Power transmission line hidden danger detection method and system based on traveling wave time-frequency diagram

Abstract

The invention discloses a power transmission line hidden danger detection method and system based on a traveling wave time-frequency diagram. The method comprises the following steps: acquiring traveling wave current data of different positions of a power transmission line; generating a time-frequency diagram of each traveling wave current data; obtaining a grey-scale map of each time-frequency map; identifying each grey-scale map, and determining a hidden danger identification result of the time-frequency map corresponding to each grey-scale map; carrying out difference on the traveling wave current data corresponding to the time-frequency diagram of which the hidden danger identification result is hidden danger and other traveling wave current data at the same moment on the same line to obtain a waveform after difference; determining the possible position of the hidden danger according to the differentiated waveform; and carrying out statistics on the potential hazard possible positions, and selecting the most potential hazard possible position as a potential hazard occurrence position. Accurate identification and position positioning of hidden dangers of the power transmission line are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of power transmission line monitoring, and in particular to a method and system for detecting hidden dangers of power transmission lines based on traveling wave time-frequency diagrams. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] The stable operation of transmission lines in the power system is crucial. However, due to factors such as long distances and complex environments, various discharge hazards may occur, such as discharge due to insulator contamination defects, discharge due to composite insulator cracking, discharge due to tree barriers, discharge due to hardware defects, and discharge due to ice defects. Discharge defects generally change from quantitative changes to qualitative changes. If no intervention is made, it will lead to line tripping events. If the hidden dangers of transmission lines can be identified and real-time alarms can be issued, it will help to timely detect and eliminate abnormal discharge hazards, thereby avoiding line tripping or shutdown, which is crucial to ensuring the safety of the power grid.

[0004] Electric energy is essentially transmitted in the form of electromagnetic waves. When voltage is applied to the transmission conductor, charged particles move in a directional manner, generating current on the transmission line, thereby establishing electric and magnetic fields. When an abnormality occurs in the line, the waves are reflected and refracted at different dielectric interfaces, generating power frequency current and traveling wave (high frequency) current. The relevant information of these waves can be used to identify hidden danger events. In the related art, there is a method of using the traveling wave hidden danger detection method to detect hidden dangers in transmission lines.

[0005] The traveling wave hidden danger detection method collects the traveling wave current of the transmission line and extracts the current signal of a specific frequency band in the current information through the data processing module, and determines whether a hidden danger occurs through the threshold. However, due to the variety of hidden danger discharge types, such as tree barrier defect discharge, most of the traveling wave main waves have obvious discharge pulse clusters, and some waveforms have multiple main waves, and the discharge is relatively dense, while the hardware defect discharge waveform is mostly a single-point pulse waveform, and its frequency is significantly higher than other types of discharge. Therefore, using the data processing module to extract the current signal of a specific frequency band in the current information cannot cover multiple types of discharge hidden dangers. The accuracy of using the empirical threshold needs to be further verified in lines with different voltage levels and is difficult to calibrate. In addition, the single range frequency band is extracted, the false alarm rate is high, and it is difficult to accurately identify different types of discharge hidden dangers. Summary of the invention

[0006] In order to solve the above problems, the present invention proposes a transmission line hidden danger detection method and system based on traveling wave time-frequency diagram, which can identify various hidden dangers of transmission lines, accurately determine the location of hidden dangers, and achieve full coverage of transmission line hidden danger detection.

[0007] To achieve the above object, the present invention adopts the following technical solution:

[0008] First, a method for detecting hidden dangers of power transmission lines based on traveling wave time-frequency diagram is proposed, including:

[0009] Obtain traveling wave current data at different locations of the transmission line;

[0010] Generate a time-frequency diagram of each traveling wave current data;

[0011] Get the grayscale image of each time-frequency graph;

[0012] Identify each grayscale image and determine the hidden danger identification result of the time-frequency image corresponding to each grayscale image;

[0013] Differentiate the traveling wave current data corresponding to the time-frequency diagram of the hidden danger identified as a hidden danger from the other traveling wave current data on the same line at the same time to obtain a differential waveform; determine the possible location of the hidden danger based on the differential waveform;

[0014] Make statistics on the possible locations of hidden dangers and select the location with the largest number of possible hidden dangers as the location where the hidden dangers occur.

