Power transmission line fault monitoring method, device, medium and apparatus

Abstract

A power transmission line fault monitoring method, device, medium and equipment are disclosed, which calculates the root mean square values of current and electric field data collected in real time, and calls a calculation model to generate a similarity vector of current and electric field when the root mean square values continue to drop by more than or equal to a preset threshold. Then, the similarity confidence is calculated by matching the degree of the two similarity vectors. If the confidence is greater than or equal to the preset threshold, it is determined that the line has a fault. This method can more accurately detect faults, reduce false positives, and improve the accuracy and efficiency of fault identification by comprehensively analyzing the changes in current and electric field signals.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power transmission lines, in particular to a fault monitoring method, device, medium and equipment for power transmission lines. BACKGROUND

[0002] Power transmission lines are facing increasingly severe forest fire threats, including frequent natural forest fires due to climate change, forest fires caused by power grid failures and improper human operations, and increased fire risks due to external environmental changes. These factors can cause line damage, large-scale power outages and safety hazards, forming a vicious cycle of forest fires and line failures. However, existing forest fire monitoring technologies have the disadvantages of not timely monitoring, low accuracy and lack of intelligent analysis means, and cannot effectively cope with these threats. SUMMARY

[0003] Therefore, it is necessary to provide a fault monitoring method, device, medium and equipment for power transmission lines to solve the problems of existing forest fire monitoring technologies, such as not timely monitoring, low accuracy and lack of intelligent analysis means.

[0004] A fault monitoring method for a power transmission line, the method comprising:

[0005] real-time acquisition of current values and electric field values of the power transmission line, calculation of a plurality of first root mean square values of the current values and calculation of a plurality of second root mean square values of the electric field values within a preset number of periods;

[0006] if the plurality of first root mean square values continuously decrease and the maximum decrease amplitude is greater than or equal to a preset first decrease amplitude, and the plurality of second root mean square values continuously decrease and the maximum decrease amplitude is greater than or equal to a preset second decrease amplitude, then a current calculation model is called to calculate a current similarity vector according to the plurality of first root mean square values, and an electric field calculation model is called to calculate an electric field similarity vector according to the plurality of second root mean square values;

[0007] calculation of a similarity confidence according to the matching degree between the current similarity vector and the electric field similarity vector, and if the similarity confidence is greater than or equal to a preset similarity threshold, then it is determined that the power transmission line currently has a fault.

[0008] In one of the embodiments, the method further comprises:

[0009] real-time acquisition of acceleration values of the power transmission line, calculation of a plurality of third root mean square values of the current values within a preset number of periods;

[0010] calling an acceleration calculation model to calculate an acceleration similarity vector according to the plurality of third root mean square values;

[0011] If the first distance between the acceleration similarity vector and the current similarity vector and the second distance between the acceleration similarity vector and the electric field similarity vector are both less than a preset distance threshold, a preset floating confidence is added on the basis of the calculated similarity confidence.

[0012] In one of the embodiments, the method further comprises:

[0013] The current value and the electric field value are segmented to obtain a plurality of electrical segments.

[0014] The total harmonic distortion and the root mean square value of each electrical segment are calculated, and the electrical segment with the total harmonic distortion greater than or equal to a preset first threshold value and / or the root mean square value greater than or equal to a preset second threshold value is removed from the plurality of electrical segments.

[0015] In one of the embodiments, the method further comprises:

[0016] The sample electric field data and the sample current data are obtained as a training set, wherein each data is labeled by a sample label.

[0017] The electric field rise rate and the electric field drop rate are extracted as electric field features from the training set, and the current rise rate, the current drop rate and the surge current are extracted as current features.

[0018] The electric field features are input into a first classifier based on a decision tree, and the current features are input into a second classifier based on a decision tree, and a predicted label output is obtained.

[0019] The first classifier and the second classifier are adjusted according to the difference between the sample label and the predicted label until the first classifier and the second classifier converge, the first classifier is taken as an electric field calculation model, and the second classifier is taken as a current calculation model.

[0020] A fault monitoring device, comprising:

[0021] A root mean square value calculation module is configured to collect current values and electric field values of a power transmission line in real time, calculate a plurality of first root mean square values of the current values, and calculate a plurality of second root mean square values of the electric field values within a preset number of periods.

