Using Wavelet Neural Network To Analyze The Data Of Wet Gas Flowmeter

The metering of gas-liquid two phase flow rates of wet gas is an essentially reference for the monitoring and controlling of gas well and gas reservoir dynamics , but the traditional wet gas flowmeter is disturbed by many factors, so it has large error. The reality is unsatisfactory due to the unsolved problem of two phase flow metering. The paper provides and studies the method of processing the data of wet gas measurement based on wavelet neural network (WNN).Firstly, this thesis makes systematic studies of structures of wavelet neural networks; it also compares wavelet neural networks with BP neural networks in detail. Secondly, two kinds of wavelet neural networks are built. The first one, based on continuous wavelet transform theory, the thesis presents continuous parameter wavelet neural network, whose activation functions are continuous wavelet functions. Furthermore, the algorithm of network is given. The second one, according to multi-resolution analysis and orthonormal wavelet decomposition theory, multi-resolution wavelet neural network is built, whose activation functions consist of orthonormal wavelet and orthonormal scaling functions, and hierarchical approximation algorithm is given, and process the data of two-phase flow measurement with two kinds of network separately. Lastly, combining network with signal characteristic, comparing many wavelet functions, we choose the best.The inputs of the neural network are differential pressure, absolute pressure and temperature of two-phase flow;the outputs of the neural network are gas and liquid flow rates. Based on MATLAB, we achieve the prediction of wet gas flow. Following conclusion can be drawn: using wavelet neural networks to forecast wet gas flow is feasible, and get a nice result. It provides a basis for the practical production.