1
|
Application of Lightweight Convolutional Neural Network for Damage Detection of Conveyor Belt. APPLIED SCIENCES-BASEL 2021. [DOI: 10.3390/app11167282] [Citation(s) in RCA: 4] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 11/16/2022]
Abstract
Aiming at the problem that mining conveyor belts are easily damaged under severe working conditions, the paper proposed a deep learning-based conveyor belt damage detection method. To further explore the possibility of the application of lightweight CNNs in the detection of conveyor belt damage, the paper deeply integrates the MobileNet and Yolov4 network to achieve the lightweight of Yolov4, and performs a test on the exiting conveyor belt damage dataset containing 3000 images. The test results show that the lightweight network can effectively detect the damage of the conveyor belt, with the fastest test speed 70.26 FPS, and the highest test accuracy 93.22%. Compared with the original Yolov4, the accuracy increased by 3.5% with the speed increased by 188%. By comparing other existing detection methods, the strong generalization ability of the model is verified, which provides technical support and empirical reference for the visual monitoring and intelligent development of belt conveyors.
Collapse
|
2
|
Abstract
Conveyors are one of the important components of transport systems and are used in almost all branches of mechanical engineering. This paper investigates the dynamics of the intermittent motion conveyor mechanical system. The mechanical transmission is a planetary mechanism with elliptical gears, in which the intermittent motion of the output shaft is provided by a variable gear ratio of non-circular gears. A single-mass dynamic model is built by reducing the masses, forces and moments to the initial link, which is the input shaft of the mechanism. The solutions of the equations of initial link motion were obtained using two methods, the energy-mass method and the third-order Hermite method. Dynamic studies by the energy-mass method made it possible to determine flywheel moment of inertia to reduce the coefficient of initial link rotation irregularity. The convergence of the functions of the initial link angular velocity obtained by both methods was confirmed. The results can be used for further force analysis, strength calculations, design and manufacture of the conveyor.
Collapse
|