Search

CN-120879395-B - Power transmission line external damage monitoring system based on machine vision

CN120879395BCN 120879395 BCN120879395 BCN 120879395BCN-120879395-B

Abstract

The invention relates to the technical field of electric wire monitoring, in particular to an electric transmission line external damage monitoring system based on machine vision, which comprises a data acquisition module, a region division module, a monitoring type division module and a resource allocation module, wherein the region division module is used for dividing a loop line risk region in a construction region based on a voltage level, a space trend track and an electric transmission line height of an electric transmission line, the monitoring type division module is used for calculating a construction region risk tendency value based on an average transverse overlapping area and an average longitudinal invasion depth of a dynamic target in the loop line risk region and dividing a monitoring type of the construction region based on the construction region risk tendency value, and the resource allocation module is used for determining an allocable monitoring resource allocation method of the construction region based on the monitoring type of the construction region and/or the average motion change degree of the dynamic target in the loop line risk region.

Inventors

  • ZHAO YONG

Assignees

  • 华普灵科(天津)科技有限公司

Dates

Publication Date
20260508
Application Date
20250714

Claims (10)

  1. 1. Machine vision-based power transmission line outward breaking monitoring system is characterized by comprising: the data acquisition module is used for acquiring power transmission parameter data, distribution position data and construction data of a power transmission line in a construction area; The region dividing module is connected with the data acquisition module and used for dividing a loop line risk region in a construction region based on the voltage level, the space trend track and the height of the transmission line; The monitoring type dividing module is connected with the area dividing module and is used for calculating a construction area risk trend value based on the average transverse superposition area and the average longitudinal invasion depth of the dynamic target in the loop risk area and dividing the monitoring type of the construction area based on the construction area risk trend value; the resource allocation module is connected with the monitoring type dividing module and is used for determining an allocable monitoring resource allocation method of the construction area based on the monitoring type of the construction area and/or the average motion change degree of the dynamic targets in the loop risk area, wherein the allocable monitoring resource allocation method comprises the steps of preferentially allocating a first dynamic target in the loop risk area and preferentially allocating a second dynamic target in the non-loop risk area; The monitoring module is respectively connected with the data acquisition module, the region division module, the monitoring type division module and the resource allocation module and is used for determining whether to send out an external power transmission line damage early warning based on whether the motion trail of the dynamic target is overlapped with the protection region of the power transmission line; The method comprises the steps of acquiring continuous frame images of a dynamic target in a loop risk area based on machine vision, extracting a two-dimensional outline of the dynamic target in each frame image through a target detection algorithm, calculating an overlapping area of the two-dimensional outline of the dynamic target and a transverse boundary of the loop risk area under the same coordinate system to obtain a transverse overlapping area of the dynamic target and the loop risk area in a single frame image, taking an arithmetic average value of the transverse overlapping areas of all single frame images within preset time to obtain the average transverse overlapping area, acquiring three-dimensional coordinate data of the dynamic target in the loop risk area based on a laser radar or a depth camera to determine the foremost position coordinate of the dynamic target along the axial direction of the line, calculating the distance between the foremost position coordinate and the inner boundary coordinate of the loop risk area along the axial direction to serve as the longitudinal invasion depth of the single-moment dynamic target, and taking an arithmetic average value of the longitudinal invasion depth at all moments within the preset time to obtain the average longitudinal invasion depth.
  2. 2. The machine vision-based power transmission line outward breaking monitoring system according to claim 1, wherein the region dividing module uses a space trend track of a power transmission line as a dynamic reference axis, extends outward to a preset safety distance threshold along a radial direction of the dynamic reference axis, and covers a preset length range along an axial direction of the reference axis to form a three-dimensional tubular region, namely a loop risk region.
  3. 3. The machine vision based power transmission line outward-break monitoring system of claim 2, wherein the monitoring type classification module calculates a construction area risk tendency value based on a weighted sum of an average lateral overlap area and an average longitudinal intrusion depth of dynamic objects within a loop risk area, and classifies a monitoring type of the construction area based on the construction area risk tendency value, wherein, If the risk trend value of the construction area is larger than or equal to the risk trend value of the preset construction area, the monitoring type dividing module determines that the monitoring type of the construction area is a high risk trend type; and if the risk trend value of the construction area is smaller than the risk trend value of the preset construction area, the monitoring type dividing module determines that the monitoring type of the construction area is a low risk trend type.
  4. 4. The machine vision based power transmission line outward breaking monitoring system according to claim 3, wherein the resource allocation module determines an allocatable monitoring resource allocation method of the construction area based on a monitoring type of the construction area and/or an average motion variation degree of a dynamic target in the loop risk area, wherein, If the monitoring type of the construction area is a high risk tendency type or the average motion change degree of the dynamic targets in the loop risk area is larger than the preset change degree, the resource allocation module determines to allocate the first dynamic targets in the loop risk area preferentially; And if the monitoring type of the construction area is a low risk tendency type and the average motion change degree of the dynamic targets in the loop risk area is smaller than or equal to the preset change degree, the resource allocation module determines to allocate the second dynamic targets in the non-loop risk area preferentially.
  5. 5. The machine vision based power transmission line outward breaking monitoring system according to claim 4, wherein the first dynamic target is a dynamic target in a loop circuit risk area, and the second dynamic target is a dynamic target in a non-loop circuit risk area, which meets the power transmission line outward breaking condition; The transmission line external breaking condition is that a space envelope surface formed by the allowable motion track of the dynamic target and a space trend track of the transmission line are intersected.
  6. 6. The machine vision based power transmission line outward breaking monitoring system of claim 5, wherein the resource allocation module determines the degree of motion change of the dynamic object based on the direction change rate and the speed fluctuation amplitude of the dynamic object.
  7. 7. The machine vision based power transmission line outward breaking monitoring system of claim 6, wherein the allocable monitoring resources comprise a set of monitoring hardware devices in an unsaturated operating state, remaining available computing power, unoccupied communication transmission bandwidth, and free storage capacity.
  8. 8. The machine vision based power transmission line outward breaking monitoring system of claim 6, wherein the resource allocation module determines the priority of the allocatable monitoring resource allocation object based on the rate of change of the motion trajectory of the first dynamic target.
  9. 9. The machine vision based power transmission line outward-break monitoring system of claim 6, wherein the resource allocation module determines the priority of the allocatable monitoring resource allocation object based on the rate of change of the working form of the second dynamic target.
  10. 10. The machine vision based power transmission line outward-break monitoring system of claim 9, wherein the monitoring module determines whether to issue a power transmission line outward-break warning based on whether a motion trajectory of a dynamic target overlaps a protection region of the power transmission line, wherein, And if the motion trail of the dynamic target is overlapped with the protection area of the power transmission line, the monitoring module determines to send out the early warning of the external damage of the power transmission line.

