DBNet-based secondary circuit terminal strip wiring detection method in intelligent substation

A technology of intelligent substation and detection method, applied in image data processing, image enhancement, instruments, etc., to achieve the effect of balancing rapidity and robustness, and rapid statistical collection

CN113160196APending Publication Date: 2021-07-23SOUTHEAST UNIV
0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Publication Date
2021-07-23

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses a DBNet-based secondary circuit terminal strip wiring detection method in an intelligent substation. The detection method comprises the following steps: S1, carrying out standardized preprocessing on image data to zoom out a picture; S2, extracting a multi-channel feature map of a terminal strip target and a cable marker wiring target of the image to be detected; S3, obtaining a probability graph and an adaptive threshold graph of each target; S4, obtaining a terminal strip communication region and a number tube communication region, and solving each target boundary contour; and S5, simply describing the boundary contours of the terminal strip and the number tubes by using a rotary rectangular frame, and judging the result that the number tubes are connected to the terminal strip according to the overlapping area proportion. According to the detection method disclosed by the invention, the indefinite target shape detection of the cable marker and the terminal strip under the field irregular dense arrangement working condition is realized, the corresponding connection relation between the cable marker and the terminal strip is judged by fully utilizing the overlapping area and position relation, the rapidity and robustness of the detection method are effectively balanced, and the rapid statistical collection of the switching condition of the secondary circuit of the transformer substation is realized.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to the technical field of three-dimensional modeling and intelligent inspection of intelligent substations, in particular to a DBNet-based detection method for secondary circuit terminal strip wiring in intelligent substations. Background technique

[0002] The digital technology of three-dimensional modeling of intelligent substation can be regarded as an extension of substation monitoring and remote viewing system. Through the three-dimensional modeling of the secondary equipment and its connection circuits in the actual substation, a panoramic digital image display of all key equipment in the substation can be achieved, which is helpful for the status evaluation of the protection equipment in the substation. However, since the actual equipment of the secondary circuit is connected to the number pipe by using the terminal block, the identification of the terminal block connection in the field is still quite challenging. In the ...

Examples

Embodiment Construction

[0031] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0032] see Figure 1 to Figure 10 As shown, a DBNet-based secondary circuit terminal block wiring detection method in a smart substation, the detection method includes the following steps:

[0033] The detection method is divided into the pre-model training compression stage and the actual reasoning prediction stage.

[0034] 1. The specific implementation process of the pre-model training compression stage is as follows:

[0035] S1: Raw image dataset data augme...