[0015] Furthermore, the traveling wave current data is normalized; the normalized data is subjected to short-time Fourier transform to obtain a time-frequency diagram.

[0016] Furthermore, the trained transmission line hidden danger identification model is used to identify the grayscale image, and the hidden danger identification result of the time-frequency image corresponding to the grayscale image is determined;

[0017] Among them, the transmission line hidden danger identification model extracts multi-scale features from the grayscale image, and fully connects the extracted multi-scale features to obtain the grayscale image features; the grayscale image features are identified to determine the hidden danger identification results of the time-frequency image corresponding to the grayscale image.

[0018] Furthermore, the transmission line hidden danger identification model includes a convolutional neural network, a feature pyramid network and a path aggregation network; the convolutional neural network extracts multi-layer features of the grayscale image, and inputs each layer of extracted features into the feature pyramid network; the feature pyramid network fuses high-level features with low-level features layer by layer through a top-down path and lateral connections to obtain a multi-scale feature map; the multi-scale feature map is input into the path aggregation network; the path aggregation network fuses small-scale feature maps with large-scale feature maps layer by layer through a bottom-up path to obtain multi-scale fusion features; the multi-scale fusion features are fully connected to obtain grayscale image features; the grayscale image features are identified to determine the hidden danger identification results of the time-frequency image corresponding to the grayscale image.

[0019] Further, the starting point of the wave head is determined according to the waveform after differentiation;

[0020] According to the starting point of the wave head, determine the time difference of the fault traveling wave reaching different ends;

[0021] Based on the time difference between the fault traveling wave reaching different ends, the possible location of the hidden danger is determined using the double-end distance measurement method.

[0022] Furthermore, statistics are conducted on the possible locations of hidden dangers to obtain a normal distribution of the possible locations of hidden dangers;

[0023] Select the possible hidden danger location at the peak of the normal distribution as the location where the hidden danger occurs.

[0024] Secondly, a transmission line hidden danger detection system based on traveling wave time-frequency diagram is proposed, including:

[0025] Distributed traveling wave monitoring device, used to obtain traveling wave current data at different locations of the transmission line;

[0026] A time-frequency diagram generating module, used for generating a time-frequency diagram of each traveling wave current data;

[0027] A grayscale image conversion module is used to obtain a grayscale image of each time-frequency image;

[0028] A hidden danger identification module is used to identify each grayscale image and determine the hidden danger identification result of the time-frequency image corresponding to each grayscale image;

[0029] The potential location determination module of the hidden danger is used to differentiate the traveling wave current data corresponding to the time-frequency diagram of the hidden danger identification result and the other traveling wave current data at the same time on the same line to obtain the differential waveform; and determine the possible location of the hidden danger according to the differential waveform;

[0030] The hidden danger occurrence location determination module is used to count the possible hidden danger locations and select the possible hidden danger locations with the largest number as the hidden danger occurrence locations.

[0031] In a third aspect, a computer device is provided, the device comprising:

[0032] a processor adapted to execute a computer program;

[0033] A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by the processor, a method for detecting hidden dangers of a power transmission line based on a traveling wave time-frequency diagram proposed in the first aspect is implemented.

[0034] In a fourth aspect, a computer-readable storage medium is proposed, wherein the computer-readable storage medium stores a computer program, and the computer program is suitable for being loaded by a processor and executing a transmission line hidden danger detection method based on a traveling wave time-frequency diagram proposed in the first aspect.

