Electric power operation whole-process digital management and control real-time early warning method

Abstract

The application discloses a power operation whole-process digital management and control real-time early warning method, relates to the technical field of power operation management and control early warning, and is used for solving the problem of reduced operation safety of power transmission equipment. The method comprises the following steps: identifying the attachments of the power transmission equipment by using a drone and marking the corresponding components, detecting the attachment coverage state and the number of corona ultraviolet photons within a set statistical time, analyzing the coverage growth trend and combining the photon number to classify abnormal states, judging whether to clean up, collecting the speckle state by using a laser after cleaning, analyzing the characteristic value to judge whether to clean up again, simultaneously detecting the cleaning frequency and the surface leakage current, comprehensively evaluating the safety level and deciding the alarm signal, realizing unmanned and continuous monitoring, combining the corona response and the speckle analysis to judge the cleaning demand, evaluating the safety level through the cleaning frequency and the leakage current, improving the power transmission equipment maintenance efficiency and the operation safety, reducing the power operation inspection cost, and improving the reliability of the power transmission equipment operation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power operation management and control early warning, more specifically, the present application relates to a power operation whole-process digital management and control real-time early warning method. BACKGROUND

[0002] During the long-term operation of power transmission equipment, it is easy to be polluted by dust, bird droppings, branches and other attachments. These attachments cover the key parts of the power transmission equipment, which can cause the insulation performance of the equipment to decrease, increase the corona discharge and surface leakage current, and further affect the safe operation of the power transmission equipment. The traditional inspection method mainly relies on regular cleaning, which has the problems of low efficiency, long inspection period and difficulty in continuous monitoring.

[0003] The prior art has the following disadvantages:

[0004] At present, the prior art cannot realize unmanned and continuous monitoring of the attachments on the power transmission equipment, and it is difficult to grasp the state change of the attachments in a timely manner. There is a lack of effective way to intelligently judge the abnormal state of the attachments in combination with the corona response and image features, and the cleaning decision lacks accurate basis, which leads to the reduction of the operation safety of the power transmission equipment, the increase of the maintenance cost and the accident risk. Therefore, a power operation whole-process digital management and control real-time early warning method is proposed.

[0005] The above information disclosed in the background section is only used to enhance the understanding of the background of the present disclosure, and therefore it can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0006] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present application provide a power operation whole-process digital management and control real-time early warning method, which uses unmanned aerial vehicle image recognition, corona response detection, laser speckle analysis and intelligent judgment algorithm of multi-dimensional data fusion to solve the problems proposed in the above background technology.

[0007] To achieve the above-mentioned purpose, the present application provides the following technical scheme, a power operation whole-process digital management and control real-time early warning method, comprising the following steps:

[0008] Step S1: identifying the attachments on the power transmission equipment by the unmanned aerial vehicle camera equipment, marking the corresponding equipment components of the attachments, setting the statistical time, and detecting the coverage state of the attachments and the number of corona ultraviolet photons of the marked equipment components within the statistical time;

[0009] Step S2: analyzing the growth trend of the attachments according to the coverage state, analyzing the abnormal state of the attachments in combination with the number of corona ultraviolet photons, and judging whether to clean the attachments based on the abnormal state;

[0010] Step S3: After the attachment is cleaned, the speckle state of the marked equipment component is collected by the laser lighting device, the cleaning feature is analyzed according to the speckle state, and it is judged whether to clean again by using the cleaning feature;

[0011] Step S4: The cleaning times and surface leakage current of the marked equipment component are detected, the safety level is evaluated by comprehensively considering the cleaning times and surface leakage current, and it is selected whether to generate an alarm signal according to the safety level.

[0012] In a preferred embodiment, in step S1, the surface image of the power transmission equipment is collected by the unmanned aerial vehicle carrying the high-definition visible light camera device;

[0013] Based on the surface image, the equipment component with the attachment bounding box image is marked;

[0014] The statistical time is set, the marked equipment component is binarized and segmented in the statistical time, the boundary contour of the attachment is extracted, and the number of pixel points in the boundary contour area of the attachment and the number of pixel points of the marked equipment component are counted;

[0015] The ratio of the number of pixel points in the boundary contour area of the attachment to the number of pixel points of the marked equipment component is taken as the coverage state of the attachment;

[0016] The number of corona ultraviolet photons of the marked equipment component is obtained based on the ultraviolet detector of single-photon technology.

