Line equipment state detection method, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a state detection method of line equipment, electronic equipment and a storage medium. The method relates to the technical field of power grid detection and the technical field of artificial intelligence. The method comprises the following steps: acquiring operation data and an appearance image sequence of line equipment; determining a multi-modal feature based on the operation data and the appearance image sequence; state detection is carried out based on the multi-modal features, and candidate operation states of the line equipment are obtained; under the condition that the candidate operation state is a suspected old failure state, determining a target operation state based on historical periodic maintenance data and multi-modal features of the line equipment; the target operation state is an old failure state or a non-old failure state. The method is used for achieving the effect of detecting the old failure state of the line equipment in time.

Description

Technical Field

[0001] This application relates to the technical fields of power grid detection and artificial intelligence, and particularly to a method for detecting the status of line equipment, an electronic device, and a storage medium. Background Art

[0002] The power grid includes line equipment, such as transmission line equipment, substation equipment, and distribution line equipment; the aging and failure of line equipment will cause faults in the power grid and affect the normal operation of the power system.

[0003] In the related art, it is difficult to detect the aging and failure of line equipment. Often, after a fault is caused by the aging and failure of line equipment, it is only possible to troubleshoot the aging and failed line equipment for the fault. Therefore, there is an urgent need for a solution that can detect the aging and failure status of line equipment. Summary of the Invention

[0004] Embodiments of this application provide a method for detecting the status of line equipment, an electronic device, a storage medium, and a program product, so as to achieve the effect of timely detecting the aging and failure status of line equipment.

[0005] In a first aspect, an embodiment of this application provides a method for detecting the status of line equipment, including:

[0006] Obtain the operation data and appearance image sequence of the line equipment;

[0007] Based on the operation data and appearance image sequence, determine multi-modal features;

[0008] Perform status detection based on the multi-modal features to obtain the candidate operation status of the line equipment;

[0009] In the case where the candidate operation status is a suspected aging and failure status, determine the target operation status based on the historical regular maintenance data and multi-modal features of the line equipment; the target operation status is an aging and failure status or a non-aging and failure status.

[0010] In a possible implementation manner, based on the operation data and appearance image sequence, determining multi-modal features includes: extracting features from the operation data to obtain operation data features; extracting features from the appearance image sequence to obtain image features; and fusing the operation data features and the image features to obtain multi-modal features.

[0011] In a possible implementation manner, the operation data includes: voltage data, current data, active power data, and surface temperature data; feature extraction is performed on the operation data to obtain operation data features, including: extracting voltage statistical features of the voltage data, amplitude-phase features of the current data, change trend features of the active power data, and temperature change features of the surface temperature data; according to the voltage data and the current data, correlation features are determined; according to the voltage statistical features, amplitude-phase features, change trend features, temperature change features, and correlation features, operation data features are determined.

[0012] In a possible implementation manner, state detection is performed based on multi-modal features to obtain candidate operating states of the line equipment, including: inputting the multi-modal features into a first detection module to obtain a probability vector; based on the probability vector, a candidate operating state of the line equipment is selected from multiple preset states, and the candidate operating state is one of: running-in period state, robust period state, and suspected old and invalid state.

[0013] In a possible implementation manner, the target operating state is determined based on the historical regular maintenance data and multi-modal features of the line equipment, including: extracting data features of the historical regular maintenance data; fusing the data features and the multi-modal features to obtain target fusion features; performing state detection on the target fusion features through a second detection module to obtain the target operating state.

[0014] In a possible implementation manner, the state detection method of the line equipment further includes: when the target operating state is the old and invalid state, obtaining the location information of the line equipment; generating a warning prompt message based on the location information, the old and invalid state, and the appearance image sequence; sending the warning prompt message to the control end.

[0015] In a second aspect, an embodiment of the present application provides a state detection device for line equipment, including:

[0016] An acquisition module, configured to acquire operation data and an appearance image sequence of the line equipment;

[0017] A multi-modal feature determination module, configured to determine multi-modal features based on the operation data and the appearance image sequence;

[0018] A first detection module, configured to perform state detection based on the multi-modal features to obtain candidate operating states of the line equipment;

[0019] A second detection module, configured to, when the candidate operating state is the suspected old and invalid state, determine the target operating state based on the historical regular maintenance data and the multi-modal features of the line equipment; the target operating state is the old and invalid state or the non-old and invalid state.

[0020] In a third aspect, an embodiment of the present application provides an electronic device, including: a memory and a processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory, so that the processor executes the above first aspect and / or various possible implementation manners of the first aspect.

