Two-Phase Flow Rate Measurement Utilizing Optical Carrier-Based Microwave Interferometry Integrated With Convolutional Neural Network

The precise measurement of gas–liquid two-phase flow rate is crucial for ensuring the safety and efficiency of industrial processes. However, achieving accurate measurement remains a significant challenge. A novel method for measuring flow rates of horizontal gas–liquid two-phase flow employing optical carrier-based microwave interferometry (OCMI) technology and convolutional neural network (CNN) architecture is presented in this article, marking the first application of OCMI in gas–liquid flow rate measurement. Leveraging the distributed measurement capabilities of OCMI, the method captures the distributed information of fluid behavior along the optical fiber and gathers more comprehensive data through the combination of global and distributed interference spectra. The input data are processed utilizing dimensionality reduction techniques, including Pearson correlation and principal component analysis (PCA), and small sample sizes are expanded through data augmentation to improve the accuracy and generalization ability of the model. A decomposed CNN architecture is constructed, with convolutions performed separately along the sequence and feature dimensions, effectively overcoming the limitations of traditional demodulation methods in information extraction. The experimental results demonstrate that the proposed method accurately measures gas and liquid flow rates, offering significant advantages over other variants.