[0017] In a preferred embodiment, in step S2, the coverage state change value of the attachment at adjacent collection time in the statistical time is calculated by difference value calculation;

[0018] The coverage state change value of the attachment in the statistical time is summed and averaged to obtain the growth trend of the attachment.

[0019] In a preferred embodiment, in step S2, the growth trend of the attachment and the number of corona ultraviolet photons of the marked equipment component are normalized;

[0020] The abnormal state feature value of the attachment is calculated based on the values after the normalization of the growth trend and the number of corona ultraviolet photons.

[0021] In a preferred embodiment, in step S2, the abnormal state feature value of the attachment is compared with the preset abnormal state feature threshold value:

[0022] If the abnormal state feature value is less than the abnormal state feature threshold value, the abnormal state of the attachment is a low-risk abnormal state, and the attachment is not cleaned;

[0023] On the contrary, the abnormal state of the attachment is a high-risk abnormal state, and the attachment is cleaned.

[0024] In a preferred embodiment, in step S3, a preset acquisition period is set, and a laser beam is emitted onto the marking device component through a laser illumination device to form speckle on the surface of the component;

[0025] Grayscale images of the marking device components that form speckles on their surface are acquired using a high-speed grayscale camera.

[0026] The method of edge detection is used to identify the area to be cleaned of the deposits by recognizing the grayscale image of the marked equipment parts with speckled surface.

[0027] Speckle pattern refers to the random brightness distribution phenomenon formed when coherent light irradiates a rough surface or non-uniform medium, including speckle contrast and speckle particle size;

[0028] Calculate the standard deviation and mean of pixel grayscale within the area where the attachment was cleaned, and use the ratio of the standard deviation to the mean of pixel grayscale within the area where the attachment was cleaned as the speckle contrast.

[0029] The size of speckle particles is calculated using the grayscale value of the coordinates corresponding to the area where the attachment was cleaned.

[0030] In a preferred embodiment, in step S3, the speckle contrast and speckle particle size are standardized.

[0031] The cleaning features are calculated by combining the speckle contrast and the standardized values ​​of speckle particle size.

[0032] The cleanup features are compared with a preset cleanup feature threshold to determine whether secondary cleanup is required.

[0033] If the cleanup feature is greater than or equal to the preset cleanup feature threshold, then a second cleanup is performed;

[0034] If the cleanup feature is less than the preset cleanup feature threshold, no secondary cleanup will be performed.

[0035] In a preferred embodiment, in step S4, the contact switch sensor of the cleaning device detects whether the cleaning device is in contact with the marking device component;

[0036] The contact result is sent to the recording unit via the wireless communication unit;

[0037] Matching the marking device with the recording unit yields the number of times the cleaning device and the marking device components have contacted each other, i.e., the number of times the marking device components have been cleaned.

[0038] Real-time acquisition of surface leakage current of marked equipment components using a high-impedance current sensor;

[0039] The number of cleaning cycles and surface leakage current of the marked equipment components were normalized using the Max-Min normalization method.

[0040] The safety characteristic value of the marked equipment component is calculated by synthesizing the cleaning times of the marked equipment component and the normalized value of the surface leakage current;

[0041] The mean and standard deviation of the safety characteristic value of the marked equipment component are calculated.

[0042] In a preferred embodiment, in step S4, the mean and standard deviation of the safety characteristic value of the marked equipment component are added as a safety level first threshold value;

[0043] The mean and standard deviation of the safety characteristic value of the marked equipment component are subtracted as a safety level second threshold value;

[0044] If the safety characteristic value of the marked equipment component is less than the safety level second threshold value, the safety level is a low safety risk level, and no alarm signal needs to be generated for continuing the current state operation;

[0045] If the safety characteristic value of the marked equipment component is greater than or equal to the safety level second threshold value and less than the safety level first threshold value, the safety level is a medium safety risk level, and a warning signal is generated;

[0046] If the safety characteristic value of the marked equipment component is greater than or equal to the safety level first threshold value, the safety level is a high safety risk level, and an emergency alarm signal is generated.