[0022] A similarity vector calculation module is configured to, if the plurality of first root mean square values continuously decrease and the maximum decrease amplitude is greater than or equal to a preset first decrease amplitude, and the plurality of second root mean square values continuously decrease and the maximum decrease amplitude is greater than or equal to a preset second decrease amplitude, call a current calculation model to calculate a current similarity vector according to the plurality of first root mean square values, and call an electric field calculation model to calculate an electric field similarity vector according to the plurality of second root mean square values.

[0023] a fault judging module configured to calculate a similarity confidence according to a matching degree between the current similarity vector and the electric field similarity vector, and determine that the power transmission line currently has a fault if the similarity confidence is greater than or equal to a preset similarity threshold.

[0024] In one of the embodiments, the fault monitoring device further comprises:

[0025] a confidence adjusting module configured to collect acceleration values of the power transmission line in real time, calculate a plurality of third root mean square values of the current values within a preset number of periods, call an acceleration calculation model to calculate an acceleration similarity vector according to the plurality of third root mean square values, and increase a preset floating confidence on the basis of the calculated similarity confidence if a first distance between the acceleration similarity vector and the current similarity vector and a second distance between the acceleration similarity vector and the electric field similarity vector are both less than a preset distance threshold.

[0026] In one of the embodiments, the fault monitoring device further comprises:

[0027] a filtering module configured to segment the current values and the electric field values to obtain a plurality of electrical segments, calculate a total harmonic distortion and a root mean square value of each electrical segment, and remove an electrical segment with a total harmonic distortion greater than or equal to a preset first threshold value and / or a root mean square value greater than or equal to a preset second threshold value from the plurality of electrical segments.

[0028] In one of the embodiments, the fault monitoring device further comprises:

[0029] a model training module configured to obtain sample electric field data and sample current data as a training set, wherein each data is labeled by a sample label, extract an electric field rise rate and an electric field drop rate as electric field features and extract a current rise rate, a current drop rate and a surge current as current features from the training set, input the electric field features into a first classifier based on a decision tree, input the current features into a second classifier based on a decision tree, obtain a predicted label output, adjust the first classifier and the second classifier according to a difference between the sample label and the predicted label until the first classifier and the second classifier converge, and take the first classifier as an electric field calculation model and take the second classifier as a current calculation model.

[0030] A computer readable storage medium storing a computer program, the computer program being executed by a processor to cause the processor to perform the steps of the fault monitoring method of the power transmission line.

[0031] A terminal device comprises a memory and a processor, the memory stores a computer program, and the computer program is executed by the processor to make the processor execute the steps of the power transmission line fault monitoring method.

[0032] The present application provides a power transmission line fault monitoring method, device, medium and equipment, by real-time acquisition of current and electric field data, the root mean square value is calculated respectively, and when the root mean square value continues to decline greater than or equal to the preset threshold, the similarity vector of current and electric field is generated by calling the calculation model. Then, by matching the degree of two similarity vectors, the similarity confidence is calculated. If the confidence is greater than or equal to the preset threshold, it is determined that the line has a fault. This method can more accurately detect faults, reduce false positives, and improve the accuracy and efficiency of fault identification by comprehensively analyzing the changes of current and electric field signals. BRIEF DESCRIPTION OF DRAWINGS

[0033] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0034] Among them:

[0035] Figure 1 The flowchart of the power transmission line fault monitoring method;

[0036] Figure 2 The structure diagram of the power transmission line fault monitoring device;

[0037] Figure 3 The structure block diagram of the terminal device. DETAILED DESCRIPTION

[0038] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0039] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0040] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0041] like Figure 1 As shown, Figure 1 This is a flowchart illustrating a fault monitoring method for transmission lines in one embodiment. The steps provided by the fault monitoring method for transmission lines in this embodiment include:

[0042] S101 collects the current and electric field values ​​of the transmission line in real time, and calculates multiple first root mean square values ​​of the current value and multiple second root mean square values ​​of the electric field value within a preset number of cycles.

