Power transmission trip diagnosis method and device, electronic equipment and storage medium

Abstract

The invention discloses a power transmission tripping diagnosis method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring first data, second data and third data of a fault line corresponding to a power transmission line tripping event; analyzing the first data and the second data based on a first detection model to obtain a first detection result, and determining a diagnosis deterministic index of the first detection result, which is used for reflecting the reliability degree of a fault type in the first detection result; and further comparing the diagnosis certainty index with a first threshold value and a second threshold value, thereby determining whether to introduce a second detection model or a third detection model for data analysis, determining a target detection result, and facilitating generation of an emergency disposal plan of the power transmission line tripping event based on the target detection result. According to the invention, the accuracy of the identification result is improved, and the intelligent level, the overall efficiency and the long-term adaptability of power grid fault emergency disposal are improved.

Description

Technical Field

[0001] This invention relates to the field of data acquisition and processing technology, and in particular to a method, device, electronic equipment, and storage medium for diagnosing power transmission trips. Background Technology

[0002] Transmission line tripping is a major type of fault affecting the safe and stable operation of the power grid. Rapidly and accurately diagnosing the causes of tripping and generating emergency response plans is crucial for shortening outage time and improving power supply reliability. Existing transmission line tripping diagnostic technologies primarily rely on electrical quantity information provided by the power grid data acquisition and monitoring control system, combined with preliminary analysis of protection action signals. Simultaneously, meteorological monitoring data or single image recognition technology are introduced to automatically identify common fault types by constructing simple fault classification models or rule-based expert systems. However, these methods can only identify simple and common faults, failing to recognize complex faults and compromising the accuracy of the identification results. Summary of the Invention

[0003] This invention provides a method, device, electronic equipment, and storage medium for diagnosing power transmission trips, in order to improve the accuracy of identification results and enhance the intelligence level, overall efficiency, and long-term adaptability of emergency response to power grid faults.

[0004] According to one aspect of the present invention, a method for diagnosing power transmission tripping is provided, the method comprising:

[0005] Acquire first data, second data, and third data of the faulty line corresponding to the power transmission line tripping event; the first data includes electrical quantity time-series data of the faulty line; the second data is meteorological monitoring data; and the third data is video stream data captured on the faulty line.

[0006] The first detection model is used to analyze the first data and the second data to obtain a first detection result, and a diagnostic certainty index for the first detection result is determined; the diagnostic certainty index is used to reflect the reliability of the first detection result.

[0007] If the diagnostic certainty index is less than or equal to the first threshold, then the fourth data is obtained, and the first data, the second data, the third data and the fourth data are analyzed based on the second detection model to obtain the target detection result; the fourth data is the geographical environment data corresponding to the faulty line.

[0008] If the diagnostic certainty index is greater than the first threshold and less than the second threshold, then the fifth data is determined, and the fifth data is analyzed based on the third detection model to obtain the target detection result; the fifth data includes the third data and the first reference time-series feature data; the reference time-series feature data is the multi-time-scale time-series feature extracted by the first detection model corresponding to the first detection result; the first threshold is less than the second threshold;

[0009] If the diagnostic certainty index is greater than or equal to the second threshold, then the first detection result is taken as the target detection result;

[0010] Based on the target detection results, an emergency response plan for the power transmission line tripping event is generated.

[0011] According to another aspect of the present invention, a power transmission trip diagnostic device is provided, the device comprising:

[0012] The data acquisition module is used to acquire first data, second data, and third data of the faulty line corresponding to the power transmission line tripping event; the first data includes electrical quantity time-series data of the faulty line; the second data is meteorological monitoring data; and the third data is video stream data captured on the faulty line.

[0013] The data processing module is used to analyze the first data and the second data based on the first detection model to obtain a first detection result and determine the diagnostic certainty index of the first detection result; the diagnostic certainty index is used to reflect the reliability of the first detection result.

[0014] The first judgment module is used to obtain fourth data if the diagnostic certainty index is less than or equal to the first threshold, and to analyze the first data, the second data, the third data and the fourth data based on the second detection model to obtain the target detection result; the fourth data is the geographical environment data corresponding to the faulty line.

[0015] The second judgment module is used to determine the fifth data if the diagnostic certainty index is greater than the first threshold and the diagnostic certainty index is less than the second threshold, and to analyze the fifth data based on the third detection model to obtain the target detection result; the fifth data includes the third data and the first reference time series feature data; the reference time series feature data is the multi-time scale time series feature extracted by the first detection model corresponding to the first detection result; the first threshold is less than the second threshold;

[0016] The third judgment module is used to take the first detection result as the target detection result if the diagnostic certainty index is greater than or equal to the second threshold.

[0017] The scheme determination module is used to generate an emergency response plan for the power transmission line tripping event based on the target detection results.

[0018] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0019] At least one processor; and

[0020] A memory communicatively connected to the at least one processor; wherein,

[0021] The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the power transmission trip diagnosis method according to any embodiment of the present invention.

[0022] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions, the computer instructions being configured to cause a processor to execute and implement the power transmission trip diagnosis method according to any embodiment of the present invention.

[0023] The technical solution of this invention involves acquiring first data, second data, and third data of the faulty line corresponding to a transmission line tripping event. The first data includes electrical quantity time-series data of the faulty line; the second data is meteorological monitoring data; and the third data is video stream data captured on the faulty line. First, the first and second data are analyzed based on a first detection model to obtain a first detection result, and a diagnostic deterministic index is determined to reflect the reliability of the fault type in the first detection result. Further, the diagnostic deterministic index is compared with a first threshold and a second threshold to determine whether to introduce a second or third detection model for data analysis, thereby determining the target detection result. When the diagnostic deterministic index is greater than or equal to the second threshold, the first detection result is used as the target detection result. This significantly shortens the diagnosis time. When the diagnostic certainty index is less than or equal to the first threshold, the geographical environment data corresponding to the faulty line is used as the fourth data. The first, second, third, and fourth data are analyzed based on the second detection model to obtain the target detection result, enabling the automatic activation of deeper multimodal fusion analysis and ensuring the reliability of the final diagnosis conclusion. When the diagnostic certainty index is greater than the first threshold and less than the second threshold, the fifth data, including the third data and the first reference time-series feature data, is determined. The fifth data is analyzed based on the third detection model to obtain the target detection result. The reference time-series feature data is the multi-time-scale time-series feature extracted by the first detection model corresponding to the first detection result. The introduction of visual data significantly improves fault determination. In other words, for simple faults with high determinism, this invention directly adopts preliminary results, greatly shortening the diagnosis time. When the uncertainty of the preliminary diagnosis is high, the system can sequentially initiate medium-complexity multimodal verification or high-complexity full-modal fusion analysis. By introducing multi-source information such as vision and geography at each level and performing deep feature interaction, the system significantly improves the success rate and robustness of diagnosis for complex scenarios such as compound faults and rare faults. Finally, based on the target detection results, an emergency response plan for transmission line tripping events is generated, improving the intelligence level, overall efficiency, and long-term adaptability of power grid fault emergency response.

