Identification of depth and size of subsurface defects by a multiple-voltage probe sensor: Analytical and neural network techniques

Makarov, Sergey; Apelian, Diran; Ludwig, Reinhold; Wang, Hengli
May 2000
AIP Conference Proceedings;2000, Vol. 509 Issue 1, p675
Academic Journal
A theoretical study is conducted to identify subsurface non-contacting defects in green-state powder metallurgy compacts by electrostatic voltage measurements along the specimen’s surface. Extensive three-dimensional numerical simulations were carried out to prove a simple analytical approach, which is based on the dipole approximation for the secondary electric field. Localized defects of different shapes and depths are considered. The simulations lead to the conclusion that the depth of a non-conducting defect can be predicted rather accurately, irrespectively of the actual flaw shape. We found that for a wide variety of shapes and different depth values, there is a linear proportionality between size and voltage distance between the two voltage peaks recorded along the surface. This appears as a direct response to the presence of the defects in the superimposed electric field. The size of the defect does not obey any simple law, especially, for slender defects. However, a rough analytical formula can be established in that case as well. In addition, some conventional neural network (NN) architectures (linear, backprop, Hopfield), including a preprocessor have been applied to the defect identification scheme for the purposes of automatic recognition of noisy data. Performance and recognition errors of different NN models are discussed and the best model is proposed for simultaneous depth recovery and size recognition of the non-conducting defects. © 2000 American Institute of Physics.


Related Articles

  • SINGLE-IMAGE SUPER RESOLUTION FOR MULTISPECTRAL REMOTE SENSING DATA USING CONVOLUTIONAL NEURAL NETWORKS. Liebel, L.; Körner, M. // International Archives of the Photogrammetry, Remote Sensing & S;2016, Vol. 41 Issue B3, p883 

    In optical remote sensing, spatial resolution of images is crucial for numerous applications. Space-borne systems are most likely to be affected by a lack of spatial resolution, due to their natural disadvantage of a large distance between the sensor and the sensed object. Thus, methods for...

  • A neural networks-based negative selection algorithm in fault diagnosis. Gao, X. Z.; Ovaska, S. J.; Wang, X.; Chow, M. Y. // Neural Computing & Applications;2008, Vol. 17 Issue 1, p91 

    Inspired by the self/nonself discrimination theory of the natural immune system, the negative selection algorithm (NSA) is an emerging computational intelligence method. Generally, detectors in the original NSA are first generated in a random manner. However, those detectors matching the self...

  • Neural networks and pattern recognition techniques applied to optical fibre sensors. Lyons, W.B.; Lewis, E. // Transactions of the Institute of Measurement & Control;2000, Vol. 22 Issue 5, p385 

    Some of the significant recent advances in the field of artificial neural networks (ANNs) applied to optical fibre sensors are reviewed. Particular attention is given to the use of ANNs in the enhancement of the performance of existing single point sensors, two- and three-dimensional...

  • Development of Pressure-Temperature Integrated Multifunction Sensor Using Piezo-Resistive Element. Kundu, Palash K.; Sarkar, Gautam; Dutta, Chandan // Sensors & Transducers (1726-5479);Mar2011, Vol. 126 Issue 3, p19 

    A novel attempt was made to develop a multifunction sensor using piezo resistive material for sensing pressure and temperature simultaneously as because it is well known that piezo resistive material has better selectivity to both temperature and pressure or force variables. The advantage of use...

  • An electronic tongue using potentiometric all-solid-state PVC-membrane sensors for the simultaneous quantification of ammonium and potassium ions in water. Gallardo, J.; Alegret, S.; Mu�oz, R.; De-Rom�n, M.; Leija, L.; Hern�ndez, P. R.; del Valle, M. // Analytical & Bioanalytical Chemistry;Sep2003, Vol. 377 Issue 2, p248 

    The simultaneous determination of NH4+ and K+ in solution has been attempted using a potentiometric sensor array and multivariate calibration. The sensors used are rather non-specific and of all-solid-state type, employing polymeric (PVC) membranes. The subsequent data processing is based on the...

  • A neural network approach for on-line fault detection of nitrogen sensors in alternated active sludge. F. Caccavale; P. Digiulio; M. Iamarino; S. Masi; F. Pierri // Water Science & Technology;2010, Vol. 62 Issue 12, p2760 

    In this paper, an effective strategy for fault detection of nitrogen sensors in alternated active sludge treatment plants is proposed and tested on a simulated set-up. It is based on two predictive neural networks, which are trained using a historical set of data collected during fault-free...

  • Artificial Neural Network Model for a Low Cost Failure Sensor: Performance Assessment in Pipeline Distribution. Khan, Asar; Widdop, Peter D.; Day, Andrew J.; Wood, Aliaster S.; Mounce, Steve R.; Machell, John // Enformatika;2006, Vol. 15, p195 

    This paper describes an automated event detection and location system for water distribution pipelines which is based upon low-cost sensor technology and signature analysis by an Artificial Neural Network (ANN). The development of a low cost failure sensor which measures the opacity or...

  • Modelling Indoor Air Carbon Dioxide (CO2) Concentration using Neural Network. Skön, J-P.; Johansson, M.; Raatikainen, M.; Leiviskä, K.; Kolehmainen, M. // World Academy of Science, Engineering & Technology;2012, Issue 61, p879 

    The use of neural networks is popular in various building applications such as prediction of heating load, ventilation rate and indoor temperature. Significant is, that only few papers deal with indoor carbon dioxide (CO2) prediction which is a very good indicator of indoor air quality (IAQ). In...

  • The Performance Improvement of the Target Position Determining System in Laser Tracking Based on 4Q Detector using Neural Network. Salmanpour, A.; Nejad, Sh. Mohammad // World Academy of Science, Engineering & Technology;Aug2011, Issue 56, p548 

    One of the methods for detecting the target position error in the laser tracking systems is using Four Quadrant (4Q) detectors. If the coordinates of the target center is yielded through the usual relations of the detector outputs, the results will be nonlinear, dependent on the shape, target...

  • Design of intelligent fiber-optic statistical mode sensors using novel features and artificial neural networks. Efendioglu, H. S.; Yildirim, T.; Toker, O.; Fidanboylu, K. // Fiber Optic Sensors & Systems;Jul2012, Vol. 26 Issue 7, p5 

    In this paper, novel statistical features namely first moment and second moment were proposed in the analysis of fiber-optic statistical mode sensors to design consistent and sensitive sensors. These features used first time in literature. Experiments were conducted to measure force using...


Read the Article


Sorry, but this item is not currently available from your library.

Try another library?
Sign out of this library

Other Topics