[0043] Here, current refers to the current intensity in the transmission line; changes in current can reflect line load, short circuits, or other anomalies. Electric field value refers to the electric field intensity around the transmission line; abnormal changes in the electric field indicate insulation faults or other problems in the line. The preset number of cycles refers to the time window during which the system collects data and calculates the root mean square (RMS) value. The RMS value is the effective value of the current or electric field waveform, representing the average power of the AC signal within one cycle. For a set of continuous discrete data (such as current, voltage, etc.), the formula for calculating its RMS value is:

[0044]

[0045] Where N is the total number of data points; x i It is the value of the i-th data point.

[0046] Optionally, a fixed number of cycles can be preset, such as 5 cycles, each lasting 30 seconds, for continuous data acquisition and calculation. At the end of each cycle, the root mean square values ​​of the corresponding current and electric field are calculated.

[0047] Optionally, the number of monitoring periods is adaptively adjusted according to the load or operating state of the line. For example, the system can increase the number of monitoring periods when the load changes greatly, from the initial 5 periods to 10 periods, so as to better capture fluctuations.

[0048] S102, if the plurality of first root mean square values continue to decrease and the maximum decrease amplitude is greater than or equal to a preset first decrease amplitude, and the plurality of second root mean square values continue to decrease and the maximum decrease amplitude is greater than or equal to a preset second decrease amplitude, a current calculation model is called to calculate a current similarity vector according to the plurality of first root mean square values, and an electric field calculation model is called to calculate an electric field similarity vector according to the plurality of second root mean square values.

[0049] In the embodiment, the first decrease amplitude can be set to 60%, and the second decrease amplitude can be set to 50%. The current similarity vector and the electric field similarity vector are feature vectors generated by the calculation model based on the root mean square values of the current and the electric field, and are used to quantify and analyze the similarity between the current monitored signal characteristics and the standard fault characteristics.

[0050] For example, assuming that the normal value of the current of the transmission line is A amperes, and the normal value of the electric field is B volts / meter. In 5 consecutive periods, the system observes that the current value changes as follows: the first period current root mean square value: A x 0.85, the second period current root mean square value: A x 0.75, the third period current root mean square value: A x 0.60, the fourth period current root mean square value: A x 0.50, and the fifth period current root mean square value: A x 0.40. And it is observed that the electric field value changes as follows: the first period electric field root mean square value: B x 0.90, the second period electric field root mean square value: B x 0.80, the third period electric field root mean square value: B x 0.70, the fourth period electric field root mean square value: B x 0.60, and the fifth period electric field root mean square value: B x 0.50. Then since the current value and the electric field value meet the conditions of S102, it can be preliminarily judged that the line has a fault, but further confirmation is required, so the current calculation model is called to generate a current similarity vector according to the current root mean square values of the 5 periods, and the electric field calculation model is called to generate an electric field similarity vector according to the electric field root mean square values of the 5 periods.

[0051] S103, a similarity confidence is calculated according to the matching degree between the current similarity vector and the electric field similarity vector, and if the similarity confidence is greater than or equal to a preset similarity threshold, it is determined that the transmission line currently has a fault.

[0052] The similarity confidence represents the degree of similarity between the time-varying of one signal and another signal. In other words, it can help determine whether certain features of one signal are reproduced in another signal in some form. The similarity confidence is used to detect the delay, phase difference or common characteristics between signals. For example, if a line fault causes the electric field and current to appear abnormal at the same time, the similarity confidence will increase significantly, thereby further determining the line fault. The preset similarity threshold is a set value for determining the fault. Optionally, the similarity confidence between the current similarity vector and the electric field similarity vector can be calculated by cosine similarity, Euclidean distance or other correlation metrics.

[0053] When the similarity confidence is greater than or equal to the threshold, the system can consider that the correlation between the electric field and current signals is high enough, and the fault detection result is reliable, thereby determining that the line has a fault.

[0054] The above power line fault monitoring method, by collecting current and electric field data in real time, respectively calculating the root mean square values, and when the root mean square values continue to decrease by more than or equal to the preset threshold, calling the calculation model to generate the similarity vectors of current and electric field. Then, by matching the degree of the two similarity vectors, the similarity confidence is calculated. If the confidence is greater than or equal to the preset threshold, it is determined that the line has a fault. This method can more accurately detect faults, reduce false positives, and improve the accuracy and efficiency of fault identification by comprehensively analyzing the changes of current and electric field signals.