[0024] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is a flowchart of a power transmission trip diagnosis method provided by an embodiment of the present invention;

[0027] Figure 2 This is a flowchart of another power transmission trip diagnosis method provided by an embodiment of the present invention;

[0028] Figure 3 This is a flowchart of another power transmission trip diagnosis method provided according to an embodiment of the present invention;

[0029] Figure 4 This is a flowchart of another power transmission trip diagnosis method provided by an embodiment of the present invention;

[0030] Figure 5 This is a schematic diagram of the structure of a power transmission trip diagnostic device according to an embodiment of the present invention;

[0031] Figure 6 This is a schematic diagram of the structure of an electronic device that implements the power transmission trip diagnosis method of the present invention, according to an embodiment of the present invention. Detailed Implementation

[0032] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0033] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0034] Example 1

[0035] Figure 1This is a flowchart illustrating a power transmission line tripping diagnosis method provided in an embodiment of the present invention. This embodiment is applicable to diagnosing power transmission line tripping events. The method can be executed by a power transmission line tripping diagnosis device, which can be implemented in hardware and / or software. This device can be configured in any electronic device with network communication capabilities. Figure 1 As shown, the power transmission trip diagnosis method of the present invention may include:

[0036] S110. Obtain the first data, second data and third data of the faulty line corresponding to the power transmission line tripping event; the first data includes the electrical quantity time sequence data of the faulty line; the second data is meteorological monitoring data; the third data is video stream data of the faulty line.

[0037] Specifically, in response to a transmission line tripping event signal issued by the power grid dispatch center, the system automatically triggers a data acquisition process, simultaneously collecting multi-source heterogeneous data within preset time windows before and after the tripping event. The total time window, consisting of twice the preset time window, is an optimal solution determined after extensive historical data analysis. It effectively captures the abnormal evolution process before the tripping event and completely records the changes in the field status at the moment of tripping and afterwards. For example, if the preset time window is 30 minutes, then the total time window can be 60 minutes.

[0038] The first data may include electrical quantity timing data such as voltage, current, active power, reactive power, zero-sequence current, and protection action type of the line before and after the trip.

[0039] The second data can be time-series data recorded by a miniature meteorological monitoring device at the location of the faulty line tower, obtained from a power meteorological monitoring platform or a national meteorological station; meteorological information such as wind speed, wind direction, temperature, humidity, precipitation intensity, and lightning location collected at 1-minute intervals; the second data is used to analyze whether there are extreme meteorological conditions that could cause the power outage, such as instantaneous strong winds, lightning strikes, hail, etc.

[0040] The third type of data can be remote sensing imagery and video data of the power line corridor. Specifically, this can be obtained by retrieving optical satellite images with resolutions ranging from sub-meter to 2 meters taken by on-orbit remote sensing satellites before and after the tripping time. Simultaneously, high-definition video monitoring devices and inspection drones deployed along the faulty power line corridor are activated to transmit real-time video stream data. The remote sensing imagery provides static snapshots of the power line corridor before and after the tripping time, used to detect spatial targets such as hanging objects, illegal structures, and large machinery. The video data provides temporal dynamic information, used to analyze the movement trajectory of construction machinery, the relative swaying of vegetation and conductors, and dynamic phenomena such as abnormal smoke or flames.

[0041] S120. Analyze the first data and the second data based on the first detection model to obtain the first detection result, and determine the diagnostic certainty index of the first detection result; the diagnostic certainty index is used to reflect the reliability of the first detection result.

[0042] The first detection model can be understood as a time-series analysis model with the highest computational efficiency, used to quickly and initially diagnose transmission line tripping events, generate a directional fault analysis result, and provide a decision-making basis for subsequent in-depth analysis that may be triggered.

[0043] The first detection result can be understood as a probability distribution vector. Each dimension of the probability distribution vector corresponds to a predefined fault type. The elements of each dimension of the probability distribution vector represent the confidence level of the first detection model in the corresponding fault type obtained by analyzing the first and second data. The confidence level can clarify the reliability of the fault type.

[0044] Specifically, the highest confidence level corresponding to the fault type in the first test result can be determined as the diagnostic certainty index of the first test result.

[0045] S130. If the diagnostic certainty index is less than or equal to the first threshold, then the fourth data is obtained. The first data, the second data, the third data and the fourth data are analyzed based on the second detection model to obtain the target detection result. The fourth data is the geographical environment data corresponding to the faulty line.

[0046] The second detection model can be understood as a dynamic fusion diagnostic model, which can perform fusion analysis on multiple types of data for the same event to obtain more accurate analysis results.

[0047] Specifically, if the diagnostic certainty index is less than or equal to the first threshold, it indicates that the transmission line tripping event corresponds to a highly complex and difficult scenario, and the inaccuracy of the first detection result is extremely high. This may be caused by a combination of multiple factors or the emergence of novel fault modes that have not been fully recorded in the training data. In this case, it is necessary to perform multi-dynamic fusion analysis on the first, second, third, and fourth data simultaneously to achieve cross-modal feature interaction and information complementarity, thereby obtaining the target detection result.

[0048] S140. If the diagnostic certainty index is greater than the first threshold and less than the second threshold, then the fifth data is determined, and the fifth data is analyzed based on the third detection model to obtain the target detection result; the fifth data includes the third data and the first reference time-series feature data; the reference time-series feature data is the multi-time-scale time-series feature extracted by the first detection model corresponding to the first detection result; the first threshold is less than the second threshold.

