Abstract A flow pattern recognition method based on Choi-Williams analysis and neural network is proposed. The array conductivity sensor is used to obtain the flow pattern information of gas-liquid two-phase flow in vertical rising pipeline, and the multivariate measurement data are denoised and dimensionally reduced. Further, Choi-Williams analysis is used to convert it into time-frequency spectrogram, and the data set is constructed. CNN, VGG-16 and ResNet-18 network models are built respectively, and the time-frequency spectrograms of different flow patterns are used as network input for training and testing. The recognition results show that Choi-Williams analysis can effectively extract the characteristics of different flow pattern signals, and the three networks have high recognition accuracy, among which ResNet-18 network has the highest accuracy, with an average flow pattern recognition rate of 99.4%.
Zhou F Y, Jin L P, Dong J. Review of convolutional neural network ［J］. Chinese Journal of Computers, 2017, 40(6): 1229-1251.
［16］
Taitel Y, Bornea D, Dukler A E. Modelling flow pattern transitions for steady upward gas-liquid flow in vertical tubes ［J］. AIChE Journal, 1980, 26: 345-354.
Fang L D, Wang P P, Wang S, et al. Study on slug flow mechanism of gas-liquid two-phase flow in long throat venturi ［J］. Acta Metrologica Sinica, 2020, 41(1): 48-54.
Wang X X, Wang B, Chen Y Z. Two phase flow pattern recognition based on reconstructed image of electrical capacitance tomography ［J］. Acta Metrologica Sinica, 2020, 41(8): 942-946.
Chu W J, Liu Y, Pan L Q, et al. Identification of boiling flow pattern in narrow rectangular channel based on TFA-CNN combined method ［J］. Flow Measurement and Instrumentation, 2022, 83: 102086.
［11］
Li X Y, Li L X, Zhao H X, et al. Identification of two-phase flow pattern in porous media based on signal feature extraction ［J］. Flow Measurement and Instrumentation, 2022, 83: 102123.
Jin N D, He X F, Luo T. Wigner-Ville analysis of measurement fluctuating signals of conductance sensor in gas/liquid two-phase flow ［J］. Transducer and Microsystem Technologies, 2006, 25(12): 29-31.
Liu Q C, Zhou Y L, Chen C. Flow pattern identification of fluctuating vibration gas liquid two phase flow based on CEEMDAN and probabilistic neural network ［J］. Chinese Journal of Scientific Instrument, 2021, 42(10): 84-93.
［14］
Wang P, Miao Y Y. Multi classification ERT flow pattern recognition method based on deep learning ［J］. Journal of Physics: Conference Series, 2021, 2181: 012010.
Wen R Y, Sun B, Zhao Y X, et al. Flow pattern recognition method of gas-liquid two-phase flow based on adaptive optimal kernel and convolution neural network ［J］. CIESC Journal, 2018, 69(12): 5065-5072.
Zhang L F, Wang Z, Wu S C. Flow Pattern Recognition of Gas-liquid Two-phase Flow Based on Convolutional Neural Network and Gated Recurrent Unit［J］. Acta Metrologica Sinica, 2022, 43(10): 1306-1312.
Dong F, Zhang S, Shi X, et al. Flow regimes identification-based multidomain features for gas-liquid two-phase flow in horizontal pipe ［J］. Transactions on Instrumentation and Measurement, 2021, 70: 1-11.
Zhang L F, Zhu Y F. Identification of two-phase flow based on MO-PLP-ELM and electrical capacitance tomography ［J］. Acta Metrologica Sinica, 2021, 42(3): 334-338.
Tong W G, Pang X C, Zhu G H. Gas-liquid two-phase flow pattern recognition method based on convolutional neural network ［J］. Journal of System Simulation, 2021, 33(4): 883-891.
Deng Y Q, Jin X, Liu G, et al. Hilbert-huang transform processing method of non-stationary signal in pipeline flow field ［J］. Applied Chemical Industry, 2020, 49(4): 1015-1019.
Fang L P, Li Y X, Liu C W, et al. Time-frequency analysis method study of the leakage acoustic signal in gas-liquid two-phase flow pipelines ［J］. Oil-Gas Field Surface Engineering, 2019, 38(1): 94-100.
［10］
Choi H I, Williams W J. Improved time-frequency representation of multi-component signals using exponential kernels ［J］. IEEE Transactions on Acoustics, Speech, and Signal Processing, 1989, 37(6): 862-871.
Zhang L F, Zhu Y F. Two-phase flow regime and its parameter prediction based on particle swarm optimization extreme learning machine and electrical capacitance tomography ［J］. Acta Metrologica Sinica, 2020, 41(12): 1488-1493.
Tan C, Shen Y, Dong F, et al. Gas-liquid flow pattern analysis based on graph connectivity and graph-variate dynamic connectivity of ERT ［J］. Transactions on Instrumentation and Measurement, 2019, 68(5): 1590-1601.
2023, Vol. 44 
