Geometric Feature Extraction of Point Cloud of Chemical Reactor Based on Dynamic Graph Convolution Neural Network

Zhao, Shuanfeng; Guo, Wei; Guo, Xiaojun

doi:10.1021/acsomega.1c02213

Cited by 4 publications

(2 citation statements)

References 43 publications

(66 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Advances have been made in graph neural networks (GNNs), which have greater expressive and representational power than traditional convolutional neural network architectures [4]. Another promising approach that leveraged feature extraction of molecules was introduced by Xing et al [5] proposed and validated a method for geometric feature extraction from drug point clouds using DGCNN [6], addressing challenges in existing methods and emphasizing the importance of point cloud data in chemical reactor management. The introduction of DGCNN enhances precision, recall, F1 score, and accuracy, demonstrating its superiority over traditional methods like PointNet [7] and PointNet++ [8].…”

Section: Introductionmentioning

confidence: 99%

Geom-SAC: Geometric Multi-Discrete Soft Actor Critic With Applications in De Novo Drug Design

Abdallah,

Adel,

Elkerdawy

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Finding new molecules with desirable properties has high computational and overhead costs. Much research has focused on generating candidate molecules in one-and two-dimensional spaces, which has produced some favorable results. However, extending these approaches to molecules in threedimensional space would be far more useful because the representation of molecules is more realistic, although three-dimensional methods have much higher computational costs. In this work, we developed a geometric deep reinforcement learning agent that generates and optimizes molecules that could interact with a biochemical target. The agent can be used for generating molecules from scratch or for lead optimization when it enhances the properties of a given molecule, whether by enhancing its drug-likeness or increasing its activity toward the target via implicit learning. Thus, the agent works with molecules in three-dimensional space without high computational costs.INDEX TERMS Molecule generation, molecule optimization, geometric deep learning, reinforcement learning, de novo drug design.

show abstract

Section: Introductionmentioning

confidence: 99%

Geom-SAC: Geometric Multi-Discrete Soft Actor Critic With Applications in De Novo Drug Design

Abdallah,

Adel,

Elkerdawy

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…With the increasingly complex battlefield environment in modern warfare, how to accurately identify the target in the battlefield environment is a difficult problem. The 2D image of the target obtained by the traditional visible-light imaging technology can only provide a small amount of target information [1], which is difficult to meet the needs in many cases, while the point cloud data obtained by the 3D imaging technology based on laser scanning can provide more target information, such as spatial coordinates, color, intensity et al [2][3][4]. so the target recognition problem based on point cloud data has received extensive attention [5][6][7].…”

Section: Introductionmentioning

confidence: 99%

Point Cloud Target Registration Algorithm Based on Local Feature Description

Wang

Zha

2023

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

Point cloud data has received extensive attention in the field of target recognition and target tracking due to its rich target information, and point cloud registration is one of the basic tasks in point cloud applications. Aiming at the problem that the feature descriptor in the traditional registration algorithm is complicated to calculate and cannot sufficiently describe the features of the point cloud, this paper proposes a local feature descriptor that integrates the local normal vector angle features, projection distance features, and distribution density features of the point cloud. The feature similarity is introduced to determine the corresponding point pairs between the template point cloud and the target point cloud, and the point cloud registration task is completed based on local feature matching. Experiments show that the feature descriptor in this paper is simple to calculate and can sufficiently describe the local features of the point cloud, a good correspondence between points can be established based on the feature descriptor, and the algorithm has good robustness. The registration algorithm proposed in this paper can provide a reference for the fields of target recognition and target pose estimation, etc.

show abstract

Multi-fault recognition of gear based on wavelet image fusion and deep neural network

Zhao

Guo

et al. 2021

AIP Advances

View full text Add to dashboard Cite

The coal mining environment where the plate conveyor is located often has narrow space, violent mechanical vibration, and explosion-proof requirements. Therefore, collecting vibration signals by installing sensors will have adverse problems such as difficult installation, strong noise, and potential safety hazards. In view of the weakness of the gear torsional load in the current signal, this paper proposes using three-phase current signal fusion to extract its phase difference information. At the same time, in order to extract the current information and phase information change caused by the early fault of the scraper conveyor gear, a gear fault diagnosis method based on the deep convolution neural network and three-phase current continuous wavelet image fusion is proposed. This method transforms the gear fault diagnosis problem into an image analysis problem. By fusing the time-frequency images of three-phase current, the phase difference information of the image can be obtained, and then the fluctuation state of motor torque can be determined. Then, the deep convolution neural network model is built to realize the fault feature recognition of the wavelet fusion image.

show abstract

Geometric Feature Extraction of Point Cloud of Chemical Reactor Based on Dynamic Graph Convolution Neural Network

Cited by 4 publications

References 43 publications

Geom-SAC: Geometric Multi-Discrete Soft Actor Critic With Applications in De Novo Drug Design

Geom-SAC: Geometric Multi-Discrete Soft Actor Critic With Applications in De Novo Drug Design

Point Cloud Target Registration Algorithm Based on Local Feature Description

Multi-fault recognition of gear based on wavelet image fusion and deep neural network

Contact Info

Product

Resources

About