Gas/Liquid Two-Phase Flow Regime Recognition Based on Adaptive Wavelet-Based Neural Network

Flow regime recognition of two-phase flow is of great importance in industrial process. In this paper, a new method is brought forward to recognize the gas/liquid two-phase flow regime. The information of the method that provided by electrical resistance tomography (ERT) is the measured data in horizontal pipe. A new adaptive wavelet-based neural network was introduced and it combines the wavelet transformation with neural network theory in this paper. The parameters of the wavelet are adjusted adaptively according to signal´s characteristic in the learning process, so the feature of the signal could be extracted to a large extent and the recognition results of flow regime would be better.