[0047] Technical effects and advantages of the present application:

[0048] The present application identifies the attachments on the power transmission equipment by the unmanned aerial vehicle image, marks the corresponding equipment components of the attachments, detects the attachment coverage state and the corona ultraviolet photon number of the marked equipment components within a set statistical time, analyzes the growth trend according to the attachment coverage state, classifies the abnormal state of the attachments in combination with the corona ultraviolet photon number, and determines whether cleaning is needed accordingly. After the attachments are cleaned, the speckle state of the marked equipment components is collected by using the laser illumination device, the speckle characteristics are analyzed, and the cleaning characteristic value is calculated to determine whether secondary cleaning is needed. At the same time, the system detects the cleaning times and surface leakage current of the marked equipment components, comprehensively evaluates the safety level, and decides whether an alarm signal is generated according to the safety level. The present application realizes unmanned and continuous monitoring of the attachment state, judges the cleaning requirement in combination with the corona response and speckle analysis, comprehensively evaluates the safety level through the cleaning times and surface leakage current, improves the power transmission equipment maintenance efficiency and operation safety, reduces the power operation inspection cost, and improves the reliability of the power transmission equipment operation. BRIEF DESCRIPTION OF DRAWINGS

[0049] Fig. 1 The implementation flowchart of the real-time early warning method for the digital management and control of the whole process of the power operation of the present application.

[0050] Fig. 2 The step schematic diagram of the power operation whole-process digital management and control real-time early warning method. DETAILED DESCRIPTION

[0051] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0052] The present application identifies the attachments on the power transmission equipment by the unmanned aerial vehicle image, and marks the corresponding equipment components of the attachments. The attachment coverage state and the corona ultraviolet photon number of the marked equipment components are detected within the set statistical time. The growth trend of the attachment coverage state is analyzed, the abnormal state of the attachment is classified in combination with the corona ultraviolet photon number, and it is judged whether cleaning is needed. After the attachments are cleaned, the speckle state of the marked equipment components is collected by using the laser illumination device, the speckle characteristics are analyzed and the cleaning characteristic value is calculated to judge whether secondary cleaning is needed. At the same time, the cleaning times and the surface leakage current of the marked equipment components are detected, the safety level is comprehensively evaluated, and it is decided whether an alarm signal is generated according to the safety level. The unmanned and continuous monitoring of the attachment state is realized, the cleaning requirement is judged in combination with the corona response and the speckle analysis, the safety level is comprehensively evaluated by the cleaning times and the surface leakage current, and the maintenance efficiency and the operation safety of the power transmission equipment are improved.

[0053] Embodiment 1, a power operation whole-process digital management and control real-time early warning method, as shown in the figure, includes the following steps: Figs. 1-2

[0054] Step S1: Identify the attachments on the power transmission equipment by the unmanned aerial vehicle camera equipment, mark the corresponding equipment components of the attachments, set the statistical time, and detect the coverage state of the attachments and the corona ultraviolet photon number of the marked equipment components within the statistical time;

[0055] Step S2: Analyze the growth trend of the attachments according to the coverage state, analyze the abnormal state of the attachments in combination with the corona ultraviolet photon number, and judge whether the attachments need to be cleaned based on the abnormal state;

[0056] Step S3: After the attachments are cleaned, the speckle state of the marked equipment components is collected by the laser illumination device, the cleaning characteristics are analyzed according to the speckle state, and it is judged whether secondary cleaning is needed by using the cleaning characteristics;

[0057] Step S4: Detect the cleaning times and the surface leakage current of the marked equipment components, comprehensively evaluate the safety level by the cleaning times and the surface leakage current, and select whether an alarm signal is generated according to the safety level. ​

[0058] The implementation is as follows:

[0059] In step S1, the unmanned aerial vehicle equipped with a high-definition visible light camera device automatically inspects the power transmission equipment according to a preset route and collects surface images of the power transmission equipment;

[0060] The surface images of the power transmission equipment are preprocessed according to Gaussian filtering;

[0061] An image recognition algorithm based on a convolutional neural network is used to identify the attachments on the power transmission equipment and mark the corresponding equipment components, and the specific steps are as follows:

