Electrolysis tanks are used to smeltmetals based on electrochemical principles,and the short-circuiting of the pole plates in the tanks in the production process will lead to high temperatures,thus affecting normal pr...Electrolysis tanks are used to smeltmetals based on electrochemical principles,and the short-circuiting of the pole plates in the tanks in the production process will lead to high temperatures,thus affecting normal production.Aiming at the problems of time-consuming and poor accuracy of existing infrared methods for high-temperature detection of dense pole plates in electrolysis tanks,an infrared dense pole plate anomalous target detection network YOLOv5-RMF based on You Only Look Once version 5(YOLOv5)is proposed.Firstly,we modified the Real-Time Enhanced Super-Resolution Generative Adversarial Network(Real-ESRGAN)by changing the U-shaped network(U-Net)to Attention U-Net,to preprocess the images;secondly,we propose a new Focus module that introduces the Marr operator,which can provide more boundary information for the network;again,because Complete Intersection over Union(CIOU)cannot accommodate target borders that are increasing and decreasing,replace CIOU with Extended Intersection over Union(EIOU),while the loss function is changed to Focal and Efficient IOU(Focal-EIOU)due to the different difficulty of sample detection.On the homemade dataset,the precision of our method is 94%,the recall is 70.8%,and the map@.5 is 83.6%,which is an improvement of 1.3%in precision,9.7%in recall,and 7%in map@.5 over the original network.The algorithm can meet the needs of electrolysis tank pole plate abnormal temperature detection,which can lay a technical foundation for improving production efficiency and reducing production waste.展开更多
Using seismic data recorded by Yunnan Telemetry Seismic Network from January 1, 2000 to December 31, 2003, the dominant polarization directions of fast shear-waves are obtained at 10 digital seismic stations by SAM te...Using seismic data recorded by Yunnan Telemetry Seismic Network from January 1, 2000 to December 31, 2003, the dominant polarization directions of fast shear-waves are obtained at 10 digital seismic stations by SAM technique, a systematic analysis method on shear-wave splitting, in this study. The results show that dominant directions of polarizations of fast shear-waves at most stations are mainly at nearly N-S or NNW direction in Yunnan. The dominant polarization directions of fast shear-waves at stations located on the active faults are consistent with the strike of active faults, directions of regional principal compressive strains measured from GPS data, and basically consistent with regional principal compressive stress. Only a few of stations.show complicated polarization pattern of fast shear-waves, or are not consistent with the strike of active faults and the directions of principal GPS compressive strains, which are always located at junction of several faults. The result reflects complicated fault distribution and stress field. The dominant polarization direction of fast shear-wave indicates the direction of the in-situ maximum principal compressive stress is controlled by multiple tectonic aspects such as the regional stress field and faults.展开更多
文摘Electrolysis tanks are used to smeltmetals based on electrochemical principles,and the short-circuiting of the pole plates in the tanks in the production process will lead to high temperatures,thus affecting normal production.Aiming at the problems of time-consuming and poor accuracy of existing infrared methods for high-temperature detection of dense pole plates in electrolysis tanks,an infrared dense pole plate anomalous target detection network YOLOv5-RMF based on You Only Look Once version 5(YOLOv5)is proposed.Firstly,we modified the Real-Time Enhanced Super-Resolution Generative Adversarial Network(Real-ESRGAN)by changing the U-shaped network(U-Net)to Attention U-Net,to preprocess the images;secondly,we propose a new Focus module that introduces the Marr operator,which can provide more boundary information for the network;again,because Complete Intersection over Union(CIOU)cannot accommodate target borders that are increasing and decreasing,replace CIOU with Extended Intersection over Union(EIOU),while the loss function is changed to Focal and Efficient IOU(Focal-EIOU)due to the different difficulty of sample detection.On the homemade dataset,the precision of our method is 94%,the recall is 70.8%,and the map@.5 is 83.6%,which is an improvement of 1.3%in precision,9.7%in recall,and 7%in map@.5 over the original network.The algorithm can meet the needs of electrolysis tank pole plate abnormal temperature detection,which can lay a technical foundation for improving production efficiency and reducing production waste.
基金National Natural Science Foundation of China (40274011).
文摘Using seismic data recorded by Yunnan Telemetry Seismic Network from January 1, 2000 to December 31, 2003, the dominant polarization directions of fast shear-waves are obtained at 10 digital seismic stations by SAM technique, a systematic analysis method on shear-wave splitting, in this study. The results show that dominant directions of polarizations of fast shear-waves at most stations are mainly at nearly N-S or NNW direction in Yunnan. The dominant polarization directions of fast shear-waves at stations located on the active faults are consistent with the strike of active faults, directions of regional principal compressive strains measured from GPS data, and basically consistent with regional principal compressive stress. Only a few of stations.show complicated polarization pattern of fast shear-waves, or are not consistent with the strike of active faults and the directions of principal GPS compressive strains, which are always located at junction of several faults. The result reflects complicated fault distribution and stress field. The dominant polarization direction of fast shear-wave indicates the direction of the in-situ maximum principal compressive stress is controlled by multiple tectonic aspects such as the regional stress field and faults.
