Quarry Aggregates: A Flexible Inspection Method Utilising Artificial Neural Networks
Close monitoring of particle size and shape is essential if today’s demanding material requirements for crushed rock aggregates are to be met.
An intelligent mechatronic inspection system is described here. On-line product sampling directs aggregate through an inspection chamber where a novel imaging system digitises the particle geometry. Data processing algorithms extract dimensional data and image features, allowing an artificial neural network classifier to assign qualitative shape descriptors to each particle, thus providing each sampled batch with a breakdown of constituent particle size and shape distributions.
KeywordsInspection System Pattern Vector Knowledge Base System Data Processing Algorithm Remote Laboratory
Unable to display preview. Download preview PDF.
- 1.Bayliss, D C, Parkin, R M, Bearman, R A, 1994. Artificial intelligence in the control of cone crushers, Mechatronics The Basis For New Industrial Development, pp669–704. (Computational Mechanics Publications. ISBN 1 85312 367 6 )Google Scholar
- 2.Bearman, R A, 1991. Energy utilisation in crushing, Mine & Quarry, Vol 20 No 10, pp 35–37.Google Scholar
- 4.Lloyd, P J, 1988. The characterisation of particle shape, Particle Size Analysis, pp347–355. (J Wiley & Sons Ltd. ISBN 0 7837 7654 3 )Google Scholar
- 5.British Standards Institution, 1989. Testing aggregates methods for sampling, BS812 Part 102.Google Scholar
- 6.Holt, C B, 1981. Characterisation of the shape of particles produced by crushing using harmonic series analysis, Institution Of Chemical Engineers Symposium Series No 63, pp D2/H/1-D2/H/13.Google Scholar