Distribution network contact status alarm system

Abstract

The present invention relates to a distribution network touch state alarm system, comprising: a detachment judgment device, which is used to issue a first judgment instruction when it is judged that the conductor side insulator of the pole tower of the distribution network line has not fallen off, otherwise, it issues a second judgment instruction; a touch alarm mechanism, which is used to execute an alarm action corresponding to the illegal touching of the insulator by the operating electrician before the conductor side insulator is detached when the convolutional neural network model is used to analyze the handover between the insulator and the operating electrician. The distribution network touch state alarm system of the present invention is intelligently designed, safe and reliable. Since it can intelligently analyze whether the insulator and the operating electrician have been handed over based on various on-site visual information, and when the handover between the insulator and the operating electrician is intelligently analyzed, the alarm action corresponding to the illegal touching of the insulator by the operating electrician before the conductor side insulator is detached is executed, thereby providing safety warning information for the safe operation of the operating electrician.

Description

Technical Field

[0001] The present invention relates to the field of power distribution networks, and in particular to a power distribution network touch status alarm system. Background Art

[0002] Artificial neural networks are nonlinear, adaptive information processing systems composed of a large number of interconnected processing units. They were proposed based on the research findings of modern neuroscience and attempt to process information by simulating the way brain neural networks process and memorize information. Artificial neural networks have four basic characteristics:

[0003] (1) Nonlinearity: Nonlinear relationships are a universal characteristic of nature. The brain's intelligence is a nonlinear phenomenon. Artificial neurons are in two different states: activated or inhibited. This behavior is mathematically represented as a nonlinear relationship. Networks composed of neurons with thresholds have better performance and can improve fault tolerance and storage capacity.

[0004] (2) Non-localization: A neural network is typically composed of multiple neurons that are extensively connected. The overall behavior of a system depends not only on the characteristics of individual neurons but may also be primarily determined by the interactions and connections between units. The non-localization of the brain can be simulated through the large number of connections between units. Associative memory is a typical example of non-localization.

[0005] (3) Very qualitative: Artificial neural networks have the ability to adapt, self-organize, and self-learn. Not only can the information processed by the neural network be subject to various changes, but the nonlinear dynamic system itself is also constantly changing while processing information. Iterative processes are often used to describe the evolution of dynamic systems.

[0006] (4) Non-convexity: The evolutionary direction of a system, under certain conditions, will depend on a specific state function. For example, the energy function, whose extreme values ​​correspond to relatively stable states of the system. Non-convexity means that such a function has multiple extreme values, so the system has multiple relatively stable equilibrium states, which will lead to the diversity of system evolution.

[0007] However, the various neural network models existing in the prior art cannot effectively identify whether the insulator and the operating electrician have been handed over, and thus cannot perform intelligent analysis and on-site alarms on the operating electrician's safe working environment. For example, when using a neural network model to intelligently analyze the handover between the insulator and the operating electrician, it is impossible to execute the alarm action corresponding to the operating electrician illegally touching the insulator before the insulator on the conductor side is detached. Summary of the Invention

[0008] In order to solve the technical problems in the prior art, the present invention provides a distribution network touch status alarm system, which can use a convolutional neural network model to analyze whether the insulator and the operating electrician are handed over based on the overall depth of field value and the number of pixels occupied by the image area of ​​the insulator in multiple mapped images, the overall depth of field value and the number of pixels occupied by the image area of ​​the operating electrician in multiple mapped images, the coordinate data corresponding to each pixel occupied by the image area of ​​the insulator in multiple mapped images, and the coordinate data corresponding to each pixel occupied by the image area of ​​the operating electrician in multiple mapped images, thereby completing an intelligent analysis of whether the insulator and the operating electrician are handed over, and introducing a touch alarm mechanism for executing an alarm action corresponding to the operating electrician illegally touching the insulator before the conductor-side insulator is detached when the convolutional neural network model is used to intelligently analyze the handover between the insulator and the operating electrician, and suspending the execution of the alarm action corresponding to the operating electrician illegally touching the insulator before the conductor-side insulator is detached when the convolutional neural network model is used to intelligently analyze the handover between the insulator and the operating electrician, thereby providing safety warning information for the safe operation of the operating electrician.

