A direct current transmission line operation and maintenance method and system and storage medium

Abstract

The application relates to the field of line operation and maintenance, in particular to a DC transmission line operation and maintenance method and system and a storage medium, wherein the method comprises collecting a feature data set and multi-modal data in a DC transmission line, calculating data discrete values of the feature data set; based on the data discrete values, performing a fusion operation on the multi-modal data to generate fusion data; inputting the fusion data into a space-time feature fusion model and outputting operation feature data; based on the operation feature data, generating an operation health index of the DC transmission line, and performing operation and maintenance on the DC transmission line according to the operation health index.

Description

Technical Field

[0001] This application relates to the field of line operation and maintenance, and in particular to a method, system and storage medium for the operation and maintenance of DC transmission lines. Background Technology

[0002] With the widespread application of sensor technology, smart devices, and information technology in power systems, a large amount of multi-source heterogeneous data is generated during the operation and maintenance of DC transmission lines. Multimodal data fusion technology can organically integrate data of different types and sources, improving the accuracy and reliability of the assessment of the operating status of DC transmission equipment.

[0003] In relevant DC transmission operation and maintenance technologies, most rely on single electrical quantity data and fixed model structures. However, due to the diversity of data in the DC transmission operation and maintenance process, the assessment of DC transmission operation status is highly variable and unstable. At the same time, existing technologies analyze based on single electrical parameters, which greatly reduces the flexibility and adaptability of DC transmission operation and maintenance assessment technology solutions. Summary of the Invention

[0004] The main objective of this application is to provide a method, system, and storage medium for the operation and maintenance of DC transmission lines, in order to solve the aforementioned problems.

[0005] To achieve the above objectives, one aspect of this application proposes a method for the operation and maintenance of a DC transmission line, the method comprising: Collect feature datasets and multimodal data from DC transmission lines, and calculate the discrete values ​​of the feature datasets; Based on the discrete values ​​of the data, a fusion operation is performed on the multimodal data to generate fused data; The fused data is input into the spatiotemporal feature fusion model, and the running feature data is output. Based on the operational characteristic data, operational health indicators for DC transmission lines are generated, and the DC transmission lines are maintained and operated according to these operational health indicators.

[0006] In some embodiments, calculating the discrete values ​​of the feature dataset specifically includes: The feature dataset is averaged to generate a feature comparison dataset, where the feature comparison dataset values ​​include the collection frequency, collection duration, collection data volume, and collection success rate. Calculate the degree of deviation between the feature dataset and the feature comparison dataset, perform weighted aggregation on the degree of deviation, and generate discrete data values.

[0007] In some embodiments, before performing the fusion operation on the multimodal data based on the discrete data values ​​to generate fused data, the method further includes: The interval classification of the discrete data values ​​is determined by a predefined interval of discrete data values. If the interval of the discrete data values ​​is classified as the first discrete value interval, then no correction is needed for the fusion operation; If the interval of the discrete values ​​of the data is classified as a second discrete value interval or a third discrete value interval, then the fused data is corrected.

[0008] In some embodiments, if the interval of the discrete data values ​​is classified into a second discrete value interval or a third discrete value interval, the fusion operation is modified, specifically including: If the interval of the discrete data value is classified as the second discrete value interval, the minimum value of the second discrete value interval is extracted, and the difference between the discrete data value and the minimum value of the second discrete value interval is processed to generate the discrete data deviation. Based on the discrete data deviation, the time axis span calibration value and the coordinate axis span calibration value are obtained. The initial values ​​of the axis span and coordinate axis span of the fusion operation are collected and added to the time axis span calibration value and coordinate axis span calibration value, respectively, to generate the time axis span correction value and coordinate axis span correction value. If the interval of the data discrete values ​​is classified as a third discrete value interval, the data discrete values ​​are proportionally processed to the third discrete value interval to obtain a first deviation ratio. Based on the first deviation ratio, an interval calibration value is obtained. The initial set of extreme values ​​in the fusion operation is collected and processed with the interval calibration value to obtain the data normalization correction interval.

[0009] In some embodiments, performing a fusion operation on the multimodal data based on the discrete data values ​​to generate fused data specifically includes: If the interval of the discrete values ​​of the data is classified as the first discrete value interval, then the multimodal data of the DC transmission line is extracted, and the fusion operation is performed according to the various initial values ​​of the multimodal data fusion operation to generate fused data; If the interval of the discrete values ​​of the data is classified as the second discrete value interval, then the multimodal data of the DC transmission line is extracted, and a multimodal data fusion operation is performed with the time axis span correction value and the coordinate axis span correction value to generate fused data; If the interval of the discrete data value is classified as the third discrete value interval, then the multimodal data of the DC transmission line is extracted, and a multimodal data fusion operation is performed using the time axis span correction value, the coordinate axis span correction value, and the data normalization correction interval.