[0049] The third detection model can be understood as a medium-complexity multimodal verification model. Its detection accuracy is lower than that of the second detection model, but higher than that of the first detection model. The first and second thresholds are preset critical values ​​based on statistical analysis of diagnostic deterministic indicators of a large number of historical tripping event samples. They are used to classify the reliability of the detection results of the first detection model into three levels: high, medium, and low, thereby triggering different processing paths.

[0050] Specifically, if the diagnostic certainty index is greater than the first threshold and less than the second threshold, it indicates that the transmission line tripping event corresponds to a medium-complexity difficult scenario. The first detection result is relatively inaccurate, so third data needs to be introduced. The third detection model obtains the target detection result by fusing and analyzing the visual features of the third data with the multi-timescale temporal features extracted by the first detection model.

[0051] S150. If the diagnostic certainty index is greater than or equal to the second threshold, the first test result shall be used as the target test result.

[0052] Specifically, if the diagnostic certainty index is greater than or equal to the second threshold, it indicates that the first test result is reliable and can therefore be directly used as the target test result.

[0053] S160. Based on the target detection results, generate an emergency response plan for power transmission line tripping events.

[0054] Specifically, if there is a pre-defined correlation between different test results and emergency response plans, then after obtaining the target test result, the emergency response plan for the power transmission line tripping event can be matched from the pre-defined correlation based on the target test result.

[0055] Optionally, generating an emergency response plan for the transmission line tripping event based on the target detection results may include: generating an emergency response plan for the transmission line tripping event based on the target detection results and diagnostic deterministic indicators.

[0056] Specifically, the diagnostic certainty index can reflect the complexity of the transmission line tripping event and the accuracy of the first detection result. Specifically, if the diagnostic certainty index is less than or equal to the first threshold, the transmission line tripping event corresponds to a high-complexity scenario, and the first detection result corresponds to low certainty. In this case, the emergency response plan for the transmission line tripping event is a comprehensive emergency response plan. A comprehensive emergency response plan can be understood as needing to consider the possibility of multiple faults, deploy comprehensive inspection and response measures, and prepare multiple backup plans.

[0057] If the diagnostic certainty index is greater than the first threshold and less than the second threshold, the transmission line tripping event corresponds to a medium complexity scenario, and the first detection result corresponds to medium certainty. In this case, the emergency response plan for the transmission line tripping event is a verification-enhanced response plan. The verification-enhanced response plan can be understood as adding an on-site verification and confirmation step to the standard plan to balance efficiency and accuracy.

[0058] If the diagnostic certainty index is greater than or equal to the second threshold, the transmission line tripping event corresponds to a normal scenario, and the first detection result corresponds to high certainty. In this case, the emergency response plan for the transmission line tripping event is a standard rapid response plan. The standard rapid response plan can be understood as a plan generation strategy that aims to achieve the fastest response speed, in which the system fully trusts the first detection result.

[0059] Accordingly, based on the emergency response plan type and the dominant fault type of the target detection results, the emergency response plan elements are dynamically called from the pre-configured emergency response knowledge base and assembled to form an emergency response plan. The plan elements mainly include resource allocation strategies, planned patrol routes, response plans, and notification content.

[0060] Resource allocation strategies can be based on the dominant fault type identified in the target detection results, corresponding to the resources allocated accordingly. For example, if the dominant fault type is a hanging object fault or a vegetation obstruction fault, the system automatically dispatches the nearest aerial work platform and insulated operating poles, and notifies the relevant line maintenance teams. If the dominant fault type is an external force damage fault, while dispatching engineering vehicles, it generates on-site evidence collection and safety alert tasks, and notifies security personnel or contacts relevant law enforcement departments for collaborative handling. If the dominant fault type is a lightning strike trip fault or an equipment insulation fault, it prioritizes the allocation of insulation testing equipment and spare parts, and plans a sequence of towers that patrol personnel need to climb for inspection.

[0061] The planned inspection route can be the optimal inspection route generated for the inspection personnel based on the power grid geographic information data. At the same time, the planned inspection route can be an electronic map, which can highlight the sections and equipment that need to be inspected in particular based on the dominant fault type, making it easier to observe.

[0062] The response plan and notification content can be specific operational instructions and safety measures matched to the type of emergency response plan. For example, for a standard rapid response plan, a response work order containing clear operational steps can be directly generated and sent to the corresponding work team. For a verification-enhanced response plan, a site image feedback confirmation instruction can be embedded in the work order, requiring patrol personnel to first take photos of designated equipment upon arrival, send them back to the system for final confirmation, and then execute subsequent operations. For a comprehensive emergency response plan, a composite response plan containing multiple scenario assumptions can be generated, clearly defining the judgment criteria and transition procedures under different scenarios. Simultaneously, an abnormal situation warning notification to the superior dispatch center can be generated.

[0063] Optionally, the first detection model, the second detection model, the third detection model, the first threshold, and the second threshold of the present invention can all be optimized based on historical data at preset time intervals to ensure the accuracy of the first detection model, the second detection model, the third detection model, the first threshold, and the second threshold.

[0064] The technical solution of this invention involves acquiring first data, second data, and third data of the faulty line corresponding to a transmission line tripping event. The first data includes electrical quantity time-series data of the faulty line; the second data is meteorological monitoring data; and the third data is video stream data captured on the faulty line. First, the first and second data are analyzed based on a first detection model to obtain a first detection result, and a diagnostic deterministic index is determined to reflect the reliability of the fault type in the first detection result. Further, the diagnostic deterministic index is compared with a first threshold and a second threshold to determine whether to introduce a second or third detection model for data analysis, thereby determining the target detection result. When the diagnostic deterministic index is greater than or equal to the second threshold, the first detection result is used as the target detection result. This significantly shortens the diagnosis time. When the diagnostic certainty index is less than or equal to the first threshold, the geographical environment data corresponding to the faulty line is used as the fourth data. The first, second, third, and fourth data are analyzed based on the second detection model to obtain the target detection result, enabling the automatic activation of deeper multimodal fusion analysis and ensuring the reliability of the final diagnostic conclusion. When the diagnostic certainty index is greater than the first threshold and less than the second threshold, the fifth data, including the third data and the first reference time-series feature data, is determined. The fifth data is analyzed based on the third detection model to obtain the target detection result. The reference time-series feature data is the multi-time-scale time-series feature extracted by the first detection model corresponding to the first detection result. The introduction of visual data significantly improves the determination of faults. In other words, for simple faults with high determinism, this invention directly adopts preliminary results, greatly shortening the diagnosis time. When the uncertainty of the preliminary diagnosis is high, the system can sequentially initiate medium-complexity multimodal verification or high-complexity full-modal fusion analysis. By introducing multi-source information such as vision and geography at each level and performing deep feature interaction, the system significantly improves the success rate and robustness of diagnosis for complex scenarios such as compound faults and rare faults. Finally, based on the target detection results, an emergency response plan for transmission line tripping events is generated, improving the intelligence level, overall efficiency, and long-term adaptability of power grid fault emergency response.