[0009] According to the present invention, a power distribution network touch status alarm system is provided, the system comprising:

[0010] a separation judgment device, configured to issue a first judgment instruction when it is determined that the conductor-side insulator of the tower of the power distribution network line has not fallen off, and further configured to issue a second judgment instruction when it is determined that the conductor-side insulator of the tower of the power distribution network line has fallen off;

[0011] an imaging processing device connected to the separation judgment device, and configured to perform imaging processing on the tower working scene upon receiving the first judgment instruction, so as to obtain and output a corresponding working scene image;

[0012] a multi-mapping device connected to the imaging processing device and comprising a signal filtering device, an affine transformation device, and a cubic interpolation device, for sequentially performing salt and pepper noise filtering, affine transformation, and cubic polynomial interpolation on the received image to obtain and output a corresponding multi-mapping image;

[0013] a first analyzing mechanism, connected to the multi-mapping device, configured to identify an image region of an insulator in the multi-mapping image based on a grayscale value interval corresponding to the insulator, and to identify an image region of an operating electrician in the multi-mapping image based on a reference outline pattern corresponding to the operating electrician, wherein the grayscale value interval corresponding to the insulator is within a range of 0-255, and the reference outline pattern corresponding to the operating electrician is in a JPEG encoding format;

[0014] a second analysis unit, connected to the first analysis unit, configured to analyze whether the insulator and the operating electrician are in contact with each other using a convolutional neural network model based on the overall depth of field value and the number of pixels occupied by the image area of ​​the insulator in the multiple mapped images, the overall depth of field value and the number of pixels occupied by the image area of ​​the operating electrician in the multiple mapped images, the coordinate data corresponding to each pixel occupied by the image area of ​​the insulator in the multiple mapped images, and the coordinate data corresponding to each pixel occupied by the image area of ​​the operating electrician in the multiple mapped images;

[0015] a touch alarm mechanism, connected to the second analysis mechanism, for executing an alarm action corresponding to the operating electrician illegally touching the insulator before the conductor-side insulator is detached, when the second analysis mechanism uses a convolutional neural network model to analyze the handover between the insulator and the operating electrician;

[0016] The convolutional neural network model is a convolutional neural network that has completed a set number of learning cycles, and the number of learning cycles is monotonically positively correlated with the mean of the number of pixel rows and the number of pixel columns of the multiple mapped images.

[0017] Among them, when it is determined that the conductor side insulator of the pole tower of the distribution network line has not fallen off, a first judgment instruction is issued, and it is also used to issue a second judgment instruction when it is determined that the conductor side insulator of the pole tower of the distribution network line has fallen off, including: judging whether the conductor side insulator of the pole tower of the distribution network line has fallen off based on the sensor data of the height sensor of the conductor side insulator.

[0018] Therefore, the present invention has at least the following beneficial technical effects:

[0019] First, the image area of ​​the insulator in the multiple mapped images is identified based on the grayscale value range corresponding to the insulator, and the image area of ​​the working electrician in the multiple mapped images is identified based on the reference outline pattern corresponding to the working electrician. The grayscale value range corresponding to the insulator is between 0-255, and the reference outline pattern corresponding to the working electrician is in JPEG encoding format, thereby providing key information for subsequent intelligent analysis.

[0020] Secondly: based on the overall depth of field value and the number of pixels occupied by the image area of ​​the insulator in the multiple mapped images, the overall depth of field value and the number of pixels occupied by the image area of ​​the working electrician in the multiple mapped images, the coordinate data corresponding to each pixel occupied by the image area of ​​the insulator in the multiple mapped images, and the coordinate data corresponding to each pixel occupied by the image area of ​​the working electrician in the multiple mapped images, a convolutional neural network model is used to analyze whether the insulator and the working electrician have been handed over, thereby completing an intelligent analysis of whether the insulator and the working electrician have been handed over;

[0021] Thirdly: the convolutional neural network model that performs intelligent analysis is a convolutional neural network that has completed a set number of learning cycles, and the number of learning cycles is monotonically positively correlated with the mean of the number of pixel rows and pixel columns of the multiple mapped images, thereby achieving customization of the structure of the convolutional neural network model;

[0022] Finally: A touch alarm mechanism is introduced to execute the alarm action corresponding to the working electrician illegally touching the insulator before the insulator on the conductor side is detached when the convolutional neural network model is used to intelligently analyze the handover between the insulator and the working electrician; and to temporarily suspend the execution of the alarm action corresponding to the working electrician illegally touching the insulator before the insulator on the conductor side is detached when the convolutional neural network model is used to intelligently analyze the handover between the insulator and the working electrician, thereby providing safety warning information for the safe operation of the working electrician.

