrin , 1986, Nature.

[6]  Evor L. Hines,et al.  Enhancing electronic nose performance by sensor selection using a new integer-based genetic algorithm approach , 2005 .

[7]  Teodiano Bastos,et al.  SLAM-based robotic wheelchair navigation system designed for confined spaces , 2010, 2010 IEEE International Symposium on Industrial Electronics.

[8]  Jacek M. Zurada,et al.  Sensitivity analysis for minimization of input data dimension for feedforward neural network , 1994, Proceedings of IEEE International Symposium on Circuits and Systems - ISCAS '94.

[9]  Jacek M. Zurada,et al.  Perturbation method for deleting redundant inputs of perceptron networks , 1997, Neurocomputing.

[10]  Terry Andres,et al.  Sensitivity analysis of model output: an investigation of new techniques , 1993 .

[11]  José Santos,et al.  Application of artificial neural networks to calculate the partial gas concentrations in a mixture , 2001 .

[12]  Jens Petter Wold,et al.  Detection of rancidity in freeze stored turkey meat using a commercial gas-sensor array system , 2006 .

[13]  Henrik Petersson,et al.  Sensor array optimization using variable selection and a Scanning Light Pulse Technique , 2009 .

[14]  Przemyslaw M. Szecowka,et al.  Simultaneous quantification of carbon monoxide and methane in humid air using a sensor array and an artificial neural network , 1997 .

[15]  A. Szczurek,et al.  Sensors selection for gas mixtures analysis systems , 2003, 26th International Spring Seminar on Electronics Technology: Integrated Management of Electronic Materials Production, 2003..

[16]  Andries Petrus Engelbrecht,et al.  A new pruning heuristic based on variance analysis of sensitivity information , 2001, IEEE Trans. Neural Networks.

[17]  Dominique Martinez,et al.  Quantification of gas concentrations in mixtures of known gases using an array of different tin-oxide sensors , 2003 .

[18]  Przemyslaw M. Szecowka,et al.  Application of sensor array and neural networks for quantification of organic solvent vapours in air , 1999 .

[19]  Fabrizio Davide,et al.  A composed neural network for the recognition of gas mixtures , 1995 .

[20]  I. Sayago,et al.  Integrated sensors for monitoring contaminant gases in atmospheres and soils , 1997, ISIE '97 Proceeding of the IEEE International Symposium on Industrial Electronics.

[21]  Zou Xiaobo,et al.  Vinegar Classification Based on Feature Extraction and Selection From Tin Oxide Gas Sensor Array Data , 2003 .

[22]  Changsheng Xie,et al.  A sensor array optimization method for electronic noses with sub-arrays , 2009 .

[23]  T. Sobanski,et al.  Analysis of gas mixture using sensor array and artificial neural networks , 2002 .

[24]  P. Travers,et al.  A method for selecting an optimum sensor array , 2000 .

[25]  R. Todeschini,et al.  A chemometric approach based on a novel similarity/diversity measure for the characterisation and selection of electronic nose sensors. , 2006, Analytica chimica acta.

[26]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.