Foreign objects intrusion into transmission lines can lead to serious troubles, using deep learning technology for foreign object detection has good performance and reduce losses. Due the complexity diversity of surrounding environment lines, limitations data acquisition methods, image invading used in current research are extremely rare, types background features single. Deep requires large as a driving force, Rare number images leads insufficient model fitting affects accuracy. We propose...

With the urgent global demand for sustainable development, intelligent education driven by multi-source physical information has attracted widespread attention as an innovative educational model. However, in context of dual carbon, achieving and efficient development faces many difficulties, one important challenges is how to effectively evaluate students. The application deep neural networks evaluation direction digitization. Currently, there need conduct research on value empowering with...

This paper introduces a fault diagnosis method for mine scraper conveyor gearbox gears using motor current signature analysis (MCSA). approach solves problems related to gear characteristics that are affected by coal flow load and power frequency, which difficult extract efficiently. A is proposed based on variational mode decomposition (VMD)-Hilbert spectrum ShuffleNet-V2. Firstly, the signal decomposed into series of intrinsic functions (IMF) VMD, sensitive parameters VMD optimized genetic...

Point cloud data can accurately and intuitively reflect the spatial relationship between coal wall underground fully mechanized mining equipment. However, indirect method of point feature extraction based on deep neural networks will lose some information cloud, while direct local cloud. Therefore, we propose use dynamic graph convolution network (DGCNN) to extract geometric features sphere in face (FMMF) order obtain position (marker) FMMF, thus providing a basis for subsequent...

With the aim of solving problem coal congestion caused by big blocks in underground mine scraper conveyors, this paper we proposed use a YOLO-BS (YOLO-Big Size) algorithm to detect abnormal phenomenon on conveyors. Given scale block targets, replaces last layer YOLOv4 feature extraction backbone network with transform module. The also deletes redundant branch which detects small targets PAnet module, reduces overall number parameters algorithm. As up-sampling and down-sampling operations...

Aiming at the problem of serious shutdowns conveyors caused by abnormal coal flow scraper conveyors, a monitoring method based on speckle structured light is proposed. The point cloud data body conveyor collected through acquisition system. Then, proposed PDS-Algorithm (Planar Density Simplification Algorithm) used to complete simplification and differentiation data, which provides basis for constructing geometric characteristics lineament. This paper uses processed calculate volume mass...

In the process of coal mining, a certain amount gas will be produced. Environmental perception is very important to realize intelligent and unmanned mine production operation reduce accident rate explosion other disasters. The identification geometric features working face main part environmental face. this study, we identify in large-scale point cloud (we take ball as an example) so provide method for On basis previous research, upgrade dynamic graph convolution neural network (DGCNN)...

The intelligent adjustment method of the shearer drum is key technology to improve level and safety degree fully mechanized mining face. This paper proposes a height model based on rough set significance attribute reduction (AR) fuzzy radial basis function neural network (FRRBFNN) optimized by adaptive immune genetic algorithm (AIGA). first selects parameters process monitoring importance set, establishes operation characteristic decision rule base. Next, determined space proposed. By...

The intersection line information of the point cloud between coal wall and roof can not only accurately reflect direction scraper conveyor but also provide a preliminary basis for realizing intelligent mine. However, indirect method using deep learning to segment mine working face cannot make full use rich provided by data. direct ignores local feature relationship points. Therefore, we propose dynamic graph convolution neural networks (DGCNNs) so as obtain them. First, in view...

The weigh-in-motion (WIM) system weighs the entire vehicle by identifying dynamic forces of each axle on road. load is very important to detect total weight vehicle. Different drivers have different driving behaviors, and when large trucks pass through weighing detection area, state may affect accuracy system. This paper proposes YOLOv3 network model as basis for this algorithm, which uses feature pyramid (FPN) idea achieve multi-scale prediction deep residual (ResNet) extract image...

The cooperative control of shearer and scraper conveyors is the prerequisite for realization intelligent comprehensive mining equipment unmanned workings. However, because harsh working face environment, complex process mining, many uncertainties, it difficult to establish a mathematical model precisely through operating mechanism. In era big data, data-driven has become popular trend. Therefore, according actual production this article proposed shearer–scraper conveyor based on rough set...

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, potential safety hazards. In view of weakness gear torsional load in current signal, this paper proposes using three-phase signal fusion to extract its phase difference information. At same time, order...

Aircraft failures can result in the leakage of fuel, hydraulic oil, or other lubricants onto runway during landing taxiing. Damage to fuel tanks oil lines hard landings accidents also contribute these spills. Further, improper maintenance operational errors may leave traces on before take-off after landing. Identifying spills airport videos is crucial flight safety and accident investigation. Advanced image processing techniques overcome limitations conventional RGB-based detection, which...

The integration of intelligent decision-making algorithms with urban cultural expression is becoming a hot topic in both academic and practical fields for exploring street landscapes. Exploring the application strategies regional landscape pattern recognition innovative design key step. single layout construction, deficiency, ecological imbalance, low resident participation seriously constrain overall quality improvement city. To address this dilemma, study delved into Kunming City selected...

Geometric features are an important factor for the classification of drugs and other transport objects in chemical reactors. The moving speed reactors is fast, it difficult to obtain their by imaging methods. In order avoid mistaken missed distribution objects, a method extracting geometric drug's point cloud reactor based on dynamic graph convolution neural network (DGCNN) proposed. this study, we first use MATLAB R2019a add random number noise points each file label cloud. Second,...

Abstract In the driving process, driver's visual attention area is of great significance to research intelligent decision-making behavior and dynamic behavior. Traditional driver intention recognition has problems such as large contact interference with wearing equipment, high false detection rate for drivers glasses strong light, unclear extraction field view. We use vision image taken by dash cam corresponding vehicle state data (steering wheel angle speed). Combined interpretability...

The simplification of three-dimensional (3D) models has always been a hot research topic for scholars. researchers simplified different parts the 3D point cloud data from both global and local information. Aiming at need to retain detailed features in models, neural network (NN) technology is firstly analyzed studied, algorithm regional segmentation geometric based on Graph Convolutional Neural Network (GCNN) proposed. Secondly, idea dense connection DenseNet structure, symmetric model...