[0055] In one specific embodiment, the above method further performs the following steps:

[0056] A1, real-time acquisition of acceleration values of the power line, and calculation of a plurality of third root mean square values of the current values within a preset number of periods.

[0057] Wherein, the acceleration value refers to the real-time acquisition of line vibration or motion data by the acceleration sensor installed on the power line. These data can reflect the changes of the line affected by the external environment, such as wind, vibration, etc. The preset number of periods here is consistent with S101, and the root mean square value calculation method is also consistent, so it will not be repeated here.

[0058] A2, calling an acceleration calculation model to calculate an acceleration similarity vector according to the plurality of third root mean square values.

[0059] Wherein, the acceleration similarity vector reflects the trend of the acceleration signal over time.

[0060] A3, if the first distance between the acceleration similarity vector and the current similarity vector, and the second distance between the acceleration similarity vector and the electric field similarity vector are both less than a preset distance threshold, then a preset floating confidence is added to the calculated similarity confidence.

[0061] The first distance and the second distance refer to the difference between the acceleration similarity vector and the current similarity vector, the electric field similarity vector, which is usually calculated by a certain mathematical distance (such as Euclidean distance) to represent the similarity of the acceleration data and the current and electric field data. If both the first distance and the second distance are less than the preset distance threshold, the possibility of system failure is further increased, and a preset floating confidence is added on the basis of the calculated similarity confidence, so that by introducing the acceleration data and comprehensively analyzing the current and electric field signals, the line state can be more accurately reflected, and the response to the failure caused by wind and other environmental factors is more sensitive.

[0062] In a specific embodiment, the above method further performs the following steps:

[0063] B1, segmenting the current value and the electric field value to obtain a plurality of electrical segments.

[0064] Among them, the electrical segment refers to the independent data segment obtained by segmenting the current value and the electric field value according to time or other conditions. We can segment the current value and the electric field value according to fixed time slices (for example, each 30 seconds as a segment) to calculate each electrical segment. Also according to the characteristics of the current and electric field signals, such as the frequency or amplitude change of signal fluctuation, the electrical segment is automatically segmented.

[0065] B2, calculating the total harmonic distortion and the root mean square value of each electrical segment, and removing the electrical segment whose total harmonic distortion is greater than or equal to a preset first critical value and / or whose root mean square value is greater than or equal to a preset second critical value in the plurality of electrical segments.

[0066] The total harmonic distortion (THD) and the root mean square value (RMS) of each electrical segment are calculated respectively. For the segment whose THD is greater than the preset first critical value and / or whose RMS value is greater than the second critical value, it is removed, and the remaining segment is used as valid data for further fault analysis. This helps to avoid the interference of abnormal fluctuation segments on fault judgment, and improves the stability and fault recognition accuracy of the system.

[0067] In a specific embodiment, the above method further performs the following steps:

[0068] C1, obtaining sample electric field data and sample current data as a training set.

[0069] Among them, the sample electric field data and the sample current data are the data set used to train the model, which is usually the electric field and current value obtained by historical monitoring or experiment. Each data label is marked by a sample label, indicating whether each group of samples belongs to a normal state or a fault state.

[0070] C2, extract the electric field rise rate and the electric field drop rate from the training set as electric field features, and extract the current rise rate, current drop rate and inrush current as current features.

[0071] wherein the electric field rise rate and the electric field drop rate respectively represent the rate of increase or decrease of the electric field intensity within a certain time, and are used to capture the dynamic characteristics of the electric field change. The current rise rate and the current drop rate respectively represent the rise or fall rate of the current intensity within a certain time, and can reflect the trend of the current signal change. The inrush current refers to the phenomenon that the current suddenly increases in a short time, which usually occurs when the system is abnormal or the device is started.

[0072] C3, input the electric field features into a first classifier based on a decision tree, and input the current features into a second classifier based on a decision tree, and obtain the output prediction label.

[0073] The decision tree is a tree structure-based classification algorithm that gradually divides data through a series of "yes / no" questions. Each node of the decision tree represents a judgment condition of a feature, and finally outputs a classification result. In this method, the first classifier is specifically used to process the electric field features, and the second classifier is used to process the current features. The decision tree can identify the normal or fault state from these features.