[0021] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, they are used to implement the above first aspect and / or various possible implementation manners of the first aspect.

[0022] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the above first aspect and / or various possible implementation manners of the first aspect.

[0023] The state detection method, electronic device, storage medium and program product of the line equipment provided by the embodiments of the present application determine multi-modal features according to the real-time operation data and appearance image sequence of the line equipment. Since the old and ineffective line equipment will lead to unstable operation states, and then the operation data will show abnormal fluctuations, effective state detection can be carried out based on the real-time operation data. The old and ineffective line equipment usually also causes changes in the appearance of the line equipment, such as surface cracking or deformation. The appearance image sequence is fused into the multi-modal features, so that the multi-modal features can reflect the operation conditions of the line equipment from multiple aspects, improving the reliability of state detection; multi-modal features are used for preliminary state detection. If the obtained candidate operation state is a suspected old and ineffective state, further state detection is carried out in combination with historical regular maintenance data to obtain the target operation state. When the line equipment is in an old and ineffective state, the historical regular maintenance data may also show abnormalities. Therefore, by combining the historical regular maintenance data, accurate state detection results can be obtained, avoiding misjudgment situations where the line equipment is determined to be in an old and ineffective state based on operation data and appearance image sequence due to the influence of the external environment, improving the accuracy of state detection; in addition, based on the operation data and appearance image sequence during the operation of the line equipment, the old and ineffective state can be detected when it causes minor changes, avoiding power grid failures caused by old and ineffective states, improving the timeliness of old and ineffective state detection, and ensuring the stable operation of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The drawings here are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0025] Figure 1 It is a schematic diagram of the scenario of the state detection method of the line equipment provided by the present application;

[0026] Figure 2 Schematic flow chart of the method for detecting the state of the line equipment provided by this application Figure 1 ;

[0027] Figure 3 Schematic diagram of the failure rates of three operating states of the line equipment provided by this application;

[0028] Figure 4 Schematic flow chart of the method for detecting the state of the line equipment provided by this application Figure 2 ;

[0029] Figure 5 Schematic structural diagram of the device for detecting the state of the line equipment provided by this application;

[0030] Figure 6 Schematic structural diagram of the electronic device provided by this application.

[0031] Through the above-mentioned drawings, specific embodiments of this application have been shown, and there will be more detailed descriptions hereinafter. These drawings and textual descriptions are not intended to limit the scope of the concept of this application in any way, but to illustrate the concept of this application to those skilled in the art by referring to specific embodiments. Specific Embodiments

[0032] Here, the exemplary embodiments will be described in detail, and the examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. On the contrary, they are merely examples of devices and methods consistent with some aspects of this application as detailed in the appended claims.

[0033] The method for detecting the state of the line equipment provided by the embodiments of this application can be applied to the application environment as shown in Figure 1 . Among them, the detection device 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or placed in the cloud or other network servers.

[0034] Among them, the detection device 102 is used to collect the operation data, appearance image sequence, etc. of the line equipment. The detection device 102 can be deployed near the line equipment. The server 104 is used to execute the method for detecting the state of the line equipment. The server 104 can be implemented by an independent server or a server cluster composed of multiple servers.

[0035] The technical solution of the present application and how the technical solution of the present application solves the above technical problems will be described in detail below with specific embodiments. The following several specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below with reference to the accompanying drawings.

[0036] Figure 2 Flow schematic of the method for detecting the state of the line equipment provided by the present application Figure 1 , the method for detecting the state of the line equipment can be applied to an electronic device, and the electronic device can be Figure 1 a server in; such as Figure 2 shown, the method for detecting the state of the line equipment includes:

[0037] S201. Obtain the operation data and appearance image sequence of the line equipment.

[0038] Among them, the operation data is real-time data obtained during the current prediction period, and the appearance image sequence is real-time images obtained during the current prediction period; during the operation of the line equipment, it alternately enters the prediction period and the rest period, and both the prediction period and the rest period have their respective preset durations, and the preset durations corresponding to the prediction period and the rest period can be set according to actual needs; whenever entering the prediction period, obtain the operation data and appearance image sequence of the line equipment during the corresponding time period of the current prediction period.

[0039] Optionally, the operation data may include but is not limited to voltage data, current data, active power data, and surface temperature data.

[0040] Since the current prediction period corresponds to a time period, the operation data is the operation data of the line equipment during this time period, and the appearance image sequence includes multiple appearance images of the line equipment during this time period sorted in chronological order.

