Method for identify gas and liquid two-phase flow pattern of wet dust removing equipment

摘要: The invention relates to a system and method for identify gas liquid two-phase flow patterns of dust removing equipment. A complete identification process collection pressure signals, analysis difference signals gas-liquid mixed states is formed by combining neural network with time domain wavelet analysis. high infault tolerance nonlinearity, overcoming the shortcoming errors frequency spectrum in prior art. accurate rapid. real-time monitoring effect achieved wet equipment whose main principle that collector produced impact air upon level, accuracy measured an experiment reaches 99.6%. Characteristic parameters are obtained through taken as input sample feature vector library, trained realize rapid conversion between measurement identification.