Hardware Implementation of Artificial Neural Networks for Vibroacoustic Signals Classification

• Authors: D. Dąbrowski , E. Jamro , W. Cioch
• Languages of publication: EN

Abstracts

EN

This paper studies the architecture of a neural classifier designed to identify technical condition of machines, based on vibroacoustic signals. The designed neural network is optimized for implementation on Field Programmable Gate Arrays (FPGA) programmable devices. FPGA allows massive parallelism and thus real-time classification as each neuron can execute arithmetic operations simultaneously. The classifier of vibroacoustic signals was designed and tested for the self - organized neural network. The teaching vectors are based on estimates derived from processed vibroacoustic signals generated by rotary machines. The created classifier was applied for recognizing technical state of demonstrative toothed gear DMA1 in variable operating conditions.

Keywords

EN

45.80.+r; 46.40.-f

Discipline

• 46.40.-f: Vibrations and mechanical waves(see also 43.40.+s Structural acoustics and vibration; 62.30.+d Mechanical and elastic waves; vibrations in mechanical properties of solids)
• 45.80.+r: Control of mechanical systems(see also 46.80.+j Measurement methods and techniques in continuum mechanics of solids)

• Publisher: Institute of Physics, Polish Academy of Sciences
• Journal: Acta Physica Polonica A
• Year: 2010
• Volume: 118
• Issue: 1
• Pages: 41-44

Dates

published

2010-07

Contributors

author

D. Dąbrowski

• Department of Mechanics and Vibroacoustics, Department Electronics Mining and Metallurgy Academy, al. Mickiewicza 30, 30-059 Kraków, Poland

author

E. Jamro

• Department of Mechanics and Vibroacoustics, Department Electronics Mining and Metallurgy Academy, al. Mickiewicza 30, 30-059 Kraków, Poland

author

W. Cioch

• Department of Mechanics and Vibroacoustics, Department Electronics Mining and Metallurgy Academy, al. Mickiewicza 30, 30-059 Kraków, Poland

References

• 1. R. Tadeusiewicz, Neural networks, Akademicka Oficyna Wydawnicza RM, Warszawa 1993, (in Polish)
• 2. J. Żurada, M. Barski, W. Jędruch, Artificial neural networks, Wydawnictwo Naukowe PWN, Warszawa, 1996, p. 240, (in Polish)
• 3. T. Kohonen, Neurocomputing 21, 1 (1998)
• 4. E. Jamro, K. Wiatr, Proc. of IEEE Design and Diagnostics of Electronics Circuits and Systems Workshop, Sopron, Apr. 2005, p. 121
• 5. J. Adamczyk, W. Cioch, P. Krzyworzeka, Int. Congress Diagnostyka 2000, Warszawa, p.10
• 6. J. Dybała, S. Radkowski, Diagnostyka 31, 59 (2004)
• 7. P. Czech, B. Łazarz, Diagnostyka 2, 42 (2007)
• 8. W. Bartelmus, R. Zimroz, W. Sawicki, M. Maniak, Z. Woźniak, K. Furmaniak, Górnictwo Geoinżynieria 31, 75 (2007)
• 9. C. Cempel, Vibroacoustics diagnostics, Państwowe Wydawnictwo Naukowe, Warszawa 1989, p 23, (in Polish)
• 10. G.P. Zhang, IEEE Appl. Rev. 30, 451 (2000)
• 11. K. Skahill, VHDL, Projecting Programmable Logic Devices, II Ed., WNT, Warszawa 2004, (in Polish)
• 12. J.S. Shawe-Taylor, M. Anthony, W. Kern, Neural Networks, Vol. 5, 1992, p. 971
• 13. J.C. Gallagher, S.K. Boddhu, S. Vigraham, Evolutionary Computation, Vol.3, 2005, p. 2461

Identifiers

DOI

10.12693/APhysPolA.118.41