[0062] The historical images containing the power transmission equipment and the attachments are combined to model the ResNet convolutional neural network architecture;

[0063] The cross-entropy loss is selected as the loss function, the Adam optimizer is used to update the network weight, and the model is trained through backpropagation;

[0064] The new surface images of the power transmission equipment are input into the trained convolutional neural network image recognition model to obtain the bounding boxes of the attachments and mark the equipment components with the bounding boxes of the attachments;

[0065] It should be noted that the unmanned aerial vehicle equipped with a high-definition visible light camera device refers to an unmanned aerial vehicle equipped with a high-resolution visible light camera for collecting surface images of the power transmission equipment; the preset route refers to one or more flight routes planned in advance before the unmanned aerial vehicle inspection, determined according to the power transmission equipment distribution map and equipment height, etc.; Gaussian filtering is a smoothing method in image processing, used for preprocessing the surface images of the power transmission equipment; the convolutional neural network is a deep learning model that extracts local features of images through convolutional layers, and then classifies or regresses through fully connected layers, used for identifying attachments on the power transmission equipment; ResNet is a very important network architecture in deep learning, used to build the convolutional neural network image recognition model; cross-entropy loss is the most commonly used loss function in classification tasks, used to measure the difference between the predicted probability distribution and the true label distribution; Adam optimizer is a commonly used optimization algorithm that can dynamically adjust the learning rate of each parameter to improve training efficiency and convergence speed; backpropagation is a very important algorithm in neural networks, used to optimize the weights of the network.

[0066] After marking the corresponding equipment components of the attachments, a statistical time is set and a plurality of adjacent collection time points are divided, and within the statistical time, the marked equipment components are binarized and segmented to extract the boundary contour of the attachments, and the number of pixel points in the boundary contour area of the attachments and the number of pixel points of the marked equipment components are counted;

[0067] The ratio of the number of pixels in the boundary contour region of the adherend to the number of pixels of the marked device component is taken as the coverage state of the adherend;

[0068] The single-photon technology-based ultraviolet detector obtains the number of corona ultraviolet photons of the marked device component;

[0069] The single-photon technology-based ultraviolet detector is placed in a light-free environment and kept in the same detection environment as in actual measurement to collect dark counts;

[0070] A device component without adherend is selected, kept in the same detection environment as in actual measurement, and the single-photon technology-based ultraviolet detector is used to collect environmental background counts;

[0071] The actual number of corona ultraviolet photons of the marked device component is obtained by deducting the effects of dark counts and environmental background counts on the number of corona ultraviolet photons of the marked device component.

[0072] It should be explained that the statistical time refers to the time window for collecting and analyzing the state of the adherend or sensor data. According to the flight speed of the unmanned aerial vehicle and the shooting frame rate of the camera, at least 5 images should be collected in each statistical time. Binary segmentation is a commonly used image processing technique for converting images into two categories for extracting the boundary contour of the adherend. The single-photon technology-based ultraviolet detector is an advanced device for detecting single photons in the ultraviolet waveband, which is used to obtain the number of corona ultraviolet photons of the marked device component. Dark counts refer to the false photon counts produced by the single-photon technology-based ultraviolet detector due to factors such as thermal noise and electronic noise of the device itself when there is no ultraviolet photon incident. Environmental background counts are the photon counts received by the single-photon technology-based ultraviolet detector in the actual environment that are irrelevant to the target signal.

[0073] In step S2, the coverage state change value of the adherend at adjacent collection time points in the statistical time is calculated by difference calculation to obtain the coverage state change value of the adherend;

[0074] The coverage state change values of the adherend in the statistical time are summed and averaged to obtain the growth trend of the adherend;

[0075] It should be noted that if the growth trend of the adherend is greater than 0, the adherend is in a growth state in the statistical time; if the growth trend of the adherend is less than 0, the adherend is in a decreasing state in the statistical time; if the growth trend of the adherend tends to 0, the adherend is in a stable state in the statistical time.