[0010] In some embodiments, generating operational health indicators for DC transmission lines based on the operational characteristic data specifically includes: The operational characteristic data are normalized and weighted, and then aggregated to generate operational health indicators for DC transmission lines.

[0011] In some embodiments, prior to performing maintenance on the DC transmission line based on the operational health indicators, the method further includes: The operational health indicators are compared with predefined operational health indicators at thresholds; Based on the threshold comparison results, the necessity of maintaining the DC transmission line is determined.

[0012] In some embodiments, the operational characteristic data includes the partial discharge phase asymmetry of the DC transmission line, the cross-sensor correlation coefficient of the DC transmission line, the coupling entropy of the environmental and electrical parameters of the DC transmission line, and the proportion of leakage current harmonics of the DC transmission line.

[0013] To achieve the above objectives, another aspect of this application proposes a DC transmission line operation and maintenance system, the system comprising: The data acquisition module is used to acquire feature datasets and multimodal data in DC transmission lines, and to calculate the discrete values ​​of the feature datasets. The data fusion module is used to perform a fusion operation on the multimodal data based on the discrete values ​​of the data to generate fused data; The feature extraction module is used to input the fused data into the spatiotemporal feature fusion model and output the running feature data; The operation and maintenance execution module is used to generate operation and health indicators for DC transmission lines based on the operation characteristic data, and to perform operation and maintenance on DC transmission lines according to the operation and health indicators.

[0014] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0015] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0016] The embodiments of this application include at least the following beneficial effects: This application provides a method, system, and storage medium for the operation and maintenance of DC transmission lines. By collecting feature datasets from DC transmission lines and calculating the data dispersion values ​​of these datasets, the application can measure the degree of data acquisition discrepancies in the DC transmission lines. Smaller dispersion values ​​indicate more consistent data acquisition across the DC transmission lines, suggesting higher data reliability and stability. Conversely, larger dispersion values ​​indicate acquisition discrepancies, which may affect the accuracy of data fusion and the reliability of subsequent analysis. Based on the data dispersion values, it can be determined whether the multimodal data fusion operation needs correction to improve the quality of data fusion and make the fused data more accurately reflect the actual situation of the DC transmission lines. Attached Figure Description

[0017] Figure 1 A flowchart of a DC transmission line operation and maintenance method provided in an embodiment of this application; Figure 2 This is a schematic diagram illustrating the correction and judgment of the fusion operation provided in the embodiments of this application; Figure 3 This is a schematic diagram of operation and maintenance execution provided in the embodiments of this application; Figure 4 A block diagram of a DC transmission line operation and maintenance system provided in an embodiment of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying 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 those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0019] refer to Figure 1 As shown, this application proposes a method for the operation and maintenance of DC transmission lines, which includes, but is not limited to, steps S101-S104, specifically: S101: Collect feature datasets from DC transmission lines and calculate the discrete values ​​of the feature datasets; S102: Based on the discrete values ​​of the data, perform a fusion operation on the multimodal data to generate fused data; S103: Input the fused data into the spatiotemporal feature fusion model and output the running feature data; S104: Based on the operational characteristic data, generate operational health indicators for the DC transmission line, and perform operation and maintenance on the DC transmission line according to the operational health indicators.

[0020] By collecting feature datasets from the DC transmission line through steps S101-S104 and calculating the data dispersion values ​​of these datasets, the degree of data acquisition discrepancies in the DC transmission line can be measured. Smaller dispersion values ​​indicate more consistent acquisition across the DC transmission line, suggesting higher data reliability and stability. Conversely, larger dispersion values ​​indicate acquisition discrepancies, potentially affecting the accuracy of data fusion and the reliability of subsequent analysis. Based on the data dispersion values, it can be determined whether the multimodal data fusion operation needs correction to improve the quality of data fusion and ensure that the fused data more accurately reflects the actual situation of the DC transmission line.

[0021] Specifically, step S101, calculating the discrete values ​​of the feature dataset, includes: The feature datasets of each data collection point are collected and averaged to obtain the collection feature comparison dataset of the DC transmission line. The specific feature datasets can be extracted from the execution reports of each data collection point.