[0065] Example 2

[0066] Figure 2 This is a flowchart of another power transmission trip diagnosis method provided by an embodiment of the present invention. The technical solution of this embodiment further optimizes the process of S120 in the aforementioned embodiments based on the above embodiments. This embodiment can be combined with various optional solutions in one or more of the above embodiments. Figure 2 As shown, the power transmission trip diagnosis method of the present invention may include:

[0067] S210. Obtain the first data, second data, and third data of the faulty line corresponding to the power transmission line tripping event; the first data includes the electrical quantity time sequence data of the faulty line; the second data is meteorological monitoring data; and the third data is video stream data captured on the faulty line.

[0068] S220. Analyze the first data and the second data based on the first detection model to obtain the first detection result and determine the diagnostic certainty index of the first detection result; the first detection model includes a temporal feature extraction layer, an attention-enhanced gating loop layer and multiple fault classification layers; each fault classification layer corresponds to a fault type; different fault types of fault classification layers correspond to different weight allocation strategies; the weight allocation strategy is used to describe the strategy of weighting features of a preset feature dimension; the diagnostic certainty index is used to reflect the reliability of the first detection result.

[0069] The temporal feature extraction layer includes three parallel temporal feature extraction branches: the first temporal feature extraction branch, the second temporal feature extraction branch, and the third temporal feature extraction branch. The feature tensors output by each branch are concatenated through the channel dimension to form the first temporal feature containing information at multiple time scales.

[0070] The first time-series feature extraction branch uses a convolutional kernel of size 3 to extract instantaneous abrupt change features at the second to minute level. For example, it can capture instantaneous current abrupt changes in lightning strike faults and instantaneous wind speed abrupt changes in strong wind flashover faults. The second time-series feature extraction branch uses a convolutional kernel of size 7 to extract trend change features at the minute to ten-minute level. For example, it can identify the gradual current change trend in icing galloping faults and the slow growth process of vegetation obstruction faults. The third time-series feature extraction branch uses a convolutional kernel of size 15 to extract periodic fluctuation features at the ten-minute to hour level. For example, it can analyze the periodic degradation patterns of equipment insulation faults and the seasonal patterns of animal-induced faults.

[0071] The attention-enhanced gating loop layer adds a time-attention mechanism for trip diagnosis to the traditional gating loop unit. In specific trip diagnosis scenarios, the calculation of the reset and update gates incorporates prior knowledge of fault characteristics. For example, a higher initial attention bias is assigned to the critical time period from 5 minutes before the trip to 1 minute after the trip; the reset gate activation value is enhanced for time steps where the zero-sequence current mutation exceeds the threshold; and the importance weight of the update gate is increased for time periods where wind speed mutations exceed safety limits. The attention-enhanced gating loop layer can automatically identify key time segments related to the trip event and suppress interference from irrelevant timing information.

[0072] Specifically, the analysis of the first and second data based on the first detection model to obtain the first detection result may include: extracting features from the first and second data based on the time-series feature extraction layer to obtain the first time-series features; the first time-series features include instantaneous change features, trend change features, and periodic fluctuation features; performing time-series dependency analysis on the first time-series features based on the attention-enhanced gating loop layer to obtain the feature analysis results; obtaining the attention weights of the first time-series features in different time periods based on the feature analysis results; and performing weighted fusion of the first time-series features based on the attention weights to obtain the second time-series features; the feature analysis results are the key time characteristics of the transmission line tripping event; and performing weighted calculation on the second time-series features based on the fault classification layer to obtain the first detection result.

[0073] Optionally, the feature analysis results include the first weights learned by the attention-enhanced gating recurrent layer based on the first temporal features. Obtaining the attention weights of the first temporal features for different time periods based on the feature analysis results may include: matching second weights from the prior knowledge base based on the first temporal features; and weighting and fusing the first weights and second weights according to the temporal information of the first temporal features to obtain the attention weights of the first temporal features. The process of weighting and fusing the first weights and second weights can be achieved by multiplying the first weights and second weights.

[0074] The prior knowledge base can be understood as the calculation rules for determining attention weights based on prior knowledge configuration for trip diagnosis. For example, the instant of tripping is assigned a basic weight coefficient of 1.2; the time point when zero-sequence current abnormality is detected within 10 minutes before tripping is assigned a weight coefficient of 1.1; the time point when wind speed exceeds the design standard is detected before or after tripping is assigned a weight coefficient of 1.15; and other time points maintain a basic weight coefficient of 1.0.

[0075] Furthermore, the weighted calculation of the second time-series features based on the fault classification layer to obtain the first detection result may include: obtaining the weight allocation strategy of the fault classification layer, performing weighted calculation of the second time-series features based on the weight allocation strategy, obtaining the detection result corresponding to each fault classification layer, and determining the detection result corresponding to each fault classification layer as the first detection result. The weight allocation strategy can be a differentiated setting based on the physical characteristics of transmission line tripping faults; for example, for meteorological faults, the focus is on weighting features related to wind speed, temperature, and humidity; for external force damage faults, the focus is on weighting features related to current surges and zero-sequence current; for equipment insulation faults, the focus is on weighting features related to voltage fluctuations and power factor. The process of weighted calculation of the second time-series features based on the weight allocation strategy can employ a weighted softmax function based on the physical characteristics of the fault.

[0076] In this embodiment of the invention, optionally, the first detection result includes the dominant probability strength, probability distribution dispersion, and temporal consistency score. The dominant probability strength reflects the confidence level of the fault type with the highest accuracy output by the first detection model. The probability distribution dispersion reflects the degree of concentration of the probability distribution of each fault type; the probability distribution dispersion can be calculated using normalized Shannon entropy. The lower the entropy value, the more concentrated the probability is on a certain fault type, and the higher the certainty. The temporal consistency score reflects the stability of the diagnostic results before and after a transmission line tripping event; for example, the 60-minute time window is divided into 12 sub-segments with 5-minute intervals, the reference detection results for each sub-segment are calculated, and then the consistency ratio between the dominant fault type and the reference detection results in the first detection result is calculated.

