Unlock instant, AI-driven research and patent intelligence for your innovation.
Transformer assembly identification method based on submersible robot
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A recognition method and robot technology, applied in computer parts, neural learning methods, character and pattern recognition, etc., can solve problems such as low-level integration of manual inspection functions, and achieve the effect of promoting development, improving local contrast, and improving accuracy
Pending Publication Date: 2022-03-01
SHANGHAI JIAO TONG UNIV +1
View PDF0 Cites 0 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
[0002] The inspection of high-voltage equipment is crucial to the reliability of power supply. However, the contradiction between inspection workload and human resources has become increasingly prominent. At present, manual inspection is still the mainstream, but manual inspection is prone to blind spots and low-level integration problems of functions. Although some substations have Inspection robots are used, but manual analysis is still the main method of image processing. The intelligence and comprehensiveness of inspections need to be improved, especially in the field of power transformer inspections. It is necessary to check whether the design of the transformers meets the requirements. If it is possible to use submersible robots to take pictures Transformer internal photos, and automatically analyze these photos, find the corresponding parts, and further determine the equipment status, can greatly reduce the burden of inspection personnel, and inspection personnel will be changed from on-site operation to management work, which will greatly improve the current O&M level, however, due to large-scale movement of submersible robots will cause changes in lighting and viewing angles, these issues are still unresolved
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 more
Image
Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0057] refer to Figure 1-Figure 2 , the present embodiment provides a method for identifying transformer components based on an oil submersible robot, including:
[0058] image enhancement;
[0059] Component identification.
[0060] Further, image enhancement includes:
[0061] Improve the contrast of the original image, remove irrelevant features, and improve the contrast of the original image including enhancing the local contrast.
[0062] Further, the formula for enhancing local contrast is as follows:
[0064] Among them, (m, n) is the pixel position, the original gray level is g(m, n), and the RGB value corresponding to the pixel position (m, n) is R(m, n), G(m, n), B (m, n),
[0065]
[0066]
[0067] Among them, θ is the threshold value, is the average gray level in the window, the center of the window is (m, n), and the window size is M×N.
[0068] Further, enhancing the local contrast also includ...
Embodiment 2
[0097] refer to Figure 3-Figure 4 , this embodiment provides a transformer component recognition method based on the submersible robot, and compares this method, other commonly used object detection methods Faster R-CNN, YOLO and SSD with the internal image of the transformer taken by the submersible robot.
[0098] 1. Dataset
[0099] The dataset includes 770 internal pictures of a single transformer taken by an oil submersible robot. All pictures have been manually checked and annotated. There are several types of components: bolts, leads, wooden sleepers, windings. Typical scenes in the dataset are illustrated as image 3 shown.
[0100] 2. The results of this method
[0101] The result of this method is in Figure 4 given in, such as Figure 4 As shown, all detected objects have a bounding box representing their position and size. Table 1 shows the accuracy of this method. It can be seen that this method has a very high accuracy rate for various components. In additio...
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 discloses a transformer component identification method based on a submersible robot, which comprises the steps of image enhancement and component identification, the image enhancement comprises local contrast enhancement, local contrast normalization and image RGB information recovery, and the component identification further comprises loss function calculation. The method has the advantages that a framework is provided by combining image enhancement and a transformer network, the illumination change problem of transformer internal component recognition is solved, the central idea of image enhancement is to improve the local contrast of the image, the core of component recognition is to mine context information in a transformer internal scene picture, and the recognition accuracy of the transformer internal component recognition is improved. Under the framework, the accuracy of component identification is improved, the framework has higher expandability, flexibility and efficiency, and the development of automatic monitoring of power equipment is greatly promoted.
Description
technical field [0001] The technical field of transformer component recognition related to the present invention, in particular, relates to a transformer component recognition method based on an oil submersible robot. Background technique [0002] The inspection of high-voltage equipment is crucial to the reliability of power supply. However, the contradiction between inspection workload and human resources has become increasingly prominent. At present, manual inspection is still the mainstream, but manual inspection is prone to blind spots and low-level integration problems. Although some substations have Inspection robots are used, but manual analysis is still the main method of image processing. The intelligence and comprehensiveness of the inspection needs to be improved, especially in the field of power transformer inspection. It is necessary to check whether the design of the transformer meets the requirements. Transformer internal photos, and automatically analyze the...
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
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to View More 
Login to View More