Intelligent Detection Method for Welding Seam Defects of Automobile Wheel Hub Based on YOLO

Author(s):  
Xiufeng Zhang ◽  
Chen Wang ◽  
Changfeng Xiang ◽  
Chao Liu ◽  
Yu Li
2014 ◽  
Vol 530-531 ◽  
pp. 646-649
Author(s):  
Ling Qiu ◽  
Cai Ming Liu

To dynamically discover network attacks hidden in network data, an intelligent detection method for network security is proposed. Biological immune principles and mechanisms are adopted to judge whether network data contain illegal network packets. Signature library of network attacks and section library of attack signatures are constructed. They store attack signatures and signature sections, respectively. They are used to make the initial detection ability of proposed method. Detectors are defined to simulate immune cells. They evolve dynamically to adapt the network security. Signatures of network data are extracted from IP packets. Detectors match network data's signatures which mean some attacks. Warning information is formed and sent to network administrators according to recognized attacks.


2013 ◽  
Vol 380-384 ◽  
pp. 3882-3885
Author(s):  
Xiaoan Yang

Using motion state of the equipment transducer to determine the presence of a weak signal is a common method of signal detection, whose core is to determine the system's phase change. There a many traditional ways to judge phase transition, but most of which have computational complexity and need a large amount of data which make them difficult to apply engineering practices. In order to solve these problems, this paper presents a detection method based on Lyapunov exponent classification with a small amount of data. This approach has some advantages such as requiring fewer observed values, small calculation amount, and able to automatically determine the phase transition without subjective factors involved etc. Experiments show that this method has stable performance, high effectiveness, strong practicality and promotion.


Author(s):  
Jun Chen

When the unmanned vehicle is disturbed by the outside world or carries out dangerous actions such as steering and continuous lane changing, the yaw stability of the unmanned vehicle decreases and the dangerous situation such as rollover is easy to occur. In this paper, the intelligent detection method for roll stability of unmanned vehicles based on fuzzy control is studied. The roll control system of the unmanned vehicle based on a double-layer control strategy is designed. The roll stability of the unmanned vehicle is controlled by an upper-layer fuzzy controller and lower-layer differential braking control. The dynamic model and tire model are built in MATLAB/Simulink to restore the running characteristics of unmanned vehicles. Based on the operation characteristics, the roll stability of the unmanned vehicle’s roll control system based on fuzzy control is tested from three aspects: steady-state response, roll stability and dynamic stability coefficient. The experimental results show that the transverse load’s transfer rate of the proposed method is reduced by more than 0.2% compared with the contrast method, the yaw angular velocity, centroid’s roll angle and roll angle measured under the two working conditions are closer to the actual values, which shows that the method has better control effect and detection accuracy.


Sign in / Sign up

Export Citation Format

Share Document