[0077] Accordingly, the diagnostic certainty index for determining the first test result may include: weighted fusion of the dominant probability strength, probability distribution dispersion, and temporal consistency score to determine the diagnostic certainty index for the first test result, thereby achieving accuracy in evaluating the first test result.

[0078] S230. If the diagnostic certainty index is less than or equal to the first threshold, then the fourth data is obtained. The first data, the second data, the third data and the fourth data are analyzed based on the second detection model to obtain the target detection result. The fourth data is the geographical environment data corresponding to the faulty line.

[0079] S240. If the diagnostic certainty index is greater than the first threshold and less than the second threshold, then the fifth data is determined, and the fifth data is analyzed based on the third detection model to obtain the target detection result; the fifth data includes the third data and the first reference time-series feature data; the reference time-series feature data is the multi-time-scale time-series feature extracted by the first detection model corresponding to the first detection result; the first threshold is less than the second threshold.

[0080] S250. If the diagnostic certainty index is greater than or equal to the second threshold, the first test result shall be used as the target test result.

[0081] S260. Based on the target detection results, generate an emergency response plan for power transmission line tripping events.

[0082] The technical solution of this invention involves acquiring first data, second data, and third data of the faulty line corresponding to a power transmission line tripping event. The first data includes time-series electrical quantity data of the faulty line; the second data is meteorological monitoring data; and the third data is video stream data captured on the faulty line. The first and second data are analyzed based on a first detection model to obtain a first detection result, and a diagnostic deterministic index for the first detection result is determined. The first detection model includes a time-series feature extraction layer, an attention-enhanced gating loop layer, and multiple fault classification layers. Each fault classification layer corresponds to a fault type. Different fault types correspond to different weight allocation strategies. The weight allocation strategy describes the strategy for weighting features of a preset feature dimension. The first detection model is used to quickly perform a preliminary diagnosis of the tripping event, generating a directional first detection result and determining the diagnostic deterministic index of the first detection result, providing a decision-making basis for subsequent potentially triggered in-depth analysis. Further, the diagnostic deterministic indicators are compared with the first and second thresholds to determine whether to introduce a second or third detection model for data analysis, and to determine the target detection results. This allows for the generation of emergency response plans for transmission line tripping events based on the target detection results, thereby improving the intelligence level, overall efficiency, and long-term adaptability of power grid fault emergency response.

[0083] Example 3

[0084] Figure 3 This is a flowchart of another power transmission trip diagnosis method provided by an embodiment of the present invention. The technical solution of this embodiment further optimizes the process of S130 in the aforementioned embodiments based on the above embodiments. This embodiment can be combined with various optional solutions in one or more of the above embodiments. For example... Figure 3 As shown, the power transmission trip diagnosis method of the present invention may include:

[0085] S310. Obtain the first data, second data and third data of the faulty line corresponding to the power transmission line tripping event; the first data includes the electrical quantity time sequence data of the faulty line; the second data is meteorological monitoring data; the third data is video stream data of the faulty line.

[0086] S320. Analyze the first data and the second data based on the first detection model to obtain the first detection result, and determine the diagnostic certainty index of the first detection result; the diagnostic certainty index is used to reflect the reliability of the first detection result.

[0087] S330. If the diagnostic certainty index is less than or equal to the first threshold, then the fourth data is obtained. The first data, the second data, the third data, and the fourth data are analyzed based on the second detection model to obtain the target detection result. The fourth data is the geographical environment data corresponding to the faulty line. The second detection model includes a first feature coding layer, a second feature coding layer, a third feature coding layer, a dynamic fusion control layer, and a full-modal classifier. The first feature coding layer, the second feature coding layer, and the third feature coding layer are parallel feature coding layers.

[0088] The fourth type of data can be GIS data retrieved from a GIS platform. The GIS data can include spatial attribute information such as the coordinates of the towers, the route, the elevation difference, the insulator type, the surrounding topography, and the construction permit area within the buffer zone.

[0089] The first feature encoding layer can include a Transformer encoder with a multi-head self-attention mechanism to capture long-term dependencies in meteorological and electrical quantity sequences; the second feature encoding layer can be a ResNet-50 convolutional neural network to extract spatial structure and texture features from images; the third feature encoding layer can be a graph attention network to model the power grid topology to obtain environmental correlation features. A dynamic fusion control layer is used to intelligently fuse deep feature vectors from various modalities. The full-modal classifier has a multi-layer perceptron structure and can output the probability distribution of all fault types through a Softmax function.

[0090] Specifically, features are extracted from the first and second data based on the first feature encoding layer to obtain the third temporal features; features are extracted from the third data based on the second feature encoding layer to obtain the first visual features; features are extracted from the fourth data based on the third feature encoding layer to obtain the geographic features; the third temporal features, the first visual features, and the geographic features are weighted and fused based on the dynamic fusion control layer to obtain the first fused feature vector; and the first fused feature vector is analyzed based on the full-modal classifier to determine the target detection result.

[0091] S340. If the diagnostic certainty index is greater than the first threshold and less than the second threshold, then the fifth data is determined, and the fifth data is analyzed based on the third detection model to obtain the target detection result; the fifth data includes the third data and the first reference time-series feature data; the reference time-series feature data is the multi-time-scale time-series feature extracted by the first detection model corresponding to the first detection result; the first threshold is less than the second threshold.

[0092] S350. If the diagnostic certainty index is greater than or equal to the second threshold, then the first test result shall be taken as the target test result.

[0093] S360: Based on the target detection results, generate an emergency response plan for power transmission line tripping events.