[0022] Based on the feature datasets of each data acquisition point and the comparison dataset of acquisition features of DC transmission lines, the discrete values ​​of the DC transmission line data are determined.

[0023] The characteristic datasets of each data collection point include the collection frequency, single collection duration, single collection data volume, and data collection success rate of each data collection point.

[0024] The dataset comparing the acquisition characteristics of DC transmission lines includes the average acquisition frequency, average single acquisition duration, average single acquisition data volume, and average data acquisition success rate of DC transmission lines.

[0025] It should be explained that the above average values ​​are obtained by averaging the corresponding parameters of each data acquisition point. For example, the average acquisition frequency of a DC transmission line is obtained by averaging the acquisition frequencies of each data acquisition point.

[0026] The deviations of the acquisition frequency of each data acquisition point from the average acquisition frequency of the DC transmission line, the deviations of the single acquisition duration of each data acquisition point from the average single acquisition duration of the DC transmission line, the deviations of the single acquisition data volume of each data acquisition point from the average single acquisition data volume of the DC transmission line, and the deviations of the data acquisition success rate of each data acquisition point from the average data acquisition success rate of the DC transmission line are sequentially weighted and aggregated to obtain the discrete value of the DC transmission line data. The specific formula for calculating the discrete value is as follows: ; in For discrete data values ​​of DC transmission lines, Let be the sampling frequency of the i-th data acquisition point, and let i be the number of each data acquisition point. M represents the total number of data collection points. This represents the average sampling frequency of the DC transmission line. Let be the duration of a single data collection session for the i-th data collection point. This represents the average duration of a single data collection session for a DC transmission line. Let i be the amount of data collected in a single session at the i-th data collection point. This represents the average amount of data collected in a single session for a DC transmission line. Let be the data acquisition success rate of the i-th data acquisition point. This represents the average data acquisition success rate for DC transmission lines. The weight element corresponding to the predefined acquisition frequency deviation in the power transmission operation and maintenance database. The weight element corresponding to the predefined single-sampling duration deviation in the power transmission operation and maintenance database. The weight element corresponding to the deviation of a single data acquisition is predefined in the power transmission operation and maintenance database. The weight element corresponding to the predefined data acquisition success rate deviation in the power transmission operation and maintenance database.

[0027] It should be explained that the above data collection success rate refers to the ratio of the number of data samples successfully collected in a single instance to the planned number of samples to be collected, reflecting the reliability of the collection process.

[0028] The weight elements corresponding to the sampling frequency deviation, single sampling duration deviation, single sampling data volume deviation, and data acquisition success rate deviation are all extracted from the power transmission operation and maintenance database. The mapping relationship can be one-to-one or many-to-one. For example, the sampling frequency deviation, single sampling duration deviation, single sampling data volume deviation, and data acquisition success rate deviation are respectively mapped to the preset weight elements corresponding to the sampling frequency deviation, single sampling duration deviation, single sampling data volume deviation, and data acquisition success rate deviation in the power transmission operation and maintenance database to form a mapping set. The real-time sampling frequency deviation, single sampling duration deviation, single sampling data volume deviation, and data acquisition success rate deviation are then substituted into the mapping set to obtain the weight elements corresponding to the sampling frequency deviation, single sampling duration deviation, single sampling data volume deviation, and data acquisition success rate deviation.

[0029] Furthermore, in this embodiment, multivariate analysis is performed on sampling frequency deviation, single sampling duration deviation, single sampling data volume deviation, and data acquisition success rate deviation. Specifically, the correlation between these parameters is considered. If the sampling frequency deviation is too large, for example, if the actual sampling frequency is higher than the average sampling frequency, then the single sampling duration will be shortened, resulting in an excessively large single sampling duration deviation, which negatively impacts the data dispersion value of the DC transmission line. Similarly, if the sampling frequency deviation causes a change in the actual sampling frequency, under the same data acquisition mode, the single sampling data volume will also change accordingly, increasing the single sampling data volume deviation, which also negatively impacts the data dispersion value of the DC transmission line. Therefore, a suitable sampling frequency can ensure that the data acquisition system has sufficient time to accurately acquire data. If the sampling frequency deviation is too high, the data acquisition equipment may not be able to process the data in time, resulting in data loss or errors, thereby reducing the data acquisition success rate, greatly increasing the data acquisition success rate deviation, and increasing the data dispersion of the DC transmission line.