[0076] The growth trend of the adherend and the number of corona ultraviolet photons of the marked device component are normalized, and the normalization formula is: 、 wherein, and the growth trend of the deposit and the number of corona ultraviolet photons of the marked device component at the first sampling time, the growth trend of the deposit and the number of corona ultraviolet photons of the marked device component at the first sampling time, the total number of sampling times within the statistical time, and the normalized value of the growth trend of the deposit and the number of corona ultraviolet photons of the marked device component;

[0077] The abnormal state feature value of the deposit is calculated by normalizing the value of the growth trend of the deposit and the number of corona ultraviolet photons of the marked device component, and the calculation formula is: wherein, the normalized value of the growth trend of the deposit, the normalized value of the number of corona ultraviolet photons of the marked device component, the abnormal state feature value of the deposit;

[0078] It should be noted that the greater the growth trend of the deposit and the more the number of corona ultraviolet photons of the marked device component, the more serious the pollution of the device surface and the more significant the corona discharge phenomenon, and the greater the abnormal state feature value of the deposit; the smaller the growth trend of the deposit and the fewer the number of corona ultraviolet photons of the marked device component, the less the pollution of the device surface and the weaker the corona discharge, and the smaller the abnormal state feature value of the deposit.

[0079] The abnormal state feature value of the deposit is compared with a preset abnormal state feature threshold value:

[0080] If the abnormal state feature value is less than the abnormal state feature threshold value, the abnormal state of the deposit is a low-risk abnormal state, and the deposit is not cleaned;

[0081] If the abnormal state feature value is greater than or equal to the abnormal state feature threshold value, the abnormal state of the deposit is a high-risk abnormal state, and the deposit is cleaned;

[0082] It should be explained that the preset abnormal state feature threshold value is an important parameter for judging the abnormal state of the deposit. The statistical data of the growth trend of the deposit and the number of ultraviolet photons during normal operation of the power transmission device are analyzed, and the threshold value is determined using the mean and standard deviation. The low-risk abnormal state refers to the abnormal phenomenon of the device, but has little effect on the safety, operation stability or environment of the device. The high-risk abnormal state refers to the abnormal phenomenon of the device, which may have a significant impact on the safety, operation stability or surrounding environment of the device, and if not handled in time, it may cause failure, damage or safety accidents.

[0083] In step S3, a preset collection period is set, and a light beam is emitted to the marked device component by a laser illumination device to form a speckle on the surface of the component;

[0084] acquire the gray scale image of the surface forming speckle of the marking device component by a high-speed gray scale camera;

[0085] identify and determine the adherent cleaning area based on the edge detection method for the gray scale image of the surface forming speckle of the marking device component;

[0086] The speckle state refers to the random brightness distribution phenomenon formed when coherent light is irradiated to a rough surface or a non-uniform medium, including speckle contrast and speckle particle size;

[0087] The ratio of the standard deviation to the mean of the pixel gray scale in the adherent cleaning area is taken as the speckle contrast;

[0088] It should be noted that the preset acquisition period refers to the time interval for data acquisition, monitoring and measurement, which is set according to the historical operation data and experience of the equipment, and the data change law of the power transmission equipment component under different periods is analyzed; the laser illumination device is a device that provides directional illumination or light source using laser technology, which is used to emit a light beam to the marking device component to form speckle on the component surface; speckle is an optical phenomenon caused by light interference or reflection, which usually appears as random brightness changes and granular patterns on the surface or object; the high-speed gray scale camera is a camera that captures high-speed moving objects or rapidly changing events, which is used to acquire the gray scale image of the surface forming speckle of the marking device component; edge detection is a basic method in computer vision and image processing, which is used to identify the edges or boundaries of the adherent cleaning area from the gray scale image of the surface forming speckle of the marking device component.

[0089] One vertex of the gray scale image of the surface forming speckle of the marking device component is taken as the origin of the two-dimensional coordinate, and one pixel horizontal interval is taken as the unit scale of the x-axis, and one pixel vertical interval is taken as the unit scale of the y-axis;

[0090] The speckle particle size is calculated by the gray scale value of the corresponding coordinate of the adherent cleaning area, and the calculation formula is: wherein, , and are the gray scale values of the corresponding coordinates of the adherent cleaning area, is the speckle particle size;

[0091] The speckle contrast and the speckle particle size are standardized;

[0092] The cleaning feature is calculated by integrating the standardized values of the speckle contrast and the speckle particle size, and the calculation formula is: wherein, is the standardized value of the speckle contrast, is the standardized value of the speckle particle size, to clean the feature;

[0093] It should be noted that the greater the speckle contrast, the larger the speckle particle size, the more serious the attachment on the surface of the marked equipment component, and the larger the cleaning feature; the smaller the speckle contrast, the smaller the speckle particle size, the smoother the surface of the marked equipment component, and the smaller the cleaning feature.