[0094] The technical solution of this invention involves acquiring first data, second data, and third data of the faulty line corresponding to a transmission line tripping event. The first data includes electrical quantity time-series data of the faulty line; the second data is meteorological monitoring data; and the third data is video stream data captured on the faulty line. Based on a first detection model, the first and second data are analyzed to obtain a first detection result, and a diagnostic deterministic index reflecting the reliability of the fault type in the first detection result is determined. Furthermore, the diagnostic deterministic index is compared with a first threshold and a second threshold to determine whether to introduce a second or third detection model for data analysis, thereby determining the target detection result. Furthermore, if the diagnostic certainty index is less than or equal to the first threshold, the fourth data is acquired. Based on the second detection model, the first, second, third, and fourth data are analyzed to obtain the target detection result. The fourth data is the geographical environment data corresponding to the faulty line. The second detection model includes a first feature encoding layer, a second feature encoding layer, a third feature encoding layer, a dynamic fusion control layer, and a full-modal classifier. The first, second, and third feature encoding layers are parallel feature encoding layers, enabling the system to initiate high-complexity full-modal fusion analysis when the initial diagnosis uncertainty is high. By introducing multi-source information such as visual and geographical information step by step and conducting deep feature interaction, the system significantly improves the diagnostic success rate and robustness for complex scenarios such as composite faults and rare faults. Finally, based on the target detection result, an emergency response plan for transmission line tripping events is generated, improving the intelligence level, overall efficiency, and long-term adaptability of power grid fault emergency response.

[0095] Example 4

[0096] Figure 4 This is a flowchart of another power transmission trip diagnosis method provided by an embodiment of the present invention. The technical solution of this embodiment further optimizes the process of S140 in the aforementioned embodiments based on the above embodiments. This embodiment can be combined with various optional solutions in one or more of the above embodiments. For example... Figure 4 As shown, the power transmission trip diagnosis method of the present invention may include:

[0097] S410. Obtain the first data, second data, and third data of the faulty line corresponding to the power transmission line tripping event; the first data includes the electrical quantity time sequence data of the faulty line; the second data is meteorological monitoring data; and the third data is video stream data captured on the faulty line.

[0098] S420. Analyze the first data and the second data based on the first detection model to obtain the first detection result, and determine the diagnostic certainty index of the first detection result; the diagnostic certainty index is used to reflect the reliability of the first detection result.

[0099] S430. If the diagnostic certainty index is less than or equal to the first threshold, then the fourth data is obtained. The first data, the second data, the third data and the fourth data are analyzed based on the second detection model to obtain the target detection result. The fourth data is the geographical environment data corresponding to the faulty line.

[0100] S440. If the diagnostic certainty index is greater than the first threshold and less than the second threshold, then the fifth data is determined. The fifth data is analyzed based on the third detection model to obtain the target detection result. The fifth data includes the third data and the first reference temporal feature data. The reference temporal feature data is the multi-timescale temporal feature extracted by the first detection model corresponding to the first detection result. The first threshold is less than the second threshold. The third detection model includes a visual feature encoder, a feature alignment layer, a cross-attention fusion layer, and a validation classifier.

[0101] The validation classifier is not only used for correct fault classification, but also focuses on correcting the uncertainty in the detection results obtained by the first detection model.

[0102] Specifically, the second visual feature is obtained by extracting features from the third data based on the visual feature encoder; the second visual feature and the first reference temporal feature data are mapped to a unified feature space based on the feature alignment layer to obtain the third visual feature and the second reference temporal feature data; the third visual feature and the second reference temporal feature data are processed based on the cross-attention fusion layer to obtain the second fused feature vector; and the second fused feature vector is analyzed based on the validation classifier to determine the target detection result.

[0103] Furthermore, extracting features from the third data based on the visual feature encoder to obtain the second visual feature may include: determining the target fault type with the highest confidence in the first detection result, and extracting features of the target fault type from the third data based on the visual feature encoder to obtain the second visual feature.

[0104] For example, if the target fault type is a hanging object fault, the characteristics of the target fault type can be the shape characteristics, color and texture characteristics of the linear foreign object above the conductor, and its spatial positional relationship with the conductor. If the target fault type is an external force damage fault, the characteristics of the target fault type can be the Hough circle characteristics of the construction machinery (used to identify crane booms and excavator tracks), the motion vector field characteristics, and the motion trajectory characteristics relative to the line protection zone. If the target fault type is a vegetation obstruction fault, the characteristics of the target fault type can be the time-series variation characteristics of the normalized vegetation index of the vegetation canopy and the minimum spatial distance characteristics between the vegetation outline and the conductor.

[0105] Accordingly, processing the third visual feature and the second reference temporal feature data based on the cross-attention fusion layer to obtain the second fusion feature vector may include: using the second reference temporal feature data as a query, using the third visual feature as the key and value to obtain the reference weight, using the reference weight to weight the third visual feature to obtain the weighted third visual feature, and concatenating the weighted third visual feature with the second reference temporal feature data to obtain the second fusion feature vector.

[0106] The reference weights can be represented by the following formula:

[0107] ;

[0108] Where Q represents the second reference temporal feature data, K and V represent the third visual features, and d k For feature dimensions.

[0109] In this embodiment of the invention, the process of using second reference time-series feature data as a query and third visual features as keys and values ​​to obtain weighted third visual features aims to enable time-series features to actively query visual features for evidence that can verify or refute their own hypotheses.

[0110] S450. If the diagnostic certainty index is greater than or equal to the second threshold, then the first test result shall be taken as the target test result.

[0111] S460. Based on the target detection results, generate an emergency response plan for power transmission line tripping events.

[0112] The technical solution of this invention involves acquiring first data, second data, and third data of the faulty line corresponding to a transmission line tripping event. The first data includes electrical quantity time-series data of the faulty line; the second data is meteorological monitoring data; and the third data is video stream data captured on the faulty line. Based on a first detection model, the first and second data are analyzed to obtain a first detection result, and a diagnostic deterministic index reflecting the reliability of the first detection result is determined. Furthermore, the diagnostic deterministic index is compared with a first threshold and a second threshold to determine whether to introduce a second or third detection model for data analysis, thereby determining the target detection result. When the diagnostic certainty index is greater than the first threshold and less than the second threshold, the fifth data is analyzed based on the third detection model to obtain the target detection result. The fifth data includes the third data and the first reference temporal feature data. The reference temporal feature data is the multi-timescale temporal feature extracted by the first detection model corresponding to the first detection result. The first threshold is less than the second threshold. The third detection model includes a visual feature encoder, a feature alignment layer, a cross-attention fusion layer, and a validation classifier. This enables the initiation of medium-complexity multimodal validation when uncertainty is high. By introducing visual data, the determination of faults is significantly improved. Finally, an emergency response plan for transmission line tripping events is generated based on the target detection results, improving the intelligence level, overall efficiency, and long-term adaptability of power grid fault emergency response.