[0074] The extracted electric field signal features (such as rise rate, drop rate, etc.) are input into the first classifier, and the classifier judges the state (normal or fault) corresponding to these features according to the rules learned in the training process. Similarly, the current signal features (such as inrush current, etc.) are input into the second classifier, and the classifier predicts whether the line is in a normal state or a fault state according to the different values of the current features.

[0075] The prediction label is the judgment result given by the classifier. Through the calculation of the decision tree, the first classifier gives the state label corresponding to the electric field features, and the second classifier gives the state label corresponding to the current features. These labels are usually "normal" or "fault".

[0076] In this way, by using the decision tree to independently classify the electric field and current features, the accuracy of classification can be improved, and the model of fault identification is more interpretable. In this way, the system can quickly and accurately judge the state of the line, reducing the risk of failure not being discovered in time.

[0077] C4, adjust the first classifier and the second classifier according to the difference between the sample label and the prediction label until the first classifier and the second classifier converge, and take the first classifier as the electric field calculation model and the second classifier as the current calculation model.

[0078] The difference refers to the difference between the sample label and the predicted label. If the prediction result of the classifier is inconsistent with the sample label, it means that the model is not accurate enough and needs to be further adjusted. By adjusting the first classifier and the second classifier through the difference, the system can reduce the error of the classifier prediction and ensure that the model classification of the input features is more and more accurate. When the difference between the prediction result of the classifier and the real sample label reaches the minimum, and further adjustment no longer significantly improves the accuracy of the model, the model is considered to have "converged". At this time, the classifier has learned how to identify faults or normal states from feature data, and the first classifier is used as an electric field calculation model and the second classifier is used as a current calculation model.

[0079] In this way, by continuously adjusting the classifier to match the actual sample label, the accuracy of fault identification can be greatly improved. After the classifier converges, the system can detect the fault condition in the power transmission line in real time and stably, thereby reducing the false positive rate and the false negative rate and improving the reliability and intelligent level of the system. The process of model training and optimization can ensure that the fault detection of electric field and current data is more accurate and timely.

[0080] In one embodiment, as shown in FIG. 1, a fault monitoring device for a power transmission line is provided, which comprises: Figure 2

[0081] The root mean square value calculation module 201 is configured to collect the current value and the electric field value of the power transmission line in real time, calculate a plurality of first root mean square values of the current value and a plurality of second root mean square values of the electric field value within a preset number of periods.

[0082] The similarity vector calculation module 202 is configured to, if the plurality of first root mean square values continuously decrease and the maximum decrease amplitude is greater than or equal to a preset first decrease amplitude, and the plurality of second root mean square values continuously decrease and the maximum decrease amplitude is greater than or equal to a preset second decrease amplitude, call the current calculation model to calculate a current similarity vector according to the plurality of first root mean square values, and call the electric field calculation model to calculate an electric field similarity vector according to the plurality of second root mean square values.

[0083] The fault judgment module 203 is configured to calculate a similarity confidence according to the matching degree between the current similarity vector and the electric field similarity vector, and determine that the power transmission line currently has a fault if the similarity confidence is greater than or equal to a preset similarity threshold.

[0084] ​In one specific embodiment, the fault monitoring device further includes: a confidence adjustment module, used to collect the acceleration value of the transmission line in real time, calculate multiple third root mean square values ​​of the current value within a preset number of cycles; call the acceleration calculation model to calculate the acceleration similarity vector based on the multiple third root mean square values; if the first distance between the acceleration similarity vector and the current similarity vector, and the second distance between the acceleration similarity vector and the electric field similarity vector are both less than a preset distance threshold, then a preset floating confidence level is added to the calculated similarity confidence level.

[0085] In one specific embodiment, the fault monitoring device further includes: a filtering module, used to segment the current value and electric field value to obtain multiple electrical segments; calculate the total harmonic distortion and root mean square value of each electrical segment; and among the multiple electrical segments, remove the electrical segments whose total harmonic distortion is greater than or equal to a preset first threshold value and / or whose root mean square value is greater than or equal to a preset second threshold value.