[0035] In a fifth aspect, a computer program product is proposed, which includes a computer program. When the computer program is executed by a processor, it implements a transmission line hidden danger detection method based on traveling wave time-frequency diagram proposed in the first aspect.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] The present invention proposes a method and system for detecting hidden dangers in power transmission lines based on traveling wave time-frequency diagrams. The method obtains the traveling wave current of the transmission line and generates a video image of the traveling wave current, and then converts the time-frequency diagram into a grayscale image, identifies the grayscale image, and thereby determines any type of hidden danger in the transmission line. Thereafter, the traveling wave current data identified as a hidden danger is analyzed to determine the location where the hidden danger occurs. Therefore, the method described in this embodiment can identify a variety of hidden dangers in the transmission line, and can accurately determine the location of the hidden danger, thereby achieving full coverage of hidden danger detection in the transmission line.

[0038] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The drawings in the specification, which constitute a part of the present application, are used to provide further understanding of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute improper limitations on the present application.

[0040] Figure 1 A flow chart of a method for detecting hidden dangers of a power transmission line based on a traveling wave time-frequency diagram disclosed in an embodiment;

[0041] Figure 2 This is a network structure diagram of the transmission line hidden danger identification model disclosed in the embodiment. DETAILED DESCRIPTION

[0042] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0043] It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0044] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0045] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0046] Example 1

[0047] The stable operation of transmission lines in the power system is crucial. However, due to factors such as long distances and complex environments, various discharge hazards may occur, such as discharge due to insulator contamination defects, discharge due to composite insulator cracking, discharge due to tree barriers, discharge due to hardware defects, and discharge due to ice defects. Discharge defects generally change from quantitative changes to qualitative changes. If no intervention is made, it will lead to line tripping events. If the hidden dangers of transmission lines can be identified and real-time alarms can be issued, it will help to timely detect and eliminate abnormal discharge hazards, thereby avoiding line tripping or shutdown, which is crucial to ensuring the safety of the power grid.

[0048] Electric energy is essentially transmitted in the form of electromagnetic waves. When voltage is applied to the transmission conductor, the charged particles move in a directional manner, generating current on the transmission line, thereby establishing electric and magnetic fields. When an abnormality occurs in the line, the wave is reflected and refracted at different dielectric interfaces, generating power frequency current and traveling wave (high frequency) current. The relevant information of these waves can be used to identify hidden danger events. In order to monitor these events, the distributed device will regularly collect the traveling wave current, which records the event information within 10ms. For example, if a discharge hidden danger occurs, the hidden danger traveling wave current information can be captured by regular recording. The distributed device collects the power frequency current and traveling wave current at a specific time. If a hidden danger occurs, the traveling wave current will record the change characteristics of the hidden danger current, such as instantaneous frequency change and instantaneous amplitude change, which appear multiple times in a single period or multiple periods and have certain similarities. Identifying these changes through algorithms is the key to identifying hidden dangers. Due to the continuity and intermittent nature of the discharge after the hidden danger occurs, the double-end traveling wave ranging method is used after the hidden danger is identified. The location of the hidden danger is highly repeatable for multiple positioning, and it can be determined whether a hidden danger event has occurred. The current methods used to identify hidden dangers in transmission lines mainly include: handheld monitor line inspection method, ultraviolet searchlight method, drone inspection image detection method and traveling wave hidden danger detection method.

