Aircraft anomaly detection system and abnormality detection method thereof
An anomaly detection and aircraft technology, applied in the direction of instruments, biological neural network models, character and pattern recognition, etc., can solve problems such as differences and prone to missed detection, and achieve the effect of simplifying the backbone network, simplifying the model, and reducing training time
Pending Publication Date: 2020-02-21
吉林省民航机场集团有限公司
View PDF11 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0002] The "Informatization Civil Aircraft Maintenance Information System" disclosed in the Chinese Patent Bulletin, the system stores the fault information of all airports and the corresponding maintenance method information in the database, and each airport can quickly obtain previous troubleshooting by consulting the database through the client experience to improve the speed of troubleshooting; at the same time, the new troubleshooting experience of the airport can be stored in the database to share with other airports; although the system can improve the timeliness of information, aircraft abnormalities need to be judged subjectively by technicians, and the disadvantage is that different release personnel There are large differences in the inspection results of the same aircraft, and it is very easy to miss inspections only by visual inspection
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0023] Such as figure 1 As shown, the aircraft anomaly detection system of the present invention includes a feature extraction model, an improved YOLOv3 detector and a result visualization module.
[0024] The feature extraction model includes three layers. The first and second layers each include a convolution operation unit, a maximum pooling unit, and a nonlinear transformation unit. The third layer contains only one convolution operation unit; The convolution operation unit contains 32 3×3 convolution kernels, the second and third layer convolution operation units contain 64 3×3 convolution kernels; the first and second layers of the largest pooling unit use 2 ×2 pooling window, the nonlinear transformation unit uses ReLU as the nonlinear activation function for nonlinear transformation.
[0025] The improved YOLOv3 detector generates three small-size anchor frames for small targets through a clustering algorithm for target detection to obtain fault coordinates, fault types, an...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention relates to an aircraft anomaly detection system. The system comprises a feature extraction model and an improved YOLOv3 detector. The feature extraction model comprises three layers, each of the first layer and the second layer comprises a convolution operation unit, a maximum pooling unit and a nonlinear transformation unit, and the third layer only comprises one convolution operation unit. The convolution operation unit of the first layer comprises 32 3 * 3 convolution kernels, and the convolution operation units of the second layer and the third layer comprise 64 3 * 3 convolution kernels. Maximum pooling units of the first layer and the second layer both adopt 2 * 2 pooling windows, and nonlinear transformation units both adopt ReLUs as nonlinear activation functions to carry out nonlinear transformation. The improved YOLOv3 detector generates three anchor frames through a clustering algorithm for a small target image sample to perform target detection, and fault coordinates, fault types and fault existence probability information are obtained. According to the invention, the model operation efficiency and the detection accuracy are improved.
Description
Technical field [0001] The invention belongs to the technical field of aircraft abnormality detection, and relates to an aircraft abnormality detection system. Background technique [0002] The "Information Civil Aircraft Maintenance Information System" disclosed in the China Patent Gazette. This system stores all the information on the failures and the corresponding maintenance methods that occur at all airports in the database. Each airport can quickly obtain the troubleshooting information from the database through the client terminal. Experience, improve the troubleshooting speed; at the same time, the new troubleshooting experience of the airport can be stored in the database to share with other airports; although the system can improve the timeliness of information, the aircraft abnormality needs to be subjectively judged by the technicians, and the disadvantage is that different release personnel The inspection results of the same aircraft are quite different, and it is ea...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62G06N3/04
CPCG06N3/048G06F18/23G06F18/214G06F18/24
Inventor 朱建东刘圣禹黄毅鹏
Owner 吉林省民航机场集团有限公司