[0023] The present invention's distribution network contact status alarm system features an intelligent, safe, and reliable design. It uses visual on-site information to intelligently analyze whether insulators and operating electricians have been handed over. Upon intelligent analysis of handover, it triggers an alarm if the operating electrician violates regulations by touching the insulator before the conductor-side insulator is detached, providing safety warnings for the electrician's operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The embodiments of the present invention will be described below with reference to the accompanying drawings, in which:

[0025] Figure 1 The figure is a schematic diagram of the internal structure of a distribution network touch status alarm system according to an insulator embodiment of the present invention.

[0026] Figure 2 Schematic diagram of the internal structure of a distribution network touch status alarm system according to an embodiment of the present invention.

[0027] Figure 3 Schematic diagram of the internal structure of a power distribution network touch status alarm system according to embodiment C of the present invention. DETAILED DESCRIPTION

[0028] The following is a detailed description of an embodiment of the power distribution network touch status alarm system of the present invention with reference to the accompanying drawings.

[0029] Figure 1 This is a schematic diagram of the internal structure of a distribution network touch status alarm system according to an embodiment of the insulator of the present invention. The system includes:

[0030] a separation judgment device, configured to issue a first judgment instruction when it is determined that the conductor-side insulator of the tower of the power distribution network line has not fallen off, and further configured to issue a second judgment instruction when it is determined that the conductor-side insulator of the tower of the power distribution network line has fallen off;

[0031] an imaging processing device connected to the separation judgment device, and configured to perform imaging processing on the tower working scene upon receiving the first judgment instruction, so as to obtain and output a corresponding working scene image;

[0032] For example, the imaging processing device is connected to the separation judgment device and is configured to perform imaging processing on the tower working scene upon receiving the first judgment instruction to obtain and output a corresponding working scene image, including: the imaging processing device includes a photoelectric sensing component, which is configured to perform imaging processing on the tower working scene based on photoelectric sensing;

[0033] a multi-mapping device connected to the imaging processing device and comprising a signal filtering device, an affine transformation device, and a cubic interpolation device, for sequentially performing salt and pepper noise filtering, affine transformation, and cubic polynomial interpolation on the received image to obtain and output a corresponding multi-mapping image;

[0034] a first analyzing mechanism, connected to the multi-mapping device, configured to identify an image region of an insulator in the multi-mapping image based on a grayscale value interval corresponding to the insulator, and to identify an image region of an operating electrician in the multi-mapping image based on a reference outline pattern corresponding to the operating electrician, wherein the grayscale value interval corresponding to the insulator is within a range of 0-255, and the reference outline pattern corresponding to the operating electrician is in a JPEG encoding format;

[0035] a second analysis unit, connected to the first analysis unit, configured to analyze whether the insulator and the operating electrician are in contact with each other using a convolutional neural network model based on the overall depth of field value and the number of pixels occupied by the image area of ​​the insulator in the multiple mapped images, the overall depth of field value and the number of pixels occupied by the image area of ​​the operating electrician in the multiple mapped images, the coordinate data corresponding to each pixel occupied by the image area of ​​the insulator in the multiple mapped images, and the coordinate data corresponding to each pixel occupied by the image area of ​​the operating electrician in the multiple mapped images;

[0036] a touch alarm mechanism, connected to the second analysis mechanism, for executing an alarm action corresponding to the operating electrician illegally touching the insulator before the conductor-side insulator is detached, when the second analysis mechanism uses a convolutional neural network model to analyze the handover between the insulator and the operating electrician;

[0037] The convolutional neural network model is a convolutional neural network that has completed a set number of learning cycles, and the number of learning cycles is monotonically positively correlated with the mean of the number of pixel rows and the number of pixel columns of the multiple mapped images.

[0038] The method includes: determining whether the conductor-side insulator of the tower of the power distribution network line has fallen off based on sensor data of a height sensor of the conductor-side insulator; and issuing a first judgment instruction when determining that the conductor-side insulator of the tower of the power distribution network line has fallen off.