Description

Power transmission line external damage monitoring system based on machine vision Technical Field The invention relates to the technical field of electric wire monitoring, in particular to an electric transmission line external damage monitoring system based on machine vision. Background With the continuous expansion of the scale of the power system, the power transmission line is used as a core carrier for energy transmission, and the safe operation of the power transmission line directly relates to the stability of a power grid. However, the transmission line is easily affected by external construction activities, so that an external accident is caused, the existing system mostly adopts a fixed radius to define a dangerous area, parameters such as a voltage level of an unbonded circuit, equipment height and the like are dynamically adjusted, monitoring resources (such as cameras and calculation power) are allocated in a lack of pertinence, monitoring hysteresis is possibly caused by insufficient resources in a high-risk area, and resource waste exists in a low-risk area. For example, CN117237363A discloses a method, a system, a medium and equipment for identifying external broken sources of a power transmission line, which comprise the steps of obtaining an image of the running environment of the power transmission line, extracting deep network image features of the obtained image, determining candidate target areas of the obtained image, detecting and identifying the determined candidate target areas by adopting a preset external broken source identification model to complete the external broken source identification of the power transmission line, wherein the preset external broken source identification model adopts a deep convolutional neural network which introduces an attention mechanism, perceives feature weights of the deep network image features in the determined candidate target areas, and detects external broken source hidden dangers in the identification image. According to the intelligent external broken source identification method based on machine vision, the image of the running environment of the power transmission line is monitored in real time, the potential safety hazard of the external broken source is found in time, and the running safety and reliability of the power transmission line are improved. However, the prior art has the problems that the classification of the risk of the external damage of the power transmission line is not fine enough, the judgment of the stability of the target movement is not enough, the monitoring strategy cannot be dynamically adjusted according to the movement characteristics, the monitoring resource allocation is unreasonable, and false alarm or missing alarm is easy to occur. Disclosure of Invention Therefore, the invention provides a machine vision-based power transmission line outward breaking monitoring system, which is used for solving the problems that in the prior art, the classification of the power transmission line outward breaking risk is not fine enough, the judgment of the target movement stability is insufficient, the monitoring strategy cannot be dynamically adjusted according to the movement characteristics, the monitoring resource allocation is unreasonable, and false alarm or missing alarm is easy to occur. In order to achieve the above object, the present invention provides a machine vision-based power transmission line outward breaking monitoring system, comprising: the data acquisition module is used for acquiring power transmission parameter data, distribution position data and construction data of a power transmission line in a construction area; The region dividing module is connected with the data acquisition module and used for dividing a loop line risk region in a construction region based on the voltage level, the space trend track and the height of the transmission line; The monitoring type dividing module is connected with the area dividing module and is used for calculating a construction area risk trend value based on the average transverse superposition area and the average longitudinal invasion depth of the dynamic target in the loop risk area and dividing the monitoring type of the construction area based on the construction area risk trend value; the resource allocation module is connected with the monitoring type dividing module and is used for determining an allocable monitoring resource allocation method of the construction area based on the monitoring type of the construction area and/or the average motion change degree of the dynamic targets in the loop risk area, wherein the allocable monitoring resource allocation method comprises the steps of preferentially allocating a first dynamic target in the loop risk area and preferentially allocating a second dynamic target in the non-loop risk area; The monitoring module is respectively connected with the data acquisition module, the area dividing module, the monitor