[0041] Specifically, the operation data of the line equipment can be collected in real time by a detection device, and then the detection device can collect the operation data of the line equipment during the current prediction period; the appearance images of the line equipment can be periodically captured by a camera device deployed near the line equipment, and at the end of each prediction period, the appearance images captured during the current prediction period are arranged in chronological order to obtain an appearance image sequence, and the appearance image sequence is sent to the detection device, so that the detection device obtains the appearance image sequence; the detection device sends the operation data and appearance image sequence of the current prediction period to the electronic device, so that the electronic device obtains the operation data and appearance image sequence of the line equipment during the current prediction period.

[0042] S202. Determine multi-modal features based on the operation data and the appearance image sequence.

[0043] Among them, the multi-modal features are obtained by the interaction of at least two types of data features. In this embodiment, the at least two types of data features include the features of operation data and the features of appearance image sequences.

[0044] Optionally, the electronic device extracts the features of operation data, extracts the features of appearance image sequences, and fuses the features of operation data and the features of appearance image sequences to obtain multi-modal features; the multi-modal features can comprehensively reflect the operation status of line equipment.

[0045] Optionally, the electronic device performs shallow feature extraction on operation data to obtain first shallow features, performs shallow feature extraction on appearance image sequences to obtain second shallow features, fuses the first shallow features and the second shallow features to obtain shallow fusion features, the electronic device performs deep feature extraction on the shallow fusion features to obtain first deep features, performs deep feature extraction on the second shallow features to obtain second deep features, and fuses the first deep features and the second deep features to obtain multi-modal features; through the interaction of shallow and deep features, the quality of multi-modal features can be improved.

[0046] Optionally, the electronic device extracts the features of operation data, the electronic device determines whether there are differences between multiple appearance images in the appearance image sequence. If there are differences, it can extract the features of the appearance image sequence, fuse the features of operation data and the features of the appearance image sequence to obtain multi-modal features, and can also extract at least two appearance images with differences in the appearance image sequence, extract the features of the at least two appearance images with differences, and fuse the features of the at least two appearance images with differences and the features of operation data to obtain multi-modal features; if there are no differences between multiple appearance images in the appearance image sequence, then extract the features of any one appearance image in the appearance image sequence, and fuse the features of the appearance image and the features of operation data to obtain multi-modal features; by comparing differences, in some cases, the amount of data in the feature extraction process can be reduced, and the processing efficiency is improved.

[0047] S203. Perform state detection based on the multi-modal features to obtain candidate operation states of the line equipment.

[0048] Among them, the candidate operation state is one of the running-in state, the robust state, and the suspected old and invalid state.

[0049] The running-in state usually appears in the initial stage when the line equipment is put into use. Due to minor defects in internal components and the running-in situation, the failure rate is relatively high. However, the equipment indicators are normal, but the running state is not stable enough, and the equipment performance is average.

[0050] The robust period state usually appears in the middle stage after the running-in period. The line equipment operates smoothly, the failure rate reaches the minimum value in the whole life cycle of the line equipment, the equipment indicators are the best, the operating state is stable, and the production performance is the strongest. This state occupies the longest time in the whole life cycle of the line equipment.

[0051] The suspected old and invalid state can be understood as the decline period state, which is usually caused by the wear and aging of the line equipment, and may lead to a gradual increase in the failure rate, abnormal fluctuations in equipment indicators, and a gradual decline in production performance.

[0052] Figure 3 The figure shows the change of the failure rate in the running-in period state, the robust period state, and the suspected old and invalid state. It can be seen that the failure rates in the running-in period state and the suspected old and invalid state are relatively high, while the failure rate in the robust period state is relatively low.

[0053] It can be seen that the running-in period state, the robust period state, and the suspected old and invalid state are all stages in the whole life cycle of the line equipment.

[0054] In one possible implementation, S203 includes: inputting the multi-modal features into the first prediction module to obtain a probability vector; based on the probability vector, selecting the candidate operating state of the line equipment from multiple preset states, and the candidate operating state is one of the running-in period state, the robust period state, and the suspected old and invalid state.

[0055] Among them, the first prediction module can handle classification tasks. In practical applications, the first prediction module can be implemented by a classifier.

[0056] Specifically, the electronic device inputs the multi-modal features into the first detection module, and the first detection module outputs a probability vector. The probability vector contains the probability values corresponding to the three preset states respectively, and the preset state corresponding to the largest probability value is used as the candidate operating state.

