Search

In the long-distance transportation of natural gas, the oleophobic coalescing filter element will experience secondary entrainment and a decrease in filtration efficiency when the gas velocity is too high. Herein, an testing device for the filtration performance of the gasliquid coalescing filter element was established according to the relevant test standards. The effects of the precision grade of the filter element, the material of the drainage layer and the punching height of the inner skeleton on the filtration performance of the oleophobic coalescing filter element under high gas velocity were studied and analyzed. The results show that the high-precision filter material is easy to cause the accumulation of liquid between the last layer of filter material and the drainage layer during filtration at high gas velocity, resulting in the high-precision filter material failing to achieve the optimal filtration effect. Instead, the filter material with moderate precision has the best filtration performance. When the material of the drainage layer is changed to primary felt and sponge,the pressure drop in the channels of the filter element is reduced, the drainage speed of the last filter material layer is accelerated, the liquid accumulation between the last filter material layer and the drainage layer is mitigated, and the secondary entrainment of the filter element is significantly weakened. The secondary entrainment of the filter element can be effectively reduced when the punching height in the inner skeleton is 2 cm from the bottom of the filter element.

In recent years, the research and application of ginseng, a famous and valuable medicinal herb, has received extensive attention at home and abroad. However, with the gradual increase in the demand for ginseng, discrepancies are inevitable when using the traditional manual method for grading the appearance and quality of ginseng. Addressing these challenges was the primary focus of this study. This study obtained a batch of ginseng samples and enhanced the dataset by data augmentation, based on which we refined the YOLOv8 network in three key dimensions: firstly, we used the C2f-DCNv2 module and the SimAM attention mechanism to augment the model's effectiveness in recognizing ginseng appearance features, followed by the use of the Slim-Neck combination (GSConv + VoVGSCSP) to lighten the model These improvements constitute our proposed DGS-YOLOv8 model, which achieved an impressive mAP50 of 95.3% for ginseng appearance quality detection. The improved model not only has a reduced number of parameters and smaller size but also improves 6.86%, 2.73%, and 3.82% in precision, mAP50, and mAP50-95 over the YOLOv8n model, which comprehensively outperforms the other related models. With its potential demonstrated in this experiment, this technology can be deployed in large-scale production lines to benefit the food and traditional Chinese medicine industries. In summary, the DGS-YOLOv8 model has the advantages of high detection accuracy, small model space occupation, easy deployment, and robustness. [ABSTRACT FROM AUTHOR]