[0049] The handheld monitor line inspection method uses a handheld detector by the inspector to detect discharge characteristics along the line. This method requires a large amount of work and cannot guarantee whether it is missed, especially for lines running in mountainous and plateau areas, where it is difficult to inspect the lines; the ultraviolet detection method determines the location of the hidden danger point by detecting ultraviolet signals. When a leakage fault occurs in the line, an ultraviolet signal will be generated. By receiving these signals and analyzing them, the location of the hidden danger point can be determined. This method has high accuracy and sensitivity, but requires special equipment and a high level of technology, and is difficult to implement, with high labor and economic costs; the drone inspection image detection method uses an image acquisition device to collect on-site environmental images of the transmission line, identify the on-site environmental images, and determine the hidden dangers and their locations. For example, patent CN113705364B discloses an artificial intelligence-based transmission line external hidden danger early warning system and method, which belongs to the field of transmission line safety monitoring. The method is characterized in that it includes a front-end device installed on the tower of the high-voltage transmission line. The front-end device is connected to the monitoring substation host, the monitoring substation host is connected to the remote background monitoring system, the monitoring substation host and the background monitoring system are connected to the neural network module, the front-end device is used to collect on-site environmental images, and send the collected data to the monitoring substation host. The monitoring substation host pre-processes and intelligently analyzes the image to determine whether there is foreign invasion, and controls the front-end device to issue an on-site alarm; Patent CN114863118A discloses a self-learning system identification system and method based on the external hidden dangers of transmission lines, belonging to the field of transmission line safety monitoring technology. The edge computing platform of the method collects images of the surrounding environment of the transmission line, and uses a hidden danger target detection model based on a self-learning mechanism to identify channel hidden dangers and determine the location of channel hidden dangers. The above two methods both use image acquisition equipment to collect information on the transmission line, obtain information and features through the image mode, and use artificial intelligence algorithms at the edge or monitoring host to analyze hidden dangers and issue warnings. However, due to the variety of hidden dangers in transmission lines, their sizes and intensities, and the influence of weather conditions such as changes in light intensity and angle, and fog, it is difficult to ensure image quality and accurately identify small target hidden dangers. In addition, the locations of hidden dangers are relatively random. Due to factors such as occlusion by different structures of poles and towers, it is difficult to cover the poles and lines from all angles through monitoring. It has great limitations and cannot accurately monitor the occurrence of hidden dangers in real time. The traveling wave hidden danger detection method collects the traveling wave current of the transmission line and extracts the current signal of a specific frequency band in the current information through a data processing module, and determines whether a hidden danger occurs through a threshold. For example, patent CN112345883A discloses a transmission line hidden danger early warning system based on traveling wave current, which relates to the field of transmission line monitoring technology. The system collects traveling wave current and extracts the current signal of a specific frequency band in the current information through a data processing module, and determines whether a hidden danger occurs through a threshold. This method overcomes the method of obtaining single information through monitoring images and realizes diversified monitoring.However, due to the variety of hidden danger discharge types, such as tree barrier defect discharge, most of the traveling wave main waves have obvious discharge pulse clusters, some waveforms have multiple main waves, and the discharge is relatively dense, while the hardware defect discharge waveform is mostly single-point pulse waveform, and its frequency is significantly higher than other types of discharge. Therefore, using the data processing module to extract the current signal of a specific frequency band in the current information cannot cover various types of discharge hazards. The accuracy of using the empirical threshold needs to be further verified in lines with different voltage levels and is difficult to calibrate. In addition, the false alarm rate is high when extracting a single range frequency band, and it is difficult to accurately identify different types of discharge hazards.

[0050] In order to solve the technical problems existing in the related art, in this embodiment, a method for detecting hidden dangers of a power transmission line based on a traveling wave time-frequency diagram is disclosed.

[0051] like Figure 1 As shown, the present embodiment discloses a method for detecting hidden dangers of a power transmission line based on a traveling wave time-frequency diagram, comprising:

[0052] S1: Obtain traveling wave current data at different locations of the transmission line.

[0053] This embodiment uses a distributed traveling wave monitoring device to obtain traveling wave current data at different locations of the transmission line. The sampling frequency of the traveling wave monitoring device is to collect traveling wave current data once every hour, and multiple sets of traveling wave monitoring devices in the line record waves simultaneously.

[0054] S2: Generate a time-frequency diagram of each traveling wave current data.

[0055] In this embodiment, the traveling wave current data is normalized; and the normalized data is subjected to short-time Fourier transform to obtain a time-frequency diagram.

