Dongxing Yu
Dongxing Yu has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- Xinyi Zhao
Sino-European Institute of Aviation Engineering, Civil Aviation University of China, Tianjin 300300, China
2 shared publications - Bing Han
Sino-European Institute of Aviation Engineering, Civil Aviation University of China, Tianjin 300300, China
1 shared publication H Guang
Tianjin Fire Research Institute of MPS
1 shared publicationH Weiping
Tianjin Fire Research Institute of MPS
1 shared publicationJ Lishuai
Tianjin Fire Research Institute of MPS
1 shared publicationJun Zhang
Qingdao Sunyfire Science and Technology Shares Co. Ltd. Qingdao China
1 shared publication
Latest Publications
- Open Access
Dual-Path CSDETR: Cascade Stochastic Attention with Object-Centric Priors for High-Accuracy Fire Detection
Detecting dynamic and amorphous objects like fire and smoke poses significant challenges in object detection. To address this, we propose Dual-Path Cascade Stochastic DETR (Dual-Path CSDETR). Unlike Cascade DETR, our model introduces cascade stochastic attention (CSA) to model the irregular morphologies of fire and smoke through variational inference, combined with a dual-path architecture that enables bidirectional feature interaction for enhanced learning efficiency. By integrating object-centric priors from bounding boxes into each decoder layer, the model refines attention mechanisms to focus on critical regions. Experiments show that Dual-Path CSDETR achieves 94% AP50 on fire/smoke detection, surpassing deterministic baselines.
Open AccessDual-Path CSDETR: Cascade Stochastic Attention with Object-Centric Priors for High-Accuracy Fire Detection
Detecting dynamic and amorphous objects like fire and smoke poses significant challenges in object detection. To address this, we propose Dual-Path Cascade Stochastic DETR (Dual-Path CSDETR). Unlike Cascade DETR, our model introduces cascade stochastic attention (CSA) to model the irregular morphologies of fire and smoke through variational inference, combined with a dual-path architecture that enables bidirectional feature interaction for enhanced learning efficiency. By integrating object-centric priors from bounding boxes into each decoder layer, the model refines attention mechanisms to focus on critical regions. Experiments show that Dual-Path CSDETR achieves 94% AP50 on fire/smoke detection, surpassing deterministic baselines.
- Open Access
Improved YOLOv5: Efficient Object Detection for Fire Images
Camera-based fire detection systems have been found to yield significantly superior results compared to their sensor-based counterparts, and are thus widely employed for this purpose across the globe. This study presents an effective and lightweight fire detection technique based on deep learning. A novel convolutional neural network has been created specifically for the purpose of identifying fire areas. This network utilizes the YOLOv5 algorithm. The default versions of the You Only Look Once (YOLO) technique demonstrate limited accuracy when applied to fire detection instances, even after undergoing training and testing. After training and testing in fire detection instances, the default versions of the YOLO technique exhibit a significantly low level of accuracy. We chose the YOLOv5 network to augment its capabilities and maximize its effectiveness in identifying fire disasters. The self-designed convolution module, also known as the improved convolution module, is implemented to substitute the regular convolution module in YOLOv5. By dividing the input features into two main portions, it helps to realize the lightweight of the network. Additionally, the network incorporates the attention mechanism. The testing findings clearly show that the proposed method is more effective than previous ways of identifying fire, with a precision of 91.22% and a recall of 93.78%.
Open AccessImproved YOLOv5: Efficient Object Detection for Fire Images
Camera-based fire detection systems have been found to yield significantly superior results compared to their sensor-based counterparts, and are thus widely employed for this purpose across the globe. This study presents an effective and lightweight fire detection technique based on deep learning. A novel convolutional neural network has been created specifically for the purpose of identifying fire areas. This network utilizes the YOLOv5 algorithm. The default versions of the You Only Look Once (YOLO) technique demonstrate limited accuracy when applied to fire detection instances, even after undergoing training and testing. After training and testing in fire detection instances, the default versions of the YOLO technique exhibit a significantly low level of accuracy. We chose the YOLOv5 network to augment its capabilities and maximize its effectiveness in identifying fire disasters. The self-designed convolution module, also known as the improved convolution module, is implemented to substitute the regular convolution module in YOLOv5. By dividing the input features into two main portions, it helps to realize the lightweight of the network. Additionally, the network incorporates the attention mechanism. The testing findings clearly show that the proposed method is more effective than previous ways of identifying fire, with a precision of 91.22% and a recall of 93.78%.