[0030] Furthermore, before performing a fusion operation on the multimodal data based on the discrete data values ​​in step S102 to generate the fused data, the following steps are also included: The interval classification of the discrete data values ​​is determined by a predefined interval of discrete data values. If the interval classification of the discrete values ​​of the data is the first discrete value interval, then the fused data is corrected; If the interval of the discrete values ​​of the data is classified as a second discrete value interval or a third discrete value interval, then the fusion operation is modified.

[0031] refer to Figure 2 As shown, specifically, the fused data is corrected, including: If the interval of the discrete data value is classified as the second discrete value interval, the minimum value of the second discrete value interval is extracted, and the difference between the discrete data value and the minimum value of the second discrete value interval is processed to generate the discrete data deviation. Based on the discrete data deviation, the time axis span calibration value and the coordinate axis span calibration value are obtained. The initial values ​​of the axis span and coordinate axis span of the fusion operation are collected and added to the time axis span calibration value and coordinate axis span calibration value, respectively, to generate the time axis span correction value and coordinate axis span correction value. If the interval of the data discrete values ​​is classified as a third discrete value interval, the data discrete values ​​are proportionally processed to the third discrete value interval to obtain a first deviation ratio. Based on the first deviation ratio, an interval calibration value is obtained. The initial set of extreme values ​​in the fusion operation is collected and processed with the interval calibration value to obtain the data normalization correction interval.

[0032] The initial values ​​for the time axis span and the coordinate axis span of the multimodal data fusion operation can both be extracted from the fusion log of the multimodal data fusion operation. These initial values ​​are the default values ​​for calibrating the time and spatial dimensions of the multimodal data when the data collection dispersion is low and belongs to the first discrete value range.

[0033] The above matching yields the time axis span calibration value. Specifically, it involves matching the discrete deviation of data acquisition of the DC transmission line with the time axis span calibration value corresponding to each predefined discrete deviation interval of data acquisition, determining the specific interval of the discrete deviation of data acquisition of the DC transmission line, and obtaining the time axis span calibration value corresponding to that interval.

[0034] The above matching yields the coordinate axis span calibration value. Specifically, it involves matching the discrete deviation of the data acquisition of the DC transmission line with the coordinate axis span calibration value corresponding to each predefined discrete deviation interval of the data acquisition, determining the specific interval of the discrete deviation of the data acquisition of the DC transmission line, and obtaining the coordinate axis span calibration value corresponding to that interval.

[0035] The aforementioned time axis span specifically defines the coverage of multimodal data from DC transmission lines in the time dimension, including start and end times. During data alignment, the time axis span is used to determine the alignment range of different datasets in time, ensuring they are compared and analyzed within the same time frame, thus eliminating the impact of time differences on data analysis.

[0036] The aforementioned coordinate axis span specifically defines the spatial coverage of multimodal data from DC transmission lines, including the start and end points of the coordinates. During data alignment, the coordinate axis span determines the alignment range of different datasets in the spatial dimension, ensuring they are compared and analyzed within the same coordinate system. This ensures that spatial data from different data sources are in the same coordinate system, eliminating the impact of coordinate differences on spatial analysis. Aligning data on the coordinate axes helps fuse data from different sensors or devices, improving data integrity and accuracy.

[0037] If the discrete values ​​of the DC transmission line data belong to the third discrete value interval, the time axis span correction value and the coordinate axis span correction value are obtained through analysis. The multimodal data of the DC transmission line are aligned in the time and spatial dimensions, respectively. The discrete values ​​of the DC transmission line data are then processed according to their proportion to the third discrete value interval. Specifically, the difference between the discrete values ​​of the DC transmission line data and the minimum value of the third discrete value interval is processed to obtain the first deviation of the data acquisition discreteness of the DC transmission line. This deviation is then compared with the span of the third discrete value interval, which is the value corresponding to the maximum value minus the minimum value of the third discrete value interval. This yields the first deviation proportion of the data discrete values. The data normalization interval calibration value is then obtained by matching the data. The set of maximum and minimum values ​​of the initial data normalization interval for the multimodal data fusion operation is collected and processed with the data normalization interval calibration value to obtain the data normalization correction interval. This correction interval is used to uniformly map the multimodal data of the DC transmission line to the data normalization correction interval.

