This work presents a new level-detection method, it is based on the acoustic resonance of a waveguide. This method can be successfully implemented for fluid-level detection in industrial plants, where problems of residues, deposits, foams, etc., can be expected. The performance of the detector is additionally improved by the use of neural networks. Neural networks enable the detector to learn from the reference detection cycles; therefore, such a system is flexible and can he easily optimized (trained) for specific operating conditions. The detector has been tested in a highly viscous medium, Styrofoam granulate and under artificially induced residues and deposits in the resonance waveguide.
https://www.sciencedirect.com/science/article/pii/S0924424797800635
English 
Slovenian