[0086] In one specific embodiment, the fault monitoring device further includes: a model training module, used to acquire sample electric field data and sample current data as a training set; wherein each data point is labeled with a sample label; extracting electric field rise rate and electric field fall rate as electric field features from the training set, and extracting current rise rate, current fall rate and surge current as current features; inputting the electric field features into a first classifier based on a decision tree, and inputting the current features into a second classifier based on a decision tree, to obtain the output predicted labels; adjusting the first classifier and the second classifier according to the difference between the sample labels and the predicted labels until the first classifier and the second classifier converge, using the first classifier as the electric field calculation model and the second classifier as the current calculation model.

[0087] Figure 3 An internal structural diagram of a terminal device in one embodiment is shown. Figure 3 As shown, the terminal device includes a processor, a memory, and a network interface connected via a system bus. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and may also store a computer program. When executed by the processor, this computer program enables the processor to implement a fault monitoring method for power transmission lines. The internal memory may also store a computer program, which, when executed by the processor, enables the processor to implement the fault monitoring method for power transmission lines. Those skilled in the art will understand that... Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the solution of this application and does not constitute a limitation on the terminal device to which the solution of this application is applied. A specific terminal device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0088] A computer readable storage medium stores a computer program, the computer program is executed by a processor to implement the following steps: collecting current values and electric field values of a power transmission line in real time, calculating a plurality of first root mean square values of the current values and a plurality of second root mean square values of the electric field values within a preset number of periods; if the plurality of first root mean square values continuously decrease and the maximum decrease amplitude is greater than or equal to a preset first decrease amplitude, and the plurality of second root mean square values continuously decrease and the maximum decrease amplitude is greater than or equal to a preset second decrease amplitude, then calling a current calculation model to calculate a current similarity vector according to the plurality of first root mean square values, and calling an electric field calculation model to calculate an electric field similarity vector according to the plurality of second root mean square values; calculating a similarity confidence according to the matching degree between the current similarity vector and the electric field similarity vector, and if the similarity confidence is greater than or equal to a preset similarity threshold, determining that the power transmission line currently has a fault.

[0089] A terminal device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the following steps when executing the computer program: collecting current values and electric field values of a power transmission line in real time, calculating a plurality of first root mean square values of the current values and a plurality of second root mean square values of the electric field values within a preset number of periods; if the plurality of first root mean square values continuously decrease and the maximum decrease amplitude is greater than or equal to a preset first decrease amplitude, and the plurality of second root mean square values continuously decrease and the maximum decrease amplitude is greater than or equal to a preset second decrease amplitude, then calling a current calculation model to calculate a current similarity vector according to the plurality of first root mean square values, and calling an electric field calculation model to calculate an electric field similarity vector according to the plurality of second root mean square values; calculating a similarity confidence according to the matching degree between the current similarity vector and the electric field similarity vector, and if the similarity confidence is greater than or equal to a preset similarity threshold, determining that the power transmission line currently has a fault.

[0090] It should be noted that the above power transmission line fault monitoring method, device, equipment and computer readable storage medium belong to one general inventive concept, and the contents in the power transmission line fault monitoring method, device, equipment and computer readable storage medium embodiments can be mutually applicable.

[0091] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer readable storage medium. When the program is executed, it can include the processes of the above-mentioned embodiment methods. Any reference to memory, storage, database or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synch link) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0092] Any combination of the technical features of the above embodiments can be made. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0093] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent of the present application. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A method for fault monitoring of transmission lines, characterized in that, The method includes: The current and electric field values ​​of the transmission line are collected in real time. Within a preset number of cycles, multiple first root mean square values ​​of the current value and multiple second root mean square values ​​of the electric field value are calculated. If the plurality of first root mean square values ​​continue to decrease and the maximum decrease is greater than or equal to a preset first decrease, and the plurality of second root mean square values ​​continue to decrease and the maximum decrease is greater than or equal to a preset second decrease, then the current calculation model is invoked to calculate the current similarity vector based on the plurality of first root mean square values, and the electric field calculation model is invoked to calculate the electric field similarity vector based on the plurality of second root mean square values. The similarity confidence level is calculated based on the degree of matching between the current similarity vector and the electric field similarity vector. If the similarity confidence level is greater than or equal to a preset similarity threshold, it is determined that there is a fault in the transmission line at present.