[0057] Optionally, when the preset state corresponding to the largest probability value in the predicted probability vector is the suspected old and invalid state, it is determined that the candidate operating state is the suspected old and invalid state.

[0058] Optionally, when the preset state corresponding to the largest probability value in the predicted probability vector is the suspected old and invalid state, and the largest probability value is greater than the preset threshold, it is determined that the candidate operating state is the suspected old and invalid state.

[0059] In the above embodiment, by classifying the multi-modal features through the first prediction module, the candidate operating state can be quickly obtained, which improves the efficiency and timeliness of state detection.

[0060] S204. When the candidate operating state is suspected to be in an old and invalid state, determine the target operating state based on the historical regular maintenance data and multi-modal features of the line equipment; the target operating state is an old and invalid state or a non-old and invalid state.

[0061] Among them, the historical regular maintenance data is the test data obtained by regularly maintaining the line equipment; the historical regular maintenance data can be the test data at the most recent maintenance inspection. In practical applications, the line equipment is usually shut down and maintained regularly to conduct functional tests on the line equipment to ensure that the line equipment can continue to operate.

[0062] It should be noted that when the line equipment is in an old and invalid state, the historical regular maintenance data when it is about to enter the old and invalid state is different from the historical regular maintenance data when the line equipment is normal. Therefore, the state can be detected through the historical regular maintenance data; in addition, the historical regular maintenance data is the data obtained by conducting specific functional tests, which is different from the operation data. Combining the two for state detection can improve the reliability of state detection.

[0063] In a possible implementation manner, determining the target operating state based on the historical regular maintenance data and multi-modal features of the line equipment includes: extracting the data features of the historical regular maintenance data; fusing the data features and multi-modal features to obtain the target fusion features; and performing state detection on the target fusion features through the second detection module to obtain the target operating state.

[0064] Specifically, when the candidate operating state is suspected to be in an old and invalid state, the electronic device obtains the historical regular maintenance data obtained from the most recent maintenance inspection of the line equipment, extracts the data features of the historical regular maintenance data through the feature extraction module, and fuses the data features and multi-modal features to obtain the target fusion features.

[0065] The historical regular maintenance data includes the insulation resistance value, leakage current value, and partial discharge signal. During historical regular maintenance, an insulation resistance test is performed on the line equipment to obtain the insulation resistance value, and the insulation material of the line equipment can be judged whether it is aged or damaged through the insulation resistance value; a leakage current test is performed on the line equipment to obtain the leakage current value, and an excessive leakage current may indicate that the insulation material of the line equipment has aged or been damaged; a partial discharge detection is performed on the line equipment to obtain the partial discharge signal, and the partial discharge signal can reflect whether there are defects in the line equipment.

[0066] The historical regular maintenance data is composed of the insulation resistance value, leakage current value, and partial discharge signal, and the historical regular maintenance data is input into the feature extraction module to obtain the data features.

[0067] Optionally, fusing the data features and the multimodal features may be concatenating the data features and the multimodal features to obtain the target fused features.

[0068] Optionally, fusing the data features and the multimodal features may also be performing weighted summation on the data features and the multimodal features according to preset weight information to obtain the target fused features.

[0069] The electronic device inputs the target fused features into the second detection module to obtain a first target probability value, and determines the target operating state according to the first target probability value; if the first target probability value is greater than the first target threshold, it is determined that the target operating state is the old and invalid state, and if the first target probability value is not greater than the first target threshold, it is determined that the target operating state is the non-old and invalid state; the first target threshold can be set according to actual requirements, and the embodiments of the present application do not limit this.

[0070] It should be noted that in the initial stage when the line device is in the old and invalid state, there may be minor differences between the historical regular maintenance data and the maintenance data in the normal operating state, resulting in the inability to determine the old and invalid state of the line device only through the historical regular maintenance data. In the case where it is initially detected through the real-time operating data and the appearance image sequence that the line device may be in the old and invalid state, the multimodal features are fused into the data features of the historical regular maintenance data, so that the differences of different types (the differences corresponding to the operating data, the appearance image sequence, and the historical regular maintenance data respectively) are superimposed, and the target operating state of the line device can be further determined, making the state detection more reliable and improving the accuracy of the state detection.

