Neural networks based applications

We have presented a neural network based spectrum classifier (NSC) aimed at ultrasonic resonance spectroscopy. The ultrasonic spectroscopy and the NSC has been evaluated in many industrial applications, such as concrete inspection, testing of aerospace composite structures, ball bearings, and aircraft multi-layer structures. The latter application has been presented in some detail. 

Coupling Fast Variable Selection Methods to Neural Network-Based Classifiers Application to Multi-Sensor Systems. 

Lohninger, H. (1993). Evaluation of Neural Networks Based on Radial Basis Functions and Their Application to the Prediction of Boiling Points from Structural Parameters. J. Chem. Inf. Comput.Sci.,33,736-744. 

Lohninger, H. (1993) Fvaluation of neural networks based on radial basis functions and their application to the prediction of boiling points from structural parameters. I. Chem. Inf. Comput. Sci., 33, 736-744. 

The application of ANN for a representation of reaction kinetics can be a very promising method to solve modelling problems. Besides intrinsic kinetics also internal diffusion resistances can be included into the neural network based model. This approach significantly reduces the time required for experimental studies. Despite that neural networks do not help to understand and develop a real reaction mechanism, they make the prediction of the reactor behaviour possible. This approach can be essential in the case of complex or uncertain kinetics - e.g. for polymerization reactions. In this study the neural network approach has been tested for a batch reactor. A trained network can be successfully implemented into any type reactor model. 

Having successfully implemented conventional MRAC techniques, the next logical step was to try to incorporate the MRAC techniques into a neural network-based adaptive control system. The ability of multilayered neural networks to approximate linear as well as nonlinear functions is well documented and has foimd extensive application in the area of system identification and adaptive control. The noise-rejection properties of neural networks makes them particularly useful in smart structure applications. Adaptive control schemes require only limited a priori knowledge about the system to be controlled. The methodology also involves identification of the plant model, followed by adaptation of the controller parameters based on a continuously updated plant model. These properties of adaptive control methods makes neural networks ideally suited for both identification and control aspects [7-11]. 

Dey D, Munshi S (2008) An artificial neural network based system for measurement of humidity and temperature using capacitive humidity sensor and thermistor. Sens Transducers J 97(10) 1-10 Di Francia G, Noce MD, Ferrara VL, LanceUotti L, MorvUlo P, Quercia L (2002) Nanostractured porous silicon for gas sensor application. Mater Sci Technol 18 767-771... 

ABSTRACT The workload of air traffic controllers plays a crucial role in the safety and efficiency of air traffic as it is the prime factor to determine airspace capacity. Workload is the result of traffic complexity which can be described by different sets of parameters. In order to find a small, yet reliable set of complexity parameters that describe controller workload in the Hungarian Air Traffic Control system, a study with multiple steps was conducted and the results are presented in this paper. Information on the most important complexity factors was aquired from experts through interviews and questionnaires and as a result, different sets of parameters were created. To validate the method for gathering information as well as the applicability of complexity factors for airspace capacity estimation, a neural network based model was used. The results indicate that even smaller sets of parameters can be helpful in estimating workload. 

In order to examine the applicability of the complexity factors introduced in Section 3, we decided to use a neural network based approach similar to the one seen in Gianazza Guittet (2006). To provide input data to our networks, air traffic data of the Hungarian airspace was obtained for two days in the past (29th July 2011 and 30th July 2012) and a few more hours from 25th October 2012 when military TRAs were in use. We took snapshots of the radar data every 30 minutes for the two full days and every 10 minutes for the time when TRAs were open. After taking these samples, we had radar data for 107 different air traffic situations which included a wide variety of traffic levels. 

This format was developed in our group and is used fruitfully in SONNIA, software for producing Kohonen Self Organizing Maps (KSOM) and Coimter-Propaga-tion (CPG) neural networks for chemical application [6]. This file format is ASCII-based, contains the entire information about patterns and usually comes with the extension "dat . 

Neural networks have been applied to IR spectrum interpreting systems in many variations and applications. Anand [108] introduced a neural network approach to analyze the presence of amino acids in protein molecules with a reliability of nearly 90%. Robb and Munk [109] used a linear neural network model for interpreting IR spectra for routine analysis purposes, with a similar performance. Ehrentreich et al. [110] used a counterpropagation network based on a strategy of Novic and Zupan [111] to model the correlation of structures and IR spectra. Penchev and co-workers [112] compared three types of spectral features derived from IR peak tables for their ability to be used in automatic classification of IR spectra. 

Chen et al. [24] provide a good review of Al techniques used for modeling environmental systems. Pongracz et al. [25] presents the application of a fuzzy-rule based modeling technique to predict regional drought. Artificial neural networks model have been applied for mountainous water-resources management in Cyprus [26] and to forecast raw-water quality parameters for the North Saskatchewan River [27]. 

Hypercubes and other new computer architectures (e.g., systems based on simulations of neural networks) represent exciting new tools for chemical engineers. A wide variety of applications central to the concerns of chemical engineers (e.g., fluid dynamics and heat flow) have already been converted to run on these architectures. The new computer designs promise to move the field of chemical engineering substantially away from its dependence on simplified models toward computer simulations and calculations that more closely represent the incredible complexity of the real world. 

For PyMS to be used for (1) routine identification of microorganisms and (2) in combination with ANNs for quantitative microbiological applications, new spectra must be comparable with those previously collected and held in a data base.127 Recent work within our laboratory has demonstrated that this problem may be overcome by the use of ANNs to correct for instrumental drift. By calibrating with standards common to both data sets, ANN models created using previously collected data gave accurate estimates of determi-nand concentrations, or bacterial identities, from newly acquired spectra.127 In this approach calibration samples were included in each of the two runs, and ANNs were set up in which the inputs were the 150 new calibration masses while the outputs were the 150 old calibration masses. These associative nets could then by used to transform data acquired on that one day to data acquired at an earlier data. For the first time PyMS was used to acquire spectra that were comparable with those previously collected and held in a database. In a further study this neural network transformation procedure was extended to allow comparison between spectra, previously collected on one machine, with spectra later collected on a different machine 129 thus calibration transfer by ANNs was affected. Wilkes and colleagues130 have also used this strategy to compensate for differences in culture conditions to construct robust microbial mass spectral databases. 

Over the last several years, the number of studies on application of artificial neural network for solving modeling problems in analytical chemistry and especially in optical fibre chemical sensor technology, has increase substantially69. The constructed sensors (e.g. the optical fibre pH sensor based on bromophenol blue immobilized in silica sol-gel film) are evaluated with respect to prediction of error of the artificial neural network, reproducibility, repeatability, photostability, response time and effect of ionic strength of the buffer solution on the sensor response. 

Suah F.B.M., Ahmad M., Taib M.N., Applications of artificial neural network on signal processing of optical fibre pH sensor based on bromophenol blue doped with sol-gel film, Sens. Actuat B 2003 90 182-188. 

Knowledge based approaches such as fuzzy logic, neural networks or multiagents model currently constitute an important axis of research and application in bioprocesses. They have shown their usefulness particularly when one does not have an analytical model but that a certain expertise is available. Harmand and Steyer [37] have addressed that when this expertise comprises a sufficiently important know-how, approaches such as fuzzy logic will be preferred. If, on the other hand, one has only a limited experience but lays out of a rather important data base, the statistical approaches such as neural networks can be used. 

---