Acoustic Emission (AE) technology is a promising way to non-intrusively measure the size of particles in pneumatic conveying pipelines. In AE-based particle sizing, extracting representative features from an AE signal and establishing the relationship between the features and particle size are essential. In this paper, AE signals from particles of different sizes are collected from a gas-solid flow test rig. Wavelet analysis is used to denoise the signals. The denoising performance of different wavelet parameters (wavelet function, decomposition level and thresholding) is compared based on signal-to-noise ratio and signal smoothness. The particle size is predicted through a neural network with energy fraction features extracted through wavelet analysis as the network inputs. Experimental results demonstrate that the relative error of the particle sizing system is no greater than 23%.
An updated version of this article was published on 31 October 2014. The Corrigendum attached to the corrected article PDF file explains the changes made to the original paper.
