Irregular Shape Particle Size Analysis Using Neural Network Approach

  • Edward Rój
Conference paper
Part of the Advances in Soft Computing book series (AINSC, volume 19)


Granular materials of irregular shape occur in great variety in nature and in different branches of industry. Chemical engineering processes as catalysis, adsorption, heat exchange and many others using the granular materials are of great practical value. Also many commercial products are of granular type and that is why the characteristics of the granular material is so important. In the paper an optical method, combined with neural network approach, developed for irregular shape particle size analyzing has been presented. The method relies on determination of a relative volume and relative surface of sample particles using grid and random secant methods, respectively. Some experimental data obtained for ammonia synthesis catalyst has also been compared with sieve analysis method results.


Granular Material Optical Method Neural Network Approach Relative Surface Image Analysis Technique 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Bobrowski, D. (1980), Probabilistic in technical applications, WNT, WarszawaGoogle Scholar
  2. 2.
    Carman A.C. (1956) Flow of gases through porus media, London Butterworth Scientific PublicationsGoogle Scholar
  3. 3.
    Duch, W., Grudzifiski, K. (1999), The weighted k-NN with selection of features and its neural realization, Proceedings of the Fourth Conference “Neural Networks and Their Applications”, CzestochowaGoogle Scholar
  4. 4.
    Rój E. (2001), Determination the particle mean diameter of a packed bed using optical method. Inz. Chem. i Proc., 22, 3E, 1231–1236Google Scholar
  5. 5.
    Ryg, J. (1970), An introduction to quantity metallography, Sl4sk Publishing House, KatowiceGoogle Scholar
  6. 6.
    Strumillo, P., Szczypiríski, P.,et al.(1997), A computer program for analysis of X-ray images of wheat grains. Scientific Works, Electronics, Lódz Technical University (In Polish)Google Scholar
  7. 7.
    Strzelecki, M. (1997), Segmentation of textured biological images using multilayer perceptron neural networks. Scientific Works, Electronics, Lódz Technical University (In Polish)Google Scholar
  8. 8.
    Zhu, Q. (1998), Minimum Cross-Entropy Approximation for Modeling of Highly Interwining Data Sets at Subclass Levels. Journal of Intelligence Information System 11, 139–152CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Edward Rój
    • 1
  1. 1.Fertilizers Research InstitutePuławyPoland

Personalised recommendations