[0071] Optionally, in the case where the candidate operating state is the suspected old and invalid state, obtain the historical regular maintenance data obtained from the most recent maintenance inspection of the line device, compare the reference regular maintenance data with the historical regular maintenance data to obtain the difference data, extract the features of the difference data, fuse the features of the difference data and the multimodal features to obtain the fused features, and perform state detection on the fused features to obtain the target operating state of the line device; specifically, the fused features can be input into the third detection module, and the third detection module outputs a second target probability value, and the target operating state is determined according to the second target probability value; if the second target probability value is greater than the second target threshold, it is determined that the target operating state is the old and invalid state, and if the second target probability value is not greater than the second target threshold, it is determined that the target operating state is the non-old and invalid state; the second target threshold can be set according to actual requirements, and the embodiments of the present application do not limit this.

[0072] The state detection method for line equipment provided by the embodiments of the present application determines multi-modal features based on the real-time operation data and appearance image sequence of the line equipment. Since the old and ineffective line equipment will cause unstable operation states, and thus abnormal fluctuations in the operation data, effective state detection can be performed based on the real-time operation data. The old and ineffective line equipment usually also causes changes in the appearance of the line equipment, such as surface cracking or deformation. The appearance image sequence is fused into the multi-modal features, so that the multi-modal features can reflect the operation conditions of the line equipment from multiple aspects, improving the reliability of state detection. The multi-modal features are used for preliminary state detection. If the obtained candidate operation state is a suspected old and ineffective state, the historical regular maintenance data is further combined for state detection to obtain the target operation state. When the line equipment is in the old and ineffective state, the historical regular maintenance data may also show abnormalities. Therefore, by combining the historical regular maintenance data, accurate state detection results can be obtained, avoiding misjudgment situations where the line equipment is determined to be in the old and ineffective state based on the operation data and appearance image sequence due to the influence of the external environment, and improving the accuracy of state detection. In addition, based on the operation data and appearance image sequence during the operation of the line equipment, the old and ineffective state can be detected when it causes minor changes, avoiding power grid failures caused by old and ineffective states, improving the timeliness of old and ineffective state detection, and ensuring the stable operation of the power grid.

[0073] In some embodiments, based on the operation data and appearance image sequence, determining multi-modal features includes: extracting features from the operation data to obtain operation data features; extracting features from the appearance image sequence to obtain image features; and fusing the operation data features and the image features to obtain multi-modal features.

[0074] Optionally, a feature extraction module can be used to extract features from the operation data to obtain operation data features; the feature extraction module is used to extract the feature maps of each appearance image respectively, and the feature maps of multiple appearance images are spliced according to the number of channels to obtain image features; the image features and the operation data features can be weighted and summed to obtain multi-modal features. The weights of the image features and the operation data features can be set according to actual needs, and the embodiments of the present application do not limit this; alternatively, the image features and the operation data features can be spliced to obtain multi-modal features.

[0075] Fusing the operation data features and the image features to obtain multi-modal features enables the multi-modal features to comprehensively reflect the operation conditions of the line equipment, improving the quality of the multi-modal features. Subsequently, state detection is performed through the multi-modal features, improving the accuracy of state detection.

[0076] Optionally, the operation data includes multiple types of data. For each type of data, feature extraction can be performed according to the characteristics of the data to obtain the features of the data, and the operation data features are composed of the features of multiple types of data.

[0077] In a possible implementation, the operation data includes: voltage data, current data, active power data, and surface temperature data; feature extraction is performed on the operation data to obtain operation data features, including:

[0078] Extract the voltage statistical features of the voltage data, the amplitude-phase features of the current data, the change trend features of the active power data, and the temperature change features of the surface temperature data; determine the correlation features according to the voltage data and the current data; determine the operation data features according to the voltage statistical features, the amplitude-phase features, the change trend features, the temperature change features, and the correlation features.

[0079] Among them, the voltage data, the current data, the active power data, and the surface temperature data are all data corresponding to the current detection period.

[0080] Specifically, the electronic device calculates the average value of the voltage data, and the average value can reflect the overall level of the voltage; calculates the standard deviation of the voltage data, and the standard deviation can measure the fluctuation degree of the voltage data, obtains the extreme value of the voltage data, which can reflect the change range of the voltage; constitutes the voltage statistical features according to the average value, the standard deviation, and the extreme value.

[0081] The electronic device performs Fourier transform on the current data to obtain spectral information, and the spectral information can reflect the components and energy distribution of the current at different frequencies; extracts the amplitude and phase at specific frequencies according to the spectral information, and constitutes the amplitude-phase features according to the amplitude and phase; the amplitude-phase features can be used to detect whether the current components are abnormal.