[0039] The touch alarm mechanism is further configured to temporarily suspend the execution of an alarm action corresponding to the operating electrician illegally touching the insulator before the conductor-side insulator is detached, when the second analysis mechanism uses a convolutional neural network model to analyze that the insulator and the operating electrician have not yet been handed over;

[0040] The imaging processing device is further configured to terminate the imaging processing of the tower working scene upon receiving the second judgment instruction.

[0041] Figure 2 Schematic diagram of the internal structure of a distribution network touch status alarm system according to an embodiment of the present invention.

[0042] and Figure 1 different, Figure 2 The distribution network contact status alarm system in the system may also include the following components:

[0043] a length sensor connected to the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device, and the cubic interpolation device, respectively, for measuring the current real-time length values ​​of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device, and the cubic interpolation device, respectively;

[0044] The length sensor is connected to the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device, and the cubic interpolation device, respectively, and is used to measure the current real-time length values ​​of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device, and the cubic interpolation device, respectively. The length sensor includes a plurality of length measurement units, which are used to be connected to the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device, and the cubic interpolation device, respectively, to complete the measurement of the current real-time length values ​​of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device, and the cubic interpolation device, respectively.

[0045] The length sensor device includes a plurality of length measurement units, which are respectively connected to the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device, and the cubic interpolation device to respectively measure the current real-time length values ​​of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device, and the cubic interpolation device. The plurality of length measurement units are a plurality of length sensors, which are respectively connected to the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device, and the cubic interpolation device to respectively measure the current real-time length values ​​of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device, and the cubic interpolation device.

[0046] The multiple length measurement units are multiple length sensors, which are respectively connected to the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device, and the cubic interpolation device to respectively measure the current real-time length values ​​of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device, and the cubic interpolation device. The multiple length sensors have the same structure.

[0047] Among them, the multiple length measurement units are multiple length sensors, which are used to connect to the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device respectively to complete the separate measurement of the current real-time length values ​​of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device. It also includes: the multiple length sensors have the same length measurement upper limit value and length measurement lower limit value.

[0048] Figure 3 Schematic diagram of the internal structure of a power distribution network touch status alarm system according to embodiment C of the present invention.

[0049] and Figure 1 different, Figure 3 The distribution network contact status alarm system in the system may also include the following components:

[0050] a length display mechanism, connected to the plurality of length measurement units of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device, and the cubic interpolation device, respectively, and configured to synchronously display the current real-time length values ​​of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device, and the cubic interpolation device;

[0051] The length display mechanism is connected to the plurality of length measurement units of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device, and the cubic interpolation device, respectively, and is used to synchronously display the current real-time length values ​​of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device, and the cubic interpolation device, respectively. The length display mechanism is a touch-screen display.

[0052] Among them, the length display mechanism is respectively connected to multiple length measurement units of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device, and is used to synchronously display the current real-time length values ​​of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device, including: the length display mechanism is a non-touch display screen.

[0053] In addition, in the distribution network touch status alarm system, a multiple mapping device is connected to the imaging processing device and includes a signal filtering device, an affine transformation device and a cubic interpolation device, which is used to perform salt and pepper noise filtering processing, affine transformation processing and cubic polynomial interpolation processing on the received image sequence to obtain and output the corresponding multiple mapping image, including: the signal filtering device, the affine transformation device and the cubic interpolation device are respectively used to perform salt and pepper noise filtering processing, affine transformation processing and cubic polynomial interpolation processing on the received image signal.

[0054] While the present invention has been described and illustrated with respect to exemplary embodiments, those skilled in the art will appreciate that the foregoing and various other changes, omissions, and additions may be made without departing from the spirit and scope of the invention.