[0056] The process of obtaining the time-frequency diagram of traveling wave current data specifically includes:

[0057] Normalize the maximum and minimum values ​​of the traveling wave current data to obtain normalized data; add a window function to the normalized data, divide the normalized data into segments, and obtain multiple segments of data; perform short-time Fourier transform (STFT) on each segment of data. STFT has a parameter N_fft, which indicates how many points in a frame are subjected to Fourier transform FFT. Each point corresponds to a frequency interval, and the frequency intervals are equally spaced, that is, the frequency resolution freq bin =F s / N fft In this way, the time variable and frequency distribution of the fluctuation can be analyzed simultaneously, thereby obtaining a heat map describing how the frequency components of the fluctuation change over time, also known as a time-frequency map. The time-frequency map is based on the time-frequency analysis method, and the value of the third dimension is presented in the form of a heat map with the depth of color. The change in the depth of color records the frequency change characteristics.

[0058] S3: Get the grayscale image of each time-frequency image.

[0059] This embodiment extracts the grayscale features of the time-frequency graph and regularizes the grayscale features to obtain a grayscale graph of the time-frequency graph.

[0060] S4: Identify each grayscale image and determine the hidden danger identification result of the time-frequency image corresponding to each grayscale image.

[0061] This embodiment uses the trained transmission line hidden danger identification model to identify the grayscale image and determine the hidden danger identification result of the time-frequency image corresponding to the grayscale image;

[0062] Among them, the transmission line hidden danger identification model extracts multi-scale features from the grayscale image, and fully connects the extracted multi-scale features to obtain the grayscale image features; the grayscale image features are identified to determine the hidden danger identification results of the time-frequency image corresponding to the grayscale image.

[0063] like Figure 2 As shown, the transmission line hidden danger identification model constructed in this embodiment includes a convolutional neural network, a feature pyramid network and a path aggregation network; the convolutional neural network extracts multi-layer features of the grayscale image, and inputs each layer of the extracted features into the feature pyramid network; the feature pyramid network fuses high-level features with low-level features layer by layer through a top-down path and lateral connection to obtain a multi-scale feature map; the multi-scale feature map is input into the path aggregation network; the path aggregation network fuses small-scale feature maps with large-scale feature maps layer by layer through a bottom-up path to obtain multi-scale fusion features; the multi-scale fusion features are fully connected to obtain grayscale image features; the grayscale image features are identified to determine the hidden danger identification result of the time-frequency image corresponding to the grayscale image.

[0064] Specifically, the transmission line hidden danger identification model takes the grayscale image of the time-frequency graph as input and the transmission line hidden danger identification model as output. The convolutional neural network uses ResNet. The grayscale image input into the transmission line hidden danger identification model first enters ResNet for feature extraction. ResNet inputs the features extracted by each feature extraction layer into the feature pyramid network (FPN). FPN fuses high-level features with low-level features layer by layer through top-down paths and lateral connections to obtain multi-scale feature maps; the multi-scale feature map generated by FPN is input into the path aggregation network (PAN). PAN fuses small-scale feature maps with large-scale feature maps layer by layer through a bottom-up path to obtain multi-scale fusion features. This step enhances the information transmission of high-level features to low-level features and promotes the comprehensive fusion of multi-scale features. The multi-scale features are input into the fully connected layer for full connection to obtain grayscale features. The grayscale features are processed by the Softmax function to obtain the hidden danger identification result of the time-frequency graph corresponding to the grayscale image.

[0065] Among them, the hidden danger identification result is the identified hidden danger type, including all possible fault hidden danger types that may occur in the transmission line, such as insulator contamination discharge defects, composite insulator cracking discharge, tree barrier discharge defects, hardware discharge defects, ice discharge defects and other defect types. When there is no hidden danger in the transmission line, the hidden danger identification result is non-hidden danger.