[0113] Example 5

[0114] Figure 5 This is a schematic diagram of a power transmission tripping diagnostic device provided in an embodiment of the present invention. This embodiment is applicable to the diagnosis of power transmission line tripping events. The power transmission tripping diagnostic device can be implemented in hardware and / or software, and can be configured in any electronic device with network communication capabilities. Figure 5 As shown, the power transmission trip diagnostic device of the present invention may include:

[0115] The data acquisition module 510 is used to acquire first data, second data, and third data of the faulty line corresponding to the power transmission line tripping event; the first data includes electrical quantity time-series data of the faulty line; the second data is meteorological monitoring data; and the third data is video stream data captured on the faulty line.

[0116] The data processing module 520 is used to analyze the first data and the second data based on the first detection model to obtain a first detection result and determine the diagnostic certainty index of the first detection result; the diagnostic certainty index is used to reflect the reliability of the first detection result.

[0117] The first judgment module 530 is used to obtain fourth data if the diagnostic certainty index is less than or equal to the first threshold, and to analyze the first data, the second data, the third data and the fourth data based on the second detection model to obtain the target detection result; the fourth data is the geographical environment data corresponding to the faulty line.

[0118] The second judgment module 540 is used to determine the fifth data if the diagnostic certainty index is greater than the first threshold and the diagnostic certainty index is less than the second threshold, and to analyze the fifth data based on the third detection model to obtain the target detection result; the fifth data includes the third data and the first reference time series feature data; the reference time series feature data is the multi-time scale time series feature extracted by the first detection model corresponding to the first detection result; the first threshold is less than the second threshold;

[0119] The third judgment module 550 is used to take the first detection result as the target detection result if the diagnostic certainty index is greater than or equal to the second threshold.

[0120] The scheme determination module 560 is used to generate an emergency response plan for the power transmission line tripping event based on the target detection results.

[0121] Based on the above embodiments, optionally, the first detection model includes a temporal feature extraction layer, an attention-enhanced gated recurrent layer, and multiple fault classification layers; each fault classification layer corresponds to a fault type; different fault type fault classification layers correspond to different weight allocation strategies; the weight allocation strategy is used to describe the strategy of weighting features of a preset feature dimension;

[0122] Accordingly, the data processing module is used to: extract features from the first data and the second data based on the time-series feature extraction layer to obtain a first time-series feature; the first time-series feature includes instantaneous change features, trend change features, and periodic fluctuation features; perform time-series dependency analysis on the first time-series feature based on the attention-enhanced gating loop layer to obtain feature analysis results; obtain attention weights for the first time-series feature in different time periods based on the feature analysis results; and perform weighted fusion of the first time-series feature based on the attention weights to obtain a second time-series feature; the feature analysis result is the key time characteristic of the transmission line tripping event; and perform weighted calculation on the second time-series feature based on the fault classification layer to obtain a first detection result.

[0123] Based on the above embodiments, optionally, the first detection result includes dominant probability strength, probability distribution dispersion, and temporal consistency score; the dominant probability strength reflects the confidence level of the fault type with the highest accuracy output by the first detection model; the probability distribution dispersion is used to reflect the degree of concentration of the probability distribution of each fault type; and the temporal consistency score is used to reflect the stability of the diagnostic results before and after the transmission line tripping event.

[0124] Accordingly, the data processing module is also used to: perform weighted fusion of the dominant probability intensity, probability distribution dispersion and temporal consistency score to determine the diagnostic deterministic index of the first detection result.

[0125] Based on the above embodiments, optionally, the second detection model includes a first feature encoding layer, a second feature encoding layer, a third feature encoding layer, a dynamic fusion control layer, and a full-modal classifier; the first feature encoding layer, the second feature encoding layer, and the third feature encoding layer are parallel feature encoding layers;

[0126] Accordingly, the first judgment module is used to: extract features from the first data and the second data based on the first feature encoding layer to obtain a third temporal feature; extract features from the third data based on the second feature encoding layer to obtain a first visual feature; extract features from the fourth data based on the third feature encoding layer to obtain a geographic feature; perform weighted fusion of the third temporal feature, the first visual feature, and the geographic feature based on the dynamic fusion control layer to obtain a first fused feature vector; and analyze the first fused feature vector based on the full-modal classifier to determine the target detection result.

[0127] Based on the above embodiments, the third detection model may optionally include a visual feature encoder, a feature alignment layer, a cross-attention fusion layer, and a verification classifier;

[0128] Accordingly, the second judgment module is used to: extract features from the third data based on the visual feature encoder to obtain a second visual feature; map the second visual feature and the first reference temporal feature data to a unified feature space based on the feature alignment layer to obtain a third visual feature and a second reference temporal feature data; process the third visual feature and the second reference temporal feature data based on the cross-attention fusion layer to obtain a second fused feature vector; and analyze the second fused feature vector based on the verification classifier to determine the target detection result.

[0129] Based on the above embodiments, optionally, the second judgment module includes a visual feature extraction unit, which is used to: determine the target fault type with the highest confidence in the first detection result, and extract the features of the target fault type in the third data based on the visual feature encoder to obtain a second visual feature.

[0130] Based on the above embodiments, optionally, the scheme determination module is used to: generate an emergency response plan for the power transmission line tripping event based on the target detection results and the diagnostic deterministic indicators.

[0131] The power transmission trip diagnosis device provided in this embodiment of the invention can execute the power transmission trip diagnosis method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0132] Example 6

[0133] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0134] Figure 6 A schematic diagram of an electronic device that can be used to implement the power transmission trip diagnosis method according to embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0135] like Figure 6 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0136] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0137] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as power transmission trip diagnosis methods.

[0138] In some embodiments, the transmission trip diagnosis method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via read-only memory (ROM) 12 and / or communication unit 19. When the computer program is loaded into random access memory (RAM) 13 and executed by processor 11, one or more steps of the transmission trip diagnosis method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the transmission trip diagnosis method by any other suitable means (e.g., by means of firmware).