[0094] The cleaning feature is compared with the preset cleaning feature threshold to determine whether to clean again:

[0095] If the cleaning feature is greater than or equal to the preset cleaning feature threshold, cleaning again is performed;

[0096] If the cleaning feature is less than the preset cleaning feature threshold, cleaning again is not performed.

[0097] It should be noted that the standardization processing method includes but is not limited to a standard linear transformation based on interval scaling, a Z-Score standardization method based on statistics, or a normalization method based on a nonlinear mapping function, and the application method of the standardization processing is not described here; the preset cleaning feature threshold is an important parameter for determining whether to clean again, and the influence of the speckle contrast and the speckle particle size of the surface of the cleaned power equipment component on the operation of the power equipment component is analyzed through historical data to select the minimum cleaning feature of the power equipment component with abnormal operation as the preset cleaning feature threshold.

[0098] In step S4, whether the cleaning equipment is in contact with the marked equipment component is detected by a contact switch sensor of the cleaning equipment;

[0099] The contact result is sent to the recording unit through a wireless communication unit;

[0100] The cleaning equipment and the recording unit are matched to obtain the number of times that the cleaning equipment is in contact with the marked equipment component, i.e., the cleaning times of the marked equipment component;

[0101] The surface leakage current of the marked equipment component is collected in real time by a high-impedance current sensor;

[0102] The cleaning times and the surface leakage current of the marked equipment component are normalized by a Max-Min normalization method, and the calculation formula is: , wherein, and are the cleaning times and the surface leakage current of the marked equipment component, and are the maximum value and the minimum value of the cleaning times of the marked equipment component, and are the maximum value and the minimum value of the surface leakage current of the marked equipment component, and a normalized value of the cleaning frequency of the marked equipment component and a normalized value of the surface leakage current of the marked equipment component are obtained;

[0103] a safety characteristic value of the marked equipment component is calculated by integrating the normalized value of the cleaning frequency of the marked equipment component and the normalized value of the surface leakage current of the marked equipment component, and the calculation formula is: wherein, is a normalized value of the cleaning frequency of the marked equipment component, is a normalized value of the surface leakage current of the marked equipment component, and is a preset weighting coefficient, is a safety characteristic value of the marked equipment component;

[0104] a mean value and a standard deviation of the safety characteristic value of the marked equipment component are calculated respectively, and the mean value and the standard deviation of the safety characteristic value of the marked equipment component are added or subtracted as a first threshold value and a second threshold value of the safety level;

[0105] the alarm signal is used to prompt the abnormal safety state of the power transmission equipment and remind the operation and maintenance personnel to take maintenance measures, and the alarm signal is divided into a warning signal and an emergency alarm signal according to different safety risks;

[0106] the safety characteristic value of the marked equipment component is compared with the first threshold value and the second threshold value of the safety level to judge:

[0107] if the safety characteristic value of the marked equipment component is less than the second threshold value of the safety level, the safety level is a low safety risk level, and the current state continues to run without generating an alarm signal;

[0108] if the safety characteristic value of the marked equipment component is greater than or equal to the second threshold value of the safety level and less than the first threshold value of the safety level, the safety level is a medium safety risk level, and a warning signal is generated;

[0109] if the safety characteristic value of the marked equipment component is greater than or equal to the first threshold value of the safety level, the safety level is a high safety risk level, and an emergency alarm signal is generated;