[0038] The initial set of extreme values ​​for the data normalization interval in the multimodal data fusion operation can be extracted from the fusion log of the multimodal data fusion operation. Here, the initial set of values ​​is the default interval extreme value for data normalization when the data collection dispersion is low and belongs to the first discrete value interval. The set of extreme values ​​includes the maximum value and the minimum value.

[0039] The above analysis yields the following specific correction values ​​for the time axis and coordinate axis span: Based on the first deviation of the data acquisition discreteness of the DC transmission line, a time axis span calibration value is obtained by matching; the initial value of the time axis span from the multi-modal data fusion operation is added to the time axis span calibration value to obtain the time axis span correction value; and based on the first deviation of the data acquisition discreteness of the DC transmission line, a coordinate axis span calibration value is obtained by matching; the initial value of the coordinate axis span from the multi-modal data fusion operation is added to the coordinate axis span calibration value to obtain the coordinate axis span correction value.

[0040] The above matching yields the data normalization interval calibration value. Specifically, it matches the first deviation percentage of the discrete data value with the data normalization interval calibration value corresponding to each predefined first deviation percentage interval to determine the specific interval of the first deviation percentage of the discrete data value, and obtains the data normalization interval calibration value corresponding to that interval.

[0041] The above-mentioned data normalization interval correction maximum and minimum values ​​are obtained by adding the initial maximum value of the data normalization interval to the data normalization interval calibration value, and subtracting the initial minimum value of the data normalization interval from the data normalization interval calibration value to obtain the data normalization interval correction maximum and minimum values, thus forming the data normalization correction interval.

[0042] It's important to explain that data normalization intervals specifically refer to mapping values ​​from different data sources to a unified numerical range; this range is the data normalization interval. By normalizing these data to a common interval (such as [0,1]), dimensional differences can be eliminated, allowing data from different data sources to be compared and analyzed on the same scale. Multi-data fusion algorithms (such as neural networks and support vector machines) are quite sensitive to the numerical range of the data. If the numerical ranges of data from different data sources differ significantly, it may lead to slow algorithm convergence, long training times, and even affect the accuracy of the algorithm. Data normalization can unify the data to a suitable interval, improving the efficiency and accuracy of the algorithm.

[0043] In this embodiment, the fusion operation is adjusted by different intervals of the discrete values ​​of the DC transmission line data. This enables on-demand adaptation and hierarchical optimization. Adjusting different intervals greatly avoids over-adjustment or under-adjustment of the fusion operation, effectively improving data fusion efficiency and significantly reducing the model evaluation misjudgment rate.

[0044] Furthermore, before performing the fusion operation on the multimodal data to generate the fused data, the following steps are also included: Data cleaning: For data of different modalities, the first step is to remove noise and erroneous data. For example, in image data, it may be necessary to remove abnormal pixels caused by image sensor malfunctions. For text data, it may be necessary to remove meaningless characters, duplicate content, etc.

[0045] Data alignment: Data from different modalities may be inconsistent in time or space. For example, video (image sequences) and audio data may have different frame rates and sampling frequencies. They need to be aligned to the same timeline, or sensor data from spatially different locations (such as cameras and temperature sensors in different locations) need to be transformed to a unified coordinate system to ensure accurate subsequent fusion.

[0046] Data normalization: Data from different modalities typically have different dimensions and numerical ranges. For example, image pixel values ​​may range from 0 to 255, while text vector values ​​may be between -1 and 1. Normalizing these data to the same numerical range (such as [0,1] or [-1,1]) can prevent data from certain modalities from dominating the entire result due to larger values ​​during the fusion process, making the data from different modalities comparable during fusion.

[0047] Multimodal data fusion refers to integrating data from multiple different modalities (such as images, text, audio, and sensor data) from DC transmission lines to obtain more comprehensive and accurate information. Furthermore, step S103, based on the judgment result, performs a fusion operation on the multimodal data to generate fused data, specifically including: If the discrete values ​​of the DC transmission line data belong to the first discrete value interval, then the multimodal data of the DC transmission line is extracted, and the operation is performed according to the various initial values ​​of the multimodal data fusion operation. Specifically, the multimodal data fusion operation performs data alignment according to the initial values ​​of the time axis span and the initial values ​​of the coordinate axis span, and performs data normalization according to the initial interval of data normalization.

[0048] When the differences in parameters such as sampling frequency and duration among various sampling points are relatively small, the original data already possesses a good spatiotemporal alignment foundation, requiring no additional correction. The initial values ​​from the multimodal data fusion operation are directly used for fusion, avoiding complex bias calculations and parameter matching, and reducing CPU / GPU resource consumption.

