Journal of Theoretical
and Applied Mechanics

40, 3, pp. 797-815, Warsaw 2002

Smart sensor for operational load measurements

Tadeusz Uhl, Maciej Petko
The important technique enabling machine health monitoring and fault localisation is operational load measurement; however direct measurement is usually difficult or even impossible. The paper deals with the realisation of the idea of developing "smart sensor" which estimates the load based on structure response. In the case study, hardware neural network has been used to obtain the load course from the vibrations in specific points of the structure. The paper covers also problems with prototyping and implementation stages during development of signal processing algorithms with emphasis placed on ASIC/FPGA based hardware platform, for which a methodology of implementation is formulated and validated by practical application to the smart sensor problem. Details of the procedure are presented along with the tools used and results obtained during its realization. Performance of the device during experiment is analysed and, finally, conclusions are shown.
Keywords: smart sensor; mechatronics; neural networks