- 15 January 2024
- Published by Wiley in Asia-Pacific Journal of Chemical Engineering
- Vol. 19 (2)
A study on the performance of a swirling–straight composite nozzle
The nozzle is the key component of water mist fire extinguishing system. As a response to the problems of small injection angle associated with straight jet nozzles and the weak axial momentum of swirl nozzles, a swirling–straight composite nozzle is designed in this work. The comparison with a straight jet nozzle and a pressure swirl nozzle shows that the swirling–straight composite nozzle has a larger axial momentum and better injection angle. Under the same pressure, the volume flow of the swirling–straight composite nozzle is more than 27% of the pressure swirl nozzle, and the injection angle was more than 65% of straight jet nozzle. The numerical model of the swirling–straight composite nozzle is established. Meantime, the internal flow field characteristics and the influence of the straight jet aperture on the performance are studied. The results demonstrate that the straight jet fluid and swirling fluid can be mixed well in the nozzle, and a larger axial momentum and tangential momentum can be obtained. With the increase of the straight jet aperture, the swirl effect in the nozzle becomes weaker, the injection angle becomes smaller, and the axial momentum improves. When the straight jet aperture increases from 1.1 to 1.9 mm, the straight jet volume flow at the nozzle inlet increases by 127%, and the injection angle reduces by 40%.
- 15 January 2024
- Published by Wiley in Asia-Pacific Journal of Chemical Engineering
- Vol. 19 (2)
A study on the performance of a swirling–straight composite nozzle
The nozzle is the key component of water mist fire extinguishing system. As a response to the problems of small injection angle associated with straight jet nozzles and the weak axial momentum of swirl nozzles, a swirling–straight composite nozzle is designed in this work. The comparison with a straight jet nozzle and a pressure swirl nozzle shows that the swirling–straight composite nozzle has a larger axial momentum and better injection angle. Under the same pressure, the volume flow of the swirling–straight composite nozzle is more than 27% of the pressure swirl nozzle, and the injection angle was more than 65% of straight jet nozzle. The numerical model of the swirling–straight composite nozzle is established. Meantime, the internal flow field characteristics and the influence of the straight jet aperture on the performance are studied. The results demonstrate that the straight jet fluid and swirling fluid can be mixed well in the nozzle, and a larger axial momentum and tangential momentum can be obtained. With the increase of the straight jet aperture, the swirl effect in the nozzle becomes weaker, the injection angle becomes smaller, and the axial momentum improves. When the straight jet aperture increases from 1.1 to 1.9 mm, the straight jet volume flow at the nozzle inlet increases by 127%, and the injection angle reduces by 40%.
- Open Access
- 11 October 2018
- Published by IOP Publishing in IOP Conference Series: Earth and Environmental Science
- Vol. 186 (5) , 012013
Design and heat release rate test of freight carriage model
The purpose of this paper was to ensure freight carriage's combustion time. Firstly, the freight carriage's fire load was obtained by reviewing papers. The freight carriage's HRR would be over 10 MW if burning time was enough. Secondly, the freight carriage model was designed referencing the real freight carriage's parameters. The standard combustible material cup boxes were used to simulate freight materials. Thirdly, the freight carriage model's HRR was tested using 10 MW cone calorimeter. Fitting curve was obtained based on t2 fire model. Further study should focus on freight carriage model's fire extinguishing experiment.
- 11 October 2018
- Published by IOP Publishing in IOP Conference Series: Earth and Environmental Science
- Vol. 186 (5) , 012013
Open AccessDesign and heat release rate test of freight carriage model
The purpose of this paper was to ensure freight carriage's combustion time. Firstly, the freight carriage's fire load was obtained by reviewing papers. The freight carriage's HRR would be over 10 MW if burning time was enough. Secondly, the freight carriage model was designed referencing the real freight carriage's parameters. The standard combustible material cup boxes were used to simulate freight materials. Thirdly, the freight carriage model's HRR was tested using 10 MW cone calorimeter. Fitting curve was obtained based on t2 fire model. Further study should focus on freight carriage model's fire extinguishing experiment.