[0066] After constructing the transmission line hidden danger identification model, this embodiment uses the traveling wave current monitoring data of known hidden danger types as training data to generate a time-frequency graph of the training data, and obtains a grayscale graph of the time-frequency graph of the training data; the hidden danger types are labeled on the grayscale graph; the constructed transmission line hidden danger identification model is trained using the labeled grayscale graph, and after the training is completed, a trained transmission line hidden danger identification model is obtained.

[0067] S5: Differentiate the traveling wave current data corresponding to the time-frequency diagram of the hidden danger identified as a hidden danger from the remaining traveling wave current data on the same line at the same time to obtain a differential waveform; determine the possible location of the hidden danger based on the differential waveform.

[0068] In this embodiment, the process of determining the possible location of the hidden danger according to the differential waveform includes:

[0069] Determine the starting point of the wave head based on the differential waveform;

[0070] According to the starting point of the wave head, determine the time difference of the fault traveling wave reaching different ends;

[0071] Based on the time difference between the fault traveling wave reaching different ends, the possible location of the hidden danger is determined using the double-end distance measurement method.

[0072] Since the frequency or amplitude will change when the hidden danger occurs, the original traveling wave current data is differentiated. After the normal waveform is differentiated, the waveform approaches zero, while the waveform with rapid frequency or amplitude changes will show large oscillations after differentiation. At this time, the starting point of the waveform after differentiation is detected. After the starting point of the wave head is detected, the start time of the hidden danger can be calculated according to the sampling rate, accurate to microseconds (us). Since multiple sets of equipment are installed in the line and recorded at the same time, the traveling wave current data of different ends where the hidden danger is identified are selected to determine the time when the fault traveling wave reaches different ends, and the time difference when the fault traveling wave arrives at different ends is calculated. The distance between the different ends where the hidden danger is identified and the time difference when the fault traveling wave arrives at different ends are analyzed using the two-end ranging method to determine the possible location of the hidden danger. The fault traveling wave data generated by the hidden danger point C is transmitted along the transmission line to the recording devices at both ends at a speed v, and the time when it arrives at the M end and the N end on both sides is t respectively. M and t N , then the distance from the possible fault point to the M end is L M , and the distance L from the fault point to the N end N , respectively LM =(L+v*(t M -t N )) / 2,L N =(Lv*(t M -L N )) / 2, where L M , L N is the distance between the potential hidden danger point and the M and N ends, and L is the distance between point M and point N.

[0073] S6: Count the possible locations of hidden dangers and select the possible locations of hidden dangers with the largest number as the locations where hidden dangers occur.

[0074] In this embodiment, statistics are collected on possible locations of hidden dangers to obtain a normal distribution of possible locations of hidden dangers; the possible location of hidden dangers at the peak of the normal distribution is selected as the location where the hidden danger occurs.

[0075] This embodiment performs statistics on the possible locations of identified hidden dangers and the corresponding traveling wave current data. Due to the system error of ranging, 1 microsecond is about 295 meters (the speed of light is 300 km / s), so the ranging result error is ±300m. Since the waveforms of hidden dangers occurring at each moment are not exactly the same, errors will occur in the ranging of hidden dangers at the same location at different moments. After multiple statistics, the possible location-frequency relationship of hidden dangers is obtained. The statistics of possible location-frequency of hidden dangers show a normal distribution trend. The peak value of the normal distribution, that is, the point where the hidden danger occurs the most times, and the possible location of the hidden danger corresponding to this point is the final location where the hidden danger occurs.

[0076] The present embodiment discloses a method for detecting hidden dangers in power transmission lines based on traveling wave time-frequency diagrams. The method obtains traveling wave currents of the transmission lines and generates video images of the traveling wave currents, and then converts the time-frequency diagrams into grayscale images, identifies the grayscale images, and thereby determines hidden dangers of any type in the power transmission lines. Thereafter, the traveling wave current data identified as hidden dangers are analyzed to determine the locations where the hidden dangers occur. Therefore, the method described in the present embodiment can identify a variety of hidden dangers in the transmission lines and accurately determine the locations where the hidden dangers occur. Compared with image analysis methods that identify hidden dangers from the essential characteristics of line currents, more comprehensive coverage of transmission lines and more efficient identification can be achieved. The traveling wave monitoring device has a long monitoring distance, and greatly reduces the economic cost compared to image monitoring.