[0110] It needs to be explained that the contact switch sensor of the cleaning device is a device for detecting whether the surface of the device contacts an object through mechanical, photoelectric or capacitive technology, which is used to detect whether the cleaning device contacts the marked device component; the wireless communication unit refers to an electronic component for wireless data transmission, which is used to send the contact result to the recording unit; the recording unit refers to a device for storing and managing data, which is used to store the contact result of the cleaning device and the marked device component; the high-impedance current sensor is a sensor for measuring very weak current, especially leakage current or other weak signals, which is used to collect the surface leakage current of the marked device component in real time; the preset weighting coefficient is used to balance the influence of the cleaning times of the marked device component and the surface leakage current of the marked device component on the safety characteristic value of the marked device component, and set the influence of the cleaning times of the marked device component and the surface leakage current of the marked device component on the safety characteristic value combined with the safety characteristic value regression analysis; the warning signal is a relatively mild warning, indicating that the state of the power transmission device has deviated from the normal range, but has not yet reached an emergency or dangerous level, and its main purpose is to remind the operation and maintenance personnel to pay attention and carry out preventive inspection or maintenance to avoid further deterioration of potential problems; the emergency alarm signal is a strong warning issued when the power transmission device has a serious failure, anomaly or dangerous situation, and its purpose is to quickly attract the attention of the operator and require emergency response measures to prevent accidents, equipment damage or personnel injury.

[0111] Finally, it also needs to be explained that in this paper, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations.

[0112] Moreover, the term "include" "contain" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device. Without more limitations, the element defined by the statement "includes a" does not exclude the presence of another identical element in the process, method, article or device including the element.

[0113] In this document, the singular forms "a," "an," and "the" include plural references unless the context clearly dictates otherwise. It will be further understood that the terms "comprises," "comprising," "includes," and "including," or the like, when used in this specification, specify the presence of stated features, integers, steps, operations, components, parts, or the like, but do not preclude the presence or addition of one or more other features, integers, steps, operations, components, parts, or the like.

[0114] Various embodiments described in this specification are described with reference to particular implementations. Embodiments can be practiced with other systems, components, materials, acts, operations, and steps than those described and / or shown in this specification, and not solely with the particular implementations described in this specification. The terms "comprise," "comprising," "include," "including," and "includes" as well as related terms such as "comprises" or "comprising," when used in this specification, include the presence of stated features, integers, steps, operations, components, parts, or the like, but do not preclude the presence or addition of one or more other features, integers, steps, operations, components, parts, or groups thereof.

[0115] The above description of disclosed embodiments is not intended to be exhaustive or to be unduly restrictive of the application. The descriptions above are intended to be illustrative rather than limiting. Many modifications and variations of the described embodiments are possible and will be apparent to those of ordinary skill in the art once the benefits of the present application are known. For example, it will be apparent to one of ordinary skill in the art that the methods described above can be implemented in a variety of ways.

Claims

1. A method for real-time early warning of electric power operation whole-process digitalization management and control, characterized in that: Comprise the following steps: Step S1: identify the attachments on the power transmission equipment by the unmanned aerial vehicle camera equipment, mark the corresponding equipment components of the attachments, set the statistical time, detect the coverage state of the attachments and the corona ultraviolet photon number of the marked equipment components within the statistical time; Step S2: analyze the growth trend of the attachments according to the coverage state, analyze the abnormal state of the attachments in combination with the corona ultraviolet photon number, and judge whether to clean the attachments based on the abnormal state; Step S3: After cleaning the attachments, collect the speckle state of the marked equipment components by the laser illumination device, analyze the cleaning characteristics according to the speckle state, and judge whether to clean again by using the cleaning characteristics; Step S4: Detect the cleaning times and surface leakage current of the marked equipment components, evaluate the safety level by comprehensively considering the cleaning times and surface leakage current, and select whether to generate an alarm signal according to the safety level.

2. The real-time early warning method for digital management and control of the whole process of power operation according to claim 1, characterized in that: In step S1, the surface image of the power transmission equipment is collected by the unmanned aerial vehicle equipped with a high-definition visible light camera device; Based on the surface image, the equipment components with the attachment bounding box image are marked; The statistical time is set, the marked equipment components are binarized and segmented within the statistical time, the boundary contour of the attachments is extracted, and the number of pixel points in the boundary contour area of the attachments and the number of pixel points of the marked equipment components are counted; The ratio of the number of pixel points in the boundary contour area of the attachments to the number of pixel points of the marked equipment components is taken as the coverage state of the attachments; The corona ultraviolet photon number of the marked equipment components is obtained based on the single-photon technology ultraviolet detector.