[0139] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0140] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0141] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0142] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0143] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0144] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through a communication network. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0145] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0146] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for diagnosing power transmission tripping, characterized in that, The method includes: Acquire first data, second data, and third data of the faulty line corresponding to the power transmission line tripping event; the first data includes electrical quantity time-series data of the faulty line; the second data is meteorological monitoring data; and the third data is video stream data captured on the faulty line. The first detection model is used to analyze the first data and the second data to obtain a first detection result, and a diagnostic certainty index for the first detection result is determined; the diagnostic certainty index is used to reflect the reliability of the first detection result. If the diagnostic certainty index is less than or equal to the first threshold, then the fourth data is obtained, and the first data, the second data, the third data and the fourth data are analyzed based on the second detection model to obtain the target detection result; the fourth data is the geographical environment data corresponding to the faulty line. If the diagnostic certainty index is greater than the first threshold and less than the second threshold, then the fifth data is determined, and the fifth data is analyzed based on the third detection model to obtain the target detection result; the fifth data includes the third data and the first reference time-series feature data; the reference time-series feature data is the multi-time-scale time-series feature extracted by the first detection model corresponding to the first detection result; the first threshold is less than the second threshold; If the diagnostic certainty index is greater than or equal to the second threshold, then the first detection result is taken as the target detection result; Based on the target detection results, an emergency response plan for the power transmission line tripping event is generated.

2. The method according to claim 1, characterized in that, The first detection model includes a temporal feature extraction layer, an attention-enhanced gated recurrent layer, and multiple fault classification layers; each fault classification layer corresponds to a fault type; different fault classification layers for different fault types correspond to different weight allocation strategies; The weighting strategy is used to describe the strategy for weighting features of a preset feature dimension; Accordingly, the first data and the second data are analyzed based on the first detection model to obtain a first detection result, including: Based on the time-series feature extraction layer, features are extracted from the first data and the second data to obtain the first time-series feature; the first time-series feature includes instantaneous change feature, trend change feature and periodic fluctuation feature; Based on the attention-enhanced gating recurrent layer, the first temporal feature is subjected to temporal dependency analysis to obtain the feature analysis result. Based on the feature analysis result, the attention weights of the first temporal feature in different time periods are obtained. Based on the attention weights, the first temporal feature is weighted and fused to obtain the second temporal feature. The feature analysis result is the key temporal characteristic of the power transmission line tripping event. The second time-series features are weighted and calculated based on the fault classification layer to obtain the first detection result.

3. The method according to claim 2, characterized in that, The first detection result includes the dominant probability strength, probability distribution dispersion, and temporal consistency score; the dominant probability strength reflects the confidence level of the fault type with the highest accuracy output by the first detection model. The probability distribution dispersion is used to reflect the degree of concentration of the probability distribution of each fault type; The timing consistency score is used to reflect the stability of diagnostic results before and after a transmission line tripping event; Accordingly, the diagnostic certainty indicators for the first test result include: The dominant probability strength, probability distribution dispersion, and temporal consistency score are weighted and fused to determine the diagnostic deterministic index of the first detection result.

4. The method according to claim 1, characterized in that, The second detection model includes a first feature encoding layer, a second feature encoding layer, a third feature encoding layer, a dynamic fusion control layer, and a full-modality classifier; the first feature encoding layer, the second feature encoding layer, and the third feature encoding layer are parallel feature encoding layers; Accordingly, based on the second detection model, the first data, the second data, the third data, and the fourth data are analyzed to obtain target detection results, including: Based on the first feature encoding layer, feature extraction is performed on the first data and the second data to obtain the third temporal feature; Based on the second feature encoding layer, feature extraction is performed on the third data to obtain the first visual feature; Based on the third feature encoding layer, feature extraction is performed on the fourth data to obtain geographical features; Based on the dynamic fusion control layer, the third temporal feature, the first visual feature, and the geographical feature are weighted and fused to obtain the first fused feature vector; The first fused feature vector is analyzed based on the full-modal classifier to determine the target detection result.

5. The method according to claim 1, characterized in that, The third detection model includes a visual feature encoder, a feature alignment layer, a cross-attention fusion layer, and a validation classifier. Accordingly, the fifth data is analyzed based on the third detection model to obtain target detection results, including: Based on the visual feature encoder, feature extraction is performed on the third data to obtain the second visual feature; Based on the feature alignment layer, the second visual feature and the first reference temporal feature data are mapped to a unified feature space to obtain the third visual feature and the second reference temporal feature data. The third visual feature and the second reference temporal feature data are processed based on the cross-attention fusion layer to obtain the second fused feature vector; The second fused feature vector is analyzed based on the verification classifier to determine the target detection result.

6. The method according to claim 5, characterized in that, Based on the visual feature encoder, feature extraction is performed on the third data to obtain second visual features, including: The target fault type with the highest confidence in the first detection result is determined, and the features of the target fault type in the third data are extracted based on the visual feature encoder to obtain the second visual feature.

7. The method according to claim 1, characterized in that, Based on the target detection results, an emergency response plan for the power transmission line tripping event is generated, including: Based on the target detection results and the diagnostic deterministic indicators, an emergency response plan for the power transmission line tripping event is generated.

8. A power transmission trip diagnostic device, characterized in that, The device includes: The data acquisition module is used to acquire first data, second data, and third data of the faulty line corresponding to the power transmission line tripping event; the first data includes electrical quantity time-series data of the faulty line; the second data is meteorological monitoring data; and the third data is video stream data captured on the faulty line. The data processing module is used to analyze the first data and the second data based on the first detection model to obtain a first detection result and determine the diagnostic certainty index of the first detection result; the diagnostic certainty index is used to reflect the reliability of the first detection result. The first judgment module is used to obtain fourth data if the diagnostic certainty index is less than or equal to the first threshold, and to analyze the first data, the second data, the third data and the fourth data based on the second detection model to obtain the target detection result; the fourth data is the geographical environment data corresponding to the faulty line. The second judgment module is used to determine the fifth data if the diagnostic certainty index is greater than the first threshold and the diagnostic certainty index is less than the second threshold, and to analyze the fifth data based on the third detection model to obtain the target detection result; the fifth data includes the third data and the first reference time series feature data; the reference time series feature data is the multi-time scale time series feature extracted by the first detection model corresponding to the first detection result; the first threshold is less than the second threshold; The third judgment module is used to take the first detection result as the target detection result if the diagnostic certainty index is greater than or equal to the second threshold. The scheme determination module is used to generate an emergency response plan for the power transmission line tripping event based on the target detection results.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the power transmission trip diagnosis method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that are used to cause a processor to execute the power transmission trip diagnosis method according to any one of claims 1-7.

Images