[0077] This embodiment uses a residual network combined with a feature pyramid (FPN) structure and a path aggregation network (PAN) to extract features from grayscale images, achieving efficient fusion of multiple features of grayscale images and significantly improving the network's feature extraction and classification performance. Through the fusion of multi-scale features, the network has higher robustness in the face of complex scenes and noise, can accurately identify various types of hidden danger waveforms, and has a fast algorithm speed and a wide range of applications.

[0078] The method disclosed in this embodiment can identify more detailed hidden danger types and locate the hidden dangers, providing richer information for line operation and maintenance, thereby being able to deal with the hidden dangers of the transmission line in a timely manner and better ensure the stable operation of the transmission line.

[0079] The method disclosed in this embodiment can be run in front-end devices, distributed computing devices, large-scale computing devices and other devices, and can realize automatic monitoring and early warning, greatly reducing manpower and economic costs.

[0080] Example 2

[0081] In this embodiment, a transmission line hidden danger detection system based on traveling wave time-frequency diagram is disclosed, comprising:

[0082] Distributed traveling wave monitoring device, used to obtain traveling wave current data at different locations of the transmission line;

[0083] A time-frequency diagram generating module, used for generating a time-frequency diagram of each traveling wave current data;

[0084] A grayscale image conversion module is used to obtain a grayscale image of each time-frequency image;

[0085] A hidden danger identification module is used to identify each grayscale image and determine the hidden danger identification result of the time-frequency image corresponding to each grayscale image;

[0086] The potential location determination module of the hidden danger is used to differentiate the traveling wave current data corresponding to the time-frequency diagram of the hidden danger identification result and the other traveling wave current data at the same time on the same line to obtain the differential waveform; and determine the possible location of the hidden danger according to the differential waveform;

[0087] The hidden danger occurrence location determination module is used to count the possible hidden danger locations and select the possible hidden danger locations with the largest number as the hidden danger occurrence locations.

[0088] The present invention also discloses a computer device, which includes:

[0089] a processor adapted to execute a computer program;

[0090] A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by the processor, a method for detecting hidden dangers of a power transmission line based on a traveling wave time-frequency diagram disclosed in Example 1 is implemented.

[0091] The present invention also discloses a computer-readable storage medium storing a computer program, wherein the computer program is suitable for being loaded by a processor and executing a method for detecting hidden dangers of a power transmission line based on a traveling wave time-frequency diagram disclosed in Example 1.

[0092] The present invention also discloses a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements a transmission line hidden danger detection method based on a traveling wave time-frequency diagram disclosed in Example 1.

[0093] The method disclosed in Example 1 can be directly embodied as a hardware processor, or can be executed by a combination of hardware and software modules in the processor. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware. To avoid repetition, it will not be described in detail here.

[0094] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with this embodiment can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0095] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A method for detecting hidden dangers of power transmission lines based on traveling wave time-frequency diagram, characterized in that: include: Obtain traveling wave current data at different locations of the transmission line; Generate a time-frequency diagram of each traveling wave current data; Get the grayscale image of each time-frequency graph; Identify each grayscale image and determine the hidden danger identification result of the time-frequency image corresponding to each grayscale image; Differentiate the traveling wave current data corresponding to the time-frequency diagram of the hidden danger identified as a hidden danger with the remaining traveling wave current data at the same time on the same line to obtain a differential waveform; Determine the possible location of hidden dangers based on the differential waveform; Make statistics on the possible locations of hidden dangers and select the location with the largest number of possible hidden dangers as the location where the hidden dangers occur.