3. The real-time early warning method for digital management and control of the whole process of power operation according to claim 2, characterized in that: In step S2, the coverage state change value of the attachments at adjacent collection time points within the statistical time is calculated by difference value calculation; The coverage state change value of the attachments within the statistical time is summed and averaged to obtain the growth trend of the attachments.

4. The real-time early warning method for digital management and control of the whole process of power operation according to claim 3, characterized in that: In step S2, the growth trend of the attachments and the corona ultraviolet photon number of the marked equipment components are normalized; The abnormal state characteristic value of the attachments is calculated based on the normalized values of the growth trend and the corona ultraviolet photon number.

5. The real-time early warning method for digital management and control of the whole process of power operation according to claim 4, characterized in that: In step S2, the abnormal state characteristic value of the attachments is compared with the preset abnormal state characteristic threshold value: If the abnormal state characteristic value is less than the abnormal state characteristic threshold value, the abnormal state of the attachments is a low-risk abnormal state, and the attachments are not cleaned; Otherwise, the abnormal state of the attachments is a high-risk abnormal state, and the attachments are cleaned.

6. The real-time early warning method for digital management and control of the whole process of power operation according to claim 1, characterized in that: In step S3, a preset collection period is set, and the laser illumination device emits a light beam to the marked equipment components to form a speckle on the component surface; The gray scale image of the surface of the marking device component forming speckle is collected by a high-speed gray scale camera; The speckle cleaning area is determined by identifying the speckle gray scale image of the surface of the marking device component based on an edge detection method; The speckle state refers to the random brightness distribution phenomenon formed when coherent light is irradiated onto a rough surface or a non-uniform medium, including speckle contrast and speckle particle size; The standard deviation and mean of the pixel gray scale in the speckle cleaning area are calculated, and the ratio of the standard deviation to the mean of the pixel gray scale in the speckle cleaning area is taken as the speckle contrast; The speckle particle size is calculated by the gray scale value of the coordinates corresponding to the speckle cleaning area.

7. The real-time early warning method for digital management and control of the whole process of electric power operation according to claim 6, characterized in that: In step S3, the speckle contrast and speckle particle size are standardized; The cleaning feature is calculated by integrating the standardized values of the speckle contrast and speckle particle size; The cleaning feature is compared with the preset cleaning feature threshold to determine whether secondary cleaning is needed: If the cleaning feature is greater than or equal to the preset cleaning feature threshold, secondary cleaning is needed; If the cleaning feature is less than the preset cleaning feature threshold, secondary cleaning is not needed.

8. The real-time early warning method for digital management and control of the whole process of electric power operation according to claim 1, characterized in that: In step S4, the contact between the cleaning device and the marking device component is detected by a contact switch sensor of the cleaning device; The contact result is sent to the recording unit through a wireless communication unit; The contact times between the cleaning device and the marking device component, i.e., the cleaning times of the marking device component, are obtained by matching the marking device and the recording unit; The surface leakage current of the marking device component is collected in real time by a high-impedance current sensor; The cleaning times and surface leakage current of the marking device component are normalized by a Max-Min normalization method; The safety feature value of the marking device component is calculated by integrating the normalized values of the cleaning times and surface leakage current of the marking device component; The mean and standard deviation of the safety feature value of the marking device component are calculated.

9. The real-time early warning method for digital management and control of the whole process of electric power operation according to claim 8, characterized in that: In step S4, the sum of the mean and standard deviation of the safety feature value of the marking device component is taken as the first threshold of the safety level; The difference between the mean and standard deviation of the safety feature value of the marking device component is taken as the second threshold of the safety level; If the safety feature value of the marking device component is less than the second threshold of the safety level, the safety level is the low safety risk level, and the current state continues to run without generating an alarm signal; If the safety feature value of the marking device component is greater than or equal to the second threshold of the safety level and less than the first threshold of the safety level, the safety level is the medium safety risk level, and a warning signal is generated; If the safety feature value of the marking device component is greater than or equal to the first threshold of the safety level, the safety level is the high safety risk level, and an emergency alarm signal is generated.

Images