[0049] If the discrete values ​​of the DC transmission line data belong to the second discrete value interval, then the multimodal data of the DC transmission line is extracted, and a multimodal data fusion operation is performed using the time axis span correction value and the coordinate axis span correction value. Specifically, the multimodal data fusion operation performs data alignment according to the time axis span correction value and the coordinate axis span correction value, and performs data normalization according to the data normalization initial interval.

[0050] For the discrete values ​​of DC transmission line data falling within the second discrete value interval, local asynchrony issues are resolved and the fusion accuracy of multimodal data is improved by correcting the time axis span and coordinate axis span. Adjusting only the spatiotemporal alignment parameters (without modifying the normalization interval) preserves the physical meaning of the data dimensions while resolving the core contradictions of time asynchrony and spatial misalignment, thus balancing the universality and specificity of the multimodal data fusion process.

[0051] If the discrete values ​​of the DC transmission line data belong to the third discrete value interval, then the multimodal data of the DC transmission line is extracted, and a multimodal data fusion operation is performed using the time axis span correction value, the coordinate axis span correction value, and the data normalization correction interval. Specifically, the multimodal data fusion operation performs data alignment according to the time axis span correction value and the coordinate axis span correction value, and performs data normalization according to the data normalization correction interval.

[0052] To address the significant differences in the dispersion of data from DC transmission lines, it is necessary to simultaneously correct the time axis (to address asynchronous sampling), the coordinate axis (to address spatial misalignment), and the normalization interval (to address dimensional explosion), employing multi-dimensional compensation to eliminate the heterogeneity barrier of multimodal data from DC transmission lines. Failure to normalize and correct extreme differences in the data may lead to training bias in the fusion model. Therefore, it is crucial to prevent the fusion failure of multimodal data from DC transmission lines and ensure the robustness of the spatiotemporal feature fusion model.

[0053] The fused data of DC transmission lines is input into the spatiotemporal feature fusion model, and the operational feature data of DC transmission lines are output.

[0054] The aforementioned spatiotemporal feature fusion model is specifically a core algorithm for processing the spatiotemporal correlation of multimodal data. It involves preprocessing and fusing the heterogeneous data (including time-series and spatial distribution characteristics) of DC transmission lines. Specifically, it generates time-dimensional feature vectors by performing LSTM or Transformer encoding on the time-series data of the DC transmission lines' multimodal data. The spatial sequence data of the DC transmission lines' multimodal data is then modeled as a graph, and spatial dependencies between nodes are calculated using GCN to generate spatial-dimensional feature vectors. Based on these time-dimensional and spatial-dimensional feature vectors, spatiotemporal coupling features reflecting the operating status of the DC transmission lines are extracted, ultimately outputting operational feature data for health assessment (such as partial discharge phase asymmetry and cross-sensor correlation coefficients).

[0055] Based on the operational characteristic data of DC transmission lines, assess the operational health indicators of DC transmission lines and determine whether to issue early warnings and conduct maintenance on the operational health of DC transmission lines.

[0056] Specifically, the operational health indicators of DC transmission lines are assessed, and the specific assessment process is as follows: The operational characteristic data of DC transmission lines include the partial discharge phase asymmetry of DC transmission lines, the cross-sensor correlation coefficient of DC transmission lines, the environmental-electrical parameter coupling entropy of DC transmission lines, and the proportion of leakage current harmonics of DC transmission lines. The specific operational characteristic data can be extracted from the output records of the spatiotemporal feature fusion model.

[0057] The partial discharge phase asymmetry, cross-sensor correlation coefficient, environmental-electrical parameter coupling entropy, and leakage current harmonic proportion of the DC transmission line are all normalized after removing units to obtain normalized results. These results are then weighted and aggregated sequentially to obtain the operational health indicators of the DC transmission line. The specific analysis method is as follows: ; In the formula, RH represents the operational health index of the DC transmission line, APD is the normalized value of the partial discharge phase asymmetry of the DC transmission line, ρ is the normalized value of the cross-sensor correlation coefficient of the DC transmission line, H is the normalized value of the environmental-electrical parameter coupling entropy of the DC transmission line, and KHC is the normalized value of the leakage current harmonic ratio of the DC transmission line. For the weight elements corresponding to the predefined partial discharge phase asymmetry in the power transmission operation and maintenance database, These are the weight elements corresponding to the predefined cross-sensor correlation coefficients in the power transmission operation and maintenance database. For the predefined environmental-electrical parameter coupling entropy in the power transmission operation and maintenance database, the weight elements are... The weight element corresponding to the predefined leakage current harmonic proportion in the power transmission operation and maintenance database.

