Soft Sensing Based on Probabilistic Neural Network

DOI

10.2991/iccet-15.2015.292How to use a DOI?

Keywords

wavelet packet; probabilistic neural network; soft sensing

Abstract

Aimed at the characteristic of nonlinear and non-stationary of pressure drop, in this article a flow regime identification soft sensing method using wavelet-packet combined with probabilistic neural network is put forward. PNN is used as classier due to its good generalization ability and fast learning capability, case of on-line updating, and sound statistical foundation in Bayesian estimation theory. The features are extracted from the differential pressure fluctuation signals of the air-water two-phase flow in the horizontal pipe and the wavelet packet energy features of various flow regimes are obtained. Then combining the energy features with probabilistic neural network, a new way to identify flow regime by soft sensing is proposed.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Qiang Wang
PY  - 2015/11
DA  - 2015/11
TI  - Soft Sensing Based on Probabilistic Neural Network
BT  - Proceedings of the 5th International Conference on Civil Engineering and Transportation 2015
PB  - Atlantis Press
SP  - 1569
EP  - 1572
SN  - 2352-5401
UR  - https://doi.org/10.2991/iccet-15.2015.292
DO  - 10.2991/iccet-15.2015.292
ID  - Wang2015/11
ER  -