Detection and Dispersion Analysis of Water Globules in Oil Samples Using Artificial Intelligence Algorithms

Fluid particle detection technology is of great importance in the oil and gas industry for improving oil-refining techniques evaluating quality refining equipment. The article discusses process creating a computer vision algorithm that allows user to detect water globules samples analyze their sizes. developing an based on convolutional neural network (CNN) YOLOv4 presented. For this study, our own empirical base was proposed, which comprised microphotographs raw materials water–oil emulsions taken at various points different operating modes refinery. number images training increased by applying authors’ augmentation algorithm. developed program makes it possible particles fluid medium with level accuracy required researcher, can be controlled stage CNN. Based results processing output data from algorithm, dispersion analysis localized carried out, supplemented frequency diagram describing ratio size found. evaluation work intelligent comparison manual method verification two distributions allow us conclude model CNN verified accepted use search medium. AP@50 = 89% AP@75 78%.