[0058] It should be explained that the aforementioned partial discharge phase asymmetry refers to the ratio of the absolute difference between the number of partial discharges in the positive and negative half-cycles of a DC transmission line within the power frequency cycle to the total number of discharges, reflecting the directional characteristics of the discharge. The power frequency cycle refers to the time required for a DC transmission line to complete one full sine wave change. The cross-sensor correlation coefficient refers to the mean of the correlation coefficients of adjacent data acquisition points in the time series, reflecting the consistency of the distribution of line parameters. The environmental-electrical parameter coupling entropy refers to the joint information entropy of environmental parameters such as humidity and temperature and electrical parameters such as voltage and current, measuring the interdependence of multimodal data. The leakage current harmonic ratio refers to the ratio of the sum of the amplitudes of the 3rd, 5th, and other odd harmonics in the leakage current signal to the amplitude of the fundamental wave, reflecting the degree of contamination on the surface of the DC transmission line insulator and the development of discharge. The fundamental wave amplitude refers to the maximum instantaneous value of the fundamental wave in a periodic signal (such as a sine wave, square wave, etc.), i.e., the lowest frequency sine wave component.

[0059] The weight elements corresponding to partial discharge phase asymmetry, cross-sensor correlation coefficient, environmental-electrical parameter coupling entropy, and leakage current harmonic proportion are all extracted from the power transmission operation and maintenance database. The mapping relationship can be one-to-one or many-to-one. For example, partial discharge phase asymmetry, cross-sensor correlation coefficient, environmental-electrical parameter coupling entropy, and leakage current harmonic proportion are mapped to the preset weight elements corresponding to partial discharge phase asymmetry, cross-sensor correlation coefficient, environmental-electrical parameter coupling entropy, and leakage current harmonic proportion in the power transmission operation and maintenance database to form a mapping set. The real-time partial discharge phase asymmetry, cross-sensor correlation coefficient, environmental-electrical parameter coupling entropy, and leakage current harmonic proportion are then substituted into the mapping set to obtain the weight elements corresponding to partial discharge phase asymmetry, cross-sensor correlation coefficient, environmental-electrical parameter coupling entropy, and leakage current harmonic proportion.

[0060] In this embodiment, multivariate analysis is performed using partial discharge phase asymmetry, cross-sensor correlation coefficient, environmental-electrical parameter coupling entropy, and leakage current harmonic proportion. Specifically, the correlation between these parameters is considered. When the partial discharge phase asymmetry is high, it means that the partial discharge of the DC transmission line is unevenly distributed within the power frequency cycle. This uneven discharge may interfere with the surrounding electric and magnetic fields, thereby affecting the measurement of electrical parameters by other sensors, reducing the cross-sensor correlation coefficient, and negatively impacting the operational health of the DC transmission line. When the environmental-electrical parameter coupling entropy increases, it means that the environment has changed, and a water film may form on the surface of the DC transmission line insulation material, reducing its insulation performance and making partial discharge more likely to occur. This leads to a change in the partial discharge phase asymmetry, which in turn greatly reduces the operational health of the DC transmission line. An increase in the proportion of leakage current harmonics indicates an increase in harmonic components in the current. These harmonics will interfere with the electrical parameters measured by other sensors, leading to a decrease in the cross-sensor correlation coefficient, which will also reduce the operational health of the DC transmission line.

[0061] Specifically, in step S105, before performing operation and maintenance on the DC transmission line based on the aforementioned operational health indicators, the following steps are also included: The operational health indicators are compared with predefined operational health indicators at thresholds; Based on the threshold comparison results, the necessity of maintaining the DC transmission line is determined.

[0062] In some embodiments, the operational health indicator thresholds include a first operational health indicator threshold, a second operational health indicator threshold, and operational health definition indicators.

[0063] If the operational health index of a DC transmission line is greater than or equal to the first threshold of operational health index, it is determined that there is no need to issue an early warning or perform maintenance on the operational health of the DC transmission line, and the operational health of the DC transmission line is continuously monitored.