[0082] The electronic device uses methods such as linear regression or time series analysis to fit the active power data to obtain fitting information, and the fitting information can reflect the increase and decrease of the active power. Determine the slope and intercept according to the change trend information, and constitute the change trend features according to the slope and intercept; the change trend features can reflect whether the active power is within the normal range.

[0083] The electronic device analyzes the surface temperature data, extracts the average value, the standard deviation, the maximum value, and the minimum value of the surface temperature data, and determines the change rate of the surface temperature data, and determines the temperature change features according to the average value, the standard deviation, the maximum value, the minimum value, and the change rate of the surface temperature data.

[0084] The electronic device takes the correlation coefficient between the current data and the voltage data as the correlation feature.

[0085] The electronic device forms a candidate vector based on voltage statistical characteristics, amplitude-phase characteristics, change trend characteristics, temperature change characteristics, and correlation characteristics, and uses a feature extraction module to extract features from the candidate vector to obtain operation data characteristics.

[0086] In the above embodiment, the operation data characteristics are determined according to the voltage statistical characteristics of voltage data, the amplitude-phase characteristics of current data, the change trend characteristics of active power data, the temperature change characteristics, and the temperature change characteristics between voltage data and current data, which reflects the characteristics of operation data from multiple aspects and improves the quality of operation data characteristics.

[0087] In a possible implementation manner, the method for detecting the state of a line device further includes: when the target operation state is an old and failed state, obtaining the location information of the line device; generating a warning prompt message based on the location information, operation data, and appearance image sequence; and sending the warning prompt message to the control end.

[0088] Specifically, the electronic device pre-stores the location information of each line device. When it is determined that the target operation state of a certain line device is an old and failed state, it searches for the location information of the line device, generates a warning prompt message according to the location information, operation data, and appearance image sequence, and sends the warning prompt message to the control end, so that the management personnel can view the warning prompt message through the control end.

[0089] Among them, the warning prompt message is used to prompt the management personnel that the line device is in an old and failed state, so that the management personnel can preliminarily determine the operation condition of the line device according to the operation data and appearance image sequence in the warning prompt message, and quickly determine the location of the line device according to the location information in the warning prompt message, so as to facilitate the processing of the line device.

[0090] In the above embodiment, when the line device is in an old and failed state, a warning prompt message is sent to the control end, so that the management personnel can process the line device in time, which improves the efficiency of controlling the line device in an old and failed state.

[0091] Exemplarily, such as Figure 4As shown, the status of the line equipment is detected by the status detection model. Specifically, the status detection model includes a feature extraction module, a first detection module, and a second detection module. After obtaining the operation data and appearance image sequence of the line equipment in the current prediction period, the feature extraction module extracts features from the operation data and appearance image sequence to obtain operation data features and image feature sequences, fuses the operation data features and image feature sequences to obtain multi-modal features, and inputs the multi-modal features into the first detection module to obtain the candidate operation status of the line equipment. When the candidate operation status is a suspected old and invalid status, the historical regular maintenance data of the line equipment is input into the feature extraction module to obtain data features, the data features and multi-modal features are fused to obtain target fusion features, and the target fusion features are input into the second detection module to obtain the target operation status.

[0092] In practical applications, the status detection model is obtained by training the initial status detection model with operation data samples, appearance image sequence samples, historical regular maintenance data samples, and status labels of the line equipment. Specifically, the initial status detection model extracts and fuses features from the operation data samples and appearance image sequence samples in the same period to obtain training multi-modal features, and performs status prediction on the training multi-modal features to obtain the initial prediction status. When the initial prediction status is a suspected candidate status, the initial status detection model performs status prediction on the historical regular maintenance data samples and training multi-modal features to obtain the target prediction status, calculates the loss value based on the target prediction status and the status label. For example, the cross-entropy loss function can be used to calculate the loss value, and the loss value is used to adjust the parameters of the initial status detection model; the initial status detection model is iteratively trained according to the above process until the initial status detection model converges, and the converged initial status detection model is used as the status detection model.