2. A method for detecting hidden dangers of power transmission lines based on traveling wave time-frequency diagram according to claim 1, characterized in that: The traveling wave current data is normalized; the normalized data is short-time Fourier transformed to obtain a time-frequency diagram.

3. A method for detecting hidden dangers of power transmission lines based on traveling wave time-frequency diagram according to claim 1, characterized in that: The grayscale image is identified using the trained transmission line hidden danger identification model to determine the hidden danger identification result of the time-frequency image corresponding to the grayscale image; Among them, the transmission line hidden danger identification model extracts multi-scale features from the grayscale image, and fully connects the extracted multi-scale features to obtain the grayscale image features; the grayscale image features are identified to determine the hidden danger identification results of the time-frequency image corresponding to the grayscale image.

4. A method for detecting hidden dangers of power transmission lines based on traveling wave time-frequency diagram as claimed in claim 3, characterized in that: The transmission line hidden danger identification model includes convolutional neural network, feature pyramid network and path aggregation network; the convolutional neural network extracts multi-layer features from the grayscale image and inputs each layer of extracted features into the feature pyramid network; The feature pyramid network fuses high-level features with low-level features layer by layer through top-down paths and lateral connections to obtain multi-scale feature maps; the multi-scale feature maps are input into the path aggregation network; the path aggregation network fuses small-scale feature maps with large-scale feature maps layer by layer through bottom-up paths to obtain multi-scale fusion features; the multi-scale fusion features are fully connected to obtain grayscale features; The grayscale image features are identified to determine the hidden danger identification results of the time-frequency image corresponding to the grayscale image.

5. The method for detecting hidden dangers of power transmission lines based on traveling wave time-frequency diagram according to claim 1, characterized in that: Determine the starting point of the wave head based on the differential waveform; According to the starting point of the wave head, determine the time difference of the fault traveling wave reaching different ends; Based on the time difference between the fault traveling wave reaching different ends, the possible location of the hidden danger is determined using the double-end distance measurement method.

6. A method for detecting hidden dangers of power transmission lines based on traveling wave time-frequency diagram according to claim 1, characterized in that: Conduct statistics on possible locations of hidden dangers to obtain normal distribution of possible locations of hidden dangers; Select the possible hidden danger location at the peak of the normal distribution as the location where the hidden danger occurs.

7. A transmission line hidden danger detection system based on traveling wave time-frequency diagram, characterized in that: include: Distributed traveling wave monitoring device, used to obtain traveling wave current data at different locations of the transmission line; A time-frequency diagram generating module, used for generating a time-frequency diagram of each traveling wave current data; A grayscale image conversion module is used to obtain a grayscale image of each time-frequency image; A hidden danger identification module is used to identify each grayscale image and determine the hidden danger identification result of the time-frequency image corresponding to each grayscale image; A module for determining the possible location of a hidden danger is used to differentiate the traveling wave current data corresponding to the time-frequency diagram of the hidden danger identification result with the rest of the traveling wave current data at the same time on the same line to obtain a differential waveform; Determine the possible location of hidden dangers based on the differential waveform; The hidden danger occurrence location determination module is used to count the possible hidden danger locations and select the possible hidden danger locations with the largest number as the hidden danger occurrence locations.

8. An electronic device, characterized in that: The device comprises: a processor adapted to execute a computer program; A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by the processor, a method for detecting hidden dangers of a power transmission line based on a traveling wave time-frequency diagram as described in any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is suitable for being loaded by a processor and executing a method for detecting hidden dangers of a power transmission line based on a traveling wave time-frequency diagram as described in any one of claims 1 to 6.

10. A computer program product, characterized in that The computer program product includes a computer program, and when the computer program is executed by a processor, it implements the method for detecting hidden dangers of a power transmission line based on a traveling wave time-frequency diagram as described in any one of claims 1 to 6.

Images