2. The method according to claim 1, characterized in that, The method further includes: The acceleration value of the transmission line is collected in real time, and multiple third root mean square values ​​of the current value are calculated within a preset number of cycles. The acceleration calculation model is invoked to calculate the acceleration similarity vector based on the multiple third root mean square values; If the first distance between the acceleration similarity vector and the current similarity vector, and the second distance between the acceleration similarity vector and the electric field similarity vector are both less than a preset distance threshold, then a preset floating confidence level is added to the calculated similarity confidence level.

3. The method according to claim 1, characterized in that, The method further includes: The current and electric field values ​​are segmented to obtain multiple electrical segments; Calculate the total harmonic distortion and root mean square value of each electrical segment, and among the plurality of electrical segments, remove the electrical segments whose total harmonic distortion is greater than or equal to a preset first threshold value and / or whose root mean square value is greater than or equal to a preset second threshold value.

4. The method according to claim 1, characterized in that, The method further includes: The sample electric field data and sample current data are obtained as the training set; each data point is labeled with a sample label. The electric field rise rate and electric field fall rate are extracted from the training set as electric field features, and the current rise rate, current fall rate and surge current are extracted as current features. The electric field features are input into a first classifier based on a decision tree, and the current features are input into a second classifier based on a decision tree to obtain the output predicted label. The first classifier and the second classifier are adjusted according to the difference between the sample label and the predicted label until the first classifier and the second classifier converge. The first classifier is used as the electric field calculation model and the second classifier is used as the current calculation model.

5. A fault monitoring device, characterized in that, The fault monitoring device includes: The root mean square (RMS) value calculation module is used to collect the current and electric field values ​​of the transmission line in real time, and calculate multiple first RMS values ​​of the current value and multiple second RMS values ​​of the electric field value within a preset number of cycles. The similarity vector calculation module is used to call the current calculation model to calculate the current similarity vector based on the multiple first root mean square values ​​and call the electric field calculation model to calculate the electric field similarity vector based on the multiple second root mean square values ​​if the multiple first root mean square values ​​continue to decrease and the maximum decrease is greater than or equal to the preset first decrease, and the multiple second root mean square values ​​continue to decrease and the maximum decrease is greater than or equal to the preset second decrease. The fault determination module is used to calculate the similarity confidence level based on the matching degree between the current similarity vector and the electric field similarity vector. If the similarity confidence level is greater than or equal to a preset similarity threshold, it is determined that there is a fault in the transmission line at present.

6. The fault monitoring device according to claim 5, characterized in that, The fault monitoring device also includes: The confidence adjustment module is used to collect the acceleration values ​​of the transmission line in real time, calculate multiple third root mean square values ​​of the current value within a preset number of cycles, call the acceleration calculation model to calculate the acceleration similarity vector based on the multiple third root mean square values, and if the first distance between the acceleration similarity vector and the current similarity vector, and the second distance between the acceleration similarity vector and the electric field similarity vector are both less than a preset distance threshold, then a preset floating confidence level is added to the calculated similarity confidence level.

7. The fault monitoring device according to claim 5, characterized in that, The fault monitoring device also includes: The filtering module is used to segment the current value and electric field value to obtain multiple electrical segments; calculate the total harmonic distortion and root mean square value of each electrical segment; and remove electrical segments whose total harmonic distortion is greater than or equal to a preset first threshold value and / or whose root mean square value is greater than or equal to a preset second threshold value from the multiple electrical segments.

8. The fault monitoring device according to claim 5, characterized in that, The fault monitoring device also includes: The model training module is used to acquire sample electric field data and sample current data as a training set; wherein each data point is labeled with a sample label; the electric field rise rate and electric field fall rate are extracted as electric field features, and the current rise rate, current fall rate, and surge current are extracted as current features from the training set; the electric field features are input into a first classifier based on a decision tree, and the current features are input into a second classifier based on a decision tree, to obtain the output predicted labels; the first classifier and the second classifier are adjusted according to the difference between the sample labels and the predicted labels until the first classifier and the second classifier converge; the first classifier is used as the electric field calculation model, and the second classifier is used as the current calculation model.

9. A computer-readable storage medium, characterized in that, The device stores a computer program that, when executed by a processor, causes the processor to perform the steps of the method as described in any one of claims 1 to 4.

10. A terminal device, characterized in that, It includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the method as described in any one of claims 1 to 4.

Images