Claims

1. A distribution network touch status alarm system, characterized in that: The system comprises: A separation judgment device, used to issue a first judgment instruction when judging that the conductor side insulator of the pole tower of the power distribution network line has not fallen off, and also used to issue a second judgment instruction when judging that the conductor side insulator of the pole tower of the power distribution network line has fallen off; An imaging processing device, connected to the separation judgment device, for performing imaging processing on the tower working scene upon receiving the first judgment instruction, so as to obtain and output a corresponding working scene image; A multiple mapping device, connected to the imaging processing device and comprising a signal filtering device, an affine transformation device and a cubic interpolation device, for performing salt and pepper noise filtering processing, affine transformation processing and cubic polynomial interpolation processing on the received image sequence to obtain and output a corresponding multiple mapping image; A first analysis mechanism is connected to the multiple mapping device and is used to identify the image area of ​​the insulator in the multiple mapping images according to the grayscale value interval corresponding to the insulator, and to identify the image area of ​​the working electrician in the multiple mapping images according to the reference shape pattern corresponding to the working electrician, wherein the grayscale value interval corresponding to the insulator is between 0 and 255, and the reference shape pattern corresponding to the working electrician is in JPEG encoding format; a second analysis mechanism, connected to the first analysis mechanism, for analyzing whether the insulator and the working electrician are handed over by using a convolutional neural network model according to the overall depth of field value and the number of pixels occupied by the image area of ​​the insulator in the multiple mapped images, the overall depth of field value and the number of pixels occupied by the image area of ​​the working electrician in the multiple mapped images, the coordinate data corresponding to each pixel occupied by the image area of ​​the insulator in the multiple mapped images, and the coordinate data corresponding to each pixel occupied by the image area of ​​the working electrician in the multiple mapped images; A touch alarm mechanism, connected to the second analysis mechanism, for executing an alarm action corresponding to the illegal touching of the insulator by the working electrician before the insulator on the conductor side is detached when the second analysis mechanism uses a convolutional neural network model to analyze the handover between the insulator and the working electrician; The convolutional neural network model is a convolutional neural network after completing a set number of learning times, and the number of learning times is monotonically positively correlated with the mean of the number of pixel rows and the number of pixel columns of the multiple mapped images; Among them, when it is judged that the conductor side insulator of the pole tower of the distribution network line has not fallen off, a first judgment instruction is issued, and it is also used to issue a second judgment instruction when it is judged that the conductor side insulator of the pole tower of the distribution network line has fallen off, including: judging whether the conductor side insulator of the pole tower of the distribution network line has fallen off based on the sensor data of the height sensor of the conductor side insulator.

2. The power distribution network contact status alarm system according to claim 1, characterized in that: The touch alarm mechanism is also used to suspend the alarm action corresponding to the working electrician illegally touching the insulator before the insulator on the conductor side is detached when the second analysis mechanism uses the convolutional neural network model to analyze that the insulator and the working electrician have not been handed over; Wherein, the imaging processing device is further used to terminate the imaging processing of the tower working scene when receiving the second judgment instruction.

3. The power distribution network contact status alarm system according to claim 2, characterized in that: The system further comprises: a length sensor, connected to the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device, respectively, for measuring the current real-time length values ​​of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device respectively; Among them, the length sensor device is respectively connected to the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device, and is used to respectively measure the current real-time length values ​​of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device. The length sensor device includes multiple length measurement units, which are respectively connected to the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device to complete the measurement of the current real-time length values ​​of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device.

4. The power distribution network contact status alarm system according to claim 3, characterized in that: The length sensor device includes a plurality of length measurement units, which are used to be respectively connected to the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device to complete the respective measurement of the current real-time length values ​​of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device. The plurality of length measurement units are a plurality of length sensors, which are used to be respectively connected to the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device to complete the respective measurement of the current real-time length values ​​of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device.

5. The power distribution network contact status alarm system according to claim 4, characterized in that: The multiple length measurement units are multiple length sensors, which are used to be connected to the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device respectively, so as to complete the measurement of the current real-time length values ​​of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device respectively, including: the structures of the multiple length sensors are the same.

6. The power distribution network contact status alarm system according to claim 5, characterized in that: The multiple length measurement units are multiple length sensors, which are used to be connected to the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device respectively to complete the measurement of the current real-time length values ​​of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device respectively, and also include: the multiple length sensors have the same length measurement upper limit value and length measurement lower limit value.

7. The power distribution network touch status alarm system according to any one of claims 3 to 6, characterized in that: The system further comprises: The length display mechanism is respectively connected to multiple length measurement units of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device, and is used to synchronously display the current real-time length values ​​of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device.

8. The power distribution network contact status alarm system according to claim 7, characterized in that: The length display mechanism is respectively connected to the multiple length measurement units of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device, and is used to synchronously display the current real-time length values ​​of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device. The length display mechanism is a touch display screen.

9. The power distribution network contact status alarm system according to claim 7, characterized in that: The length display mechanism is respectively connected to the multiple length measurement units of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device, and is used to synchronously display the current real-time length values ​​of the first analysis mechanism, the second analysis mechanism, the signal filtering device, the affine transformation device and the cubic interpolation device. The length display mechanism is a non-touch display screen.

Images