[0093] The state detection method for line equipment provided by the embodiments of the present application determines multimodal features based on the real-time operation data and appearance image sequence of the line equipment. Since the old and ineffective line equipment will lead to unstable operation states, and further abnormal fluctuations in the operation data, effective state detection can be carried out based on the real-time operation data. The old and ineffective line equipment usually also causes changes in the appearance of the line equipment, such as surface cracking or deformation. The appearance image sequence is fused into the multimodal features, so that the multimodal features can reflect the operation conditions of the line equipment from multiple aspects, improving the reliability of state detection. The multimodal features are used for preliminary state detection. If the obtained candidate operation state is a suspected old and ineffective state, the historical regular maintenance data is further combined for state detection to obtain the target operation state. When the line equipment is in the old and ineffective state, the historical regular maintenance data may also show abnormalities. Therefore, by combining the historical regular maintenance data, accurate state detection results can be obtained, avoiding misjudgment situations where it is determined that the line equipment is in the old and ineffective state based on the operation data and appearance image sequence due to the influence of the external environment, and improving the accuracy of state detection. In addition, based on the operation data and appearance image sequence during the operation of the line equipment, the old and ineffective state can be detected when it causes minor changes, avoiding power grid failures caused by old and ineffective conditions, improving the timeliness of old and ineffective state detection, and ensuring the stable operation of the power grid.

[0094] It should be understood that although the steps in the flowcharts involved in the above embodiments are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear indication in this article, the execution of these steps does not have a strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same moment, but can be executed at different moments. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0095] Figure 5 The following is a schematic structural diagram of the state detection device for line equipment provided by the present application, as Figure 5 shown, the state detection device 50 provided by this embodiment includes:

[0096] An acquisition module 501, configured to acquire the operation data and appearance image sequence of the line equipment;

[0097] A multimodal feature determination module 502, configured to determine multimodal features based on the operation data and appearance image sequence;

[0098] The first detection module 503 is configured to perform status detection based on multimodal features to obtain candidate operating states of the line equipment;

[0099] The second detection module 504 is configured to, when the candidate operating state is a suspected old and invalid state, determine a target operating state based on the historical regular maintenance data and multimodal features of the line equipment; the target operating state is an old and invalid state or a non-old and invalid state.

[0100] In a possible implementation manner, the multimodal feature determination module 502 is further configured to extract features from the operating data to obtain operating data features; extract features from the appearance image sequence to obtain image features; and fuse the operating data features and the image features to obtain multimodal features.

[0101] In a possible implementation manner, the operating data includes: voltage data, current data, active power data, and surface temperature data; the multimodal feature determination module 502 is further configured to extract voltage statistical features of the voltage data, amplitude-phase features of the current data, change trend features of the active power data, and temperature change features of the surface temperature data; determine correlation features according to the voltage data and the current data; and determine operating data features according to the voltage statistical features, amplitude-phase features, change trend features, temperature change features, and correlation features.

[0102] In a possible implementation manner, the first detection module 503 is further configured to input the multimodal features into the first detection module 503 to obtain a probability vector; and based on the probability vector, select a candidate operating state of the line equipment from multiple preset states, and the candidate operating state is one of a running-in period state, a robust period state, and a suspected old and invalid state.

[0103] In a possible implementation manner, the second detection module 504 is further configured to extract data features of the historical regular maintenance data; fuse the data features and the multimodal features to obtain target fusion features; and perform status detection on the target fusion features through the second detection module 504 to obtain a target operating state.

[0104] In a possible implementation manner, the status detection device of the line equipment further includes: a warning prompt module, configured to, when the target operating state is an old and invalid state, obtain the location information of the line equipment; generate a warning prompt message based on the location information, the old and invalid state, and the appearance image sequence; and send the warning prompt message to the control end.

[0105] The status detection device of the line equipment provided in this embodiment can execute the status detection method of the line equipment provided in the above method embodiment, and its implementation principle and technical effects are similar, which will not be elaborated here in this embodiment.

[0106] Figure 6The structural schematic diagram of the electronic device provided by this application is as follows. As Figure 6 shown, the electronic device 60 provided in this embodiment includes: at least one processor 601 and a memory 602. Optionally, the device 60 further includes a communication component 603. Among them, the processor 601, the memory 602, and the communication component 603 are connected through a bus.

[0107] In the specific implementation process, at least one processor 601 executes the computer-executable instructions stored in the memory 602, so that at least one processor 601 executes the above-mentioned method.

[0108] For the specific implementation process of the processor 601, reference can be made to the above method embodiment, and its implementation principle and technical effect are similar, which will not be elaborated here in this embodiment.

[0109] In the above embodiment, it should be understood that the processor may be a central processing unit (English: Central Processing Unit, abbreviated as: CPU), and may also be other general-purpose processors, digital signal processors (English: Digital Signal Processor, abbreviated as: DSP), application-specific integrated circuits (English: Application Specific Integrated Circuit, abbreviated as: ASIC), etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in combination with the invention can be directly embodied as being executed by a hardware processor, or executed by a combination of hardware and software modules in the processor.