[0064] If the operational health index of a DC transmission line is less than the first threshold of operational health index and greater than or equal to the second threshold of operational health index, it is determined that there is no need to issue an early warning for the operational health of the DC transmission line, but maintenance of the operational health of the DC transmission line is required. Specifically, the operational health index of the DC transmission line is compared with the first threshold of operational health index to obtain the first deviation ratio of the operational health of the DC transmission line. The adaptive electromotive force of the DC transmission line is then matched, and the smoothing reactor is started to increase the initial electromotive force of the DC transmission line to the adaptive electromotive force of the DC transmission line at a preset adjustment rate.

[0065] The above-mentioned percentage processing specifically involves taking the difference between the first threshold of the operational health index and the operational health index of the DC transmission line to obtain the first deviation of the operational health of the DC transmission line, and then taking the ratio with the first threshold of the operational health index to obtain the percentage of the operational health deviation of the DC transmission line.

[0066] The above matching obtains the adaptive electromotive force of the DC transmission line. Specifically, it matches the first deviation percentage of the operating health of the DC transmission line with the adaptive electromotive force corresponding to each predefined first deviation percentage interval of the operating health, determines the specific interval of the first deviation percentage of the operating health of the DC transmission line, and allocates the adaptive electromotive force corresponding to the interval to the DC transmission line to obtain the adaptive electromotive force of the DC transmission line.

[0067] The current in DC transmission lines is easily affected by various factors, causing fluctuations. Smoothing reactors have the characteristic of suppressing current fluctuations. When a smoothing reactor is started and its electromotive force is adjusted, the DC current is smoothed, reducing the ripple component and making the DC current more stable. This is crucial for ensuring the normal operation of equipment in DC transmission systems, reducing the risk of equipment damage due to excessive current fluctuations, and helping to maintain the power balance and voltage stability of the DC transmission system.

[0068] If the operational health index of a DC transmission line is less than the second threshold of operational health index but greater than the operational health definition index, or if the operational health index of a DC transmission line is less than or equal to the operational health definition index, then it is determined that an early warning and maintenance of the operational health of the DC transmission line is required.

[0069] In this embodiment, the process of determining whether to issue an early warning and perform maintenance on the operational health of the DC transmission line is as follows: Figure 3 As shown, Figure 3To establish a flowchart for early warning and maintenance judgment, the following steps are taken: First, the operational health indicators of the DC transmission line are compared with predefined threshold values. When the operational health indicators are greater than or equal to the first threshold, the line is considered to be in good operating condition, requiring no early warning or maintenance; continued monitoring is sufficient. If the operational health indicators are between the first and second thresholds, maintenance is required even without an early warning. In this case, the smoothing reactor is activated, and the initial electromotive force is increased to the appropriate electromotive force at a preset adjustment rate to maintain stable line operation. If the operational health indicators are below the second threshold but above the defined threshold, or directly below the defined threshold, both early warning and maintenance are required. In this case, the converter valve firing angle is adjusted to reduce the transmission voltage. If the operational health indicators remain below the defined threshold, the DC circuit breaker must be activated to implement an emergency maintenance plan to ensure the safe and stable operation of the line.

[0070] Specifically, determining the operational health of DC transmission lines requires early warning and maintenance. The specific analysis process is as follows: If the operational health index of a DC transmission line is less than the second threshold of operational health index but greater than the operational health definition index, an early warning is issued for the operational health of the DC transmission line. Simultaneously, maintenance is carried out on the operational health of the DC transmission line. Specifically, the operational health index of the DC transmission line is proportionally processed with the second threshold of operational health index to obtain the second deviation ratio of the operational health of the DC transmission line. This ratio is then matched to obtain the firing angle adjustment amount of the converter valve. The initial firing angle of the converter valve (which can be extracted from the converter valve's usage report) is collected and added to the firing angle adjustment amount of the converter valve to obtain the corrected firing angle of the converter valve. The difference between the operational health index of the DC transmission line and the operational health definition index is processed to obtain the operational health deviation of the DC transmission line. The firing angle adjustment rate is then matched to obtain the firing angle adjustment rate, which is used to increase the initial firing angle of the converter valve to the corrected firing angle of the converter valve, thereby reducing the transmission voltage of the DC transmission line converter valve.

[0071] It should be explained that the above-mentioned converter valve refers to the DC output terminal of the converter valve, that is, the DC terminal of the converter valve.

[0072] The above-mentioned percentage processing specifically involves processing the difference between the second threshold of the operational health index and the operational health index of the DC transmission line to obtain the second deviation of the operational health of the DC transmission line, and then performing ratio processing with the second threshold of the operational health index to obtain the percentage of the operational health