[0110] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (Non-volatile Memory, NVM), such as at least one disk memory.

[0111] The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, the bus in the drawings of this application is not limited to only one bus or one type of bus.

[0112] This application also provides a computer program product, including a computer program, which implements the above-mentioned method when executed by a processor.

[0113] The present application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above method.

[0114] The above-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium accessible by a general-purpose or special-purpose computer.

[0115] An exemplary readable storage medium is coupled to the processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in a device.

[0116] The division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling, direct coupling or communication connection between each other can be through some interfaces, and the indirect coupling or communication connection of devices or units can be in electrical, mechanical or other forms.

[0117] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0118] In addition, the functional units in each embodiment of the present invention can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit.

[0119] If a function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or a part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The foregoing storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs and other various media that can store program codes.

[0120] Those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps including the above method embodiments; and the foregoing storage medium includes: ROMs, RAMs, magnetic disks, or optical discs and other various media that can store program codes.

[0121] Finally, it should be noted that: After considering the specification and practicing the invention disclosed herein, those skilled in the art will readily think of other embodiments of the present invention. The present invention is intended to cover any variations, uses, or adaptations of the present invention, which follow the general principles of the present invention and include the common general knowledge or conventional technical means in the technical field not disclosed in the present invention. It is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is only limited by the appended claims.

Claims

1. A method for detecting the state of a line device, characterized in that: include: Obtaining operation data and appearance image sequences of line equipment; determining a multimodal feature based on the operational data and the appearance image sequence; Performing state detection based on the multimodal features to obtain a candidate operating state of the line equipment; When the candidate operating state is a suspected obsolete failure state, a target operating state is determined based on the historical periodic maintenance data of the line equipment and the multimodal feature; the target operating state is an obsolete failure state or a non-obsolete failure state.

2. The method according to claim 1, characterized in that The determining of multimodal features based on the operation data and the appearance image sequence includes: Extracting features from the operation data to obtain operation data features; Extracting features from the appearance image sequence to obtain image features; The operation data features and the image features are fused to obtain multimodal features.

3. The method according to claim 2, characterized in that The operation data includes: voltage data, current data, active power data and surface temperature data; The extracting features of the operation data to obtain operation data features includes: Extracting voltage statistical characteristics of the voltage data, amplitude and phase characteristics of the current data, change trend characteristics of the active power data, and temperature change characteristics of the surface temperature data; Determining a correlation feature according to the voltage data and the current data; The operation data characteristics are determined according to the voltage statistical characteristics, the amplitude and phase characteristics, the change trend characteristics, the temperature change characteristics and the correlation characteristics.

4. The method according to claim 1, characterized in that: The performing state detection based on the multimodal feature to obtain the candidate operating state of the line device includes: Inputting the multimodal features into a first detection module to obtain a probability vector; Based on the probability vector, a candidate operating state of the line equipment is selected from a plurality of preset states, and the candidate operating state is one of: a running-in state, a robust state, and a suspected old and failed state.

5. The method according to any one of claims 1 to 4, characterized in that The determining the target operating state based on the historical periodic maintenance data of the line equipment and the multimodal feature includes: Extract data features of historical periodic maintenance data; Fusing the data feature with the multimodal feature to obtain a target fusion feature; The target fusion feature is subjected to state detection by a second detection module to obtain the target operation state.

6. The method according to any one of claims 1 to 4, characterized in that The method further comprises: When the target operating state is an old and invalid state, obtaining the location information of the line equipment; Generate a warning message based on the location information, the old failure state and the appearance image sequence; Send the early warning message to the control terminal.

7. A state detection device for line equipment, characterized in that: include: An acquisition module, used to acquire operation data and appearance image sequences of line equipment; a multimodal feature determination module, configured to determine a multimodal feature based on the operation data and the appearance image sequence; A first detection module, configured to perform state detection based on the multimodal feature to obtain a candidate operating state of the line device; The second detection module is used to determine the target operating state based on the historical periodic maintenance data of the line equipment and the multimodal characteristics when the candidate operating state is a suspected old and failed state; the target operating state is an old and failed state or a non-old and failed state.

8. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 6 when executed by a processor.

10. A computer program product, characterized in that The method comprises computer-executable instructions, which implement the method according to any one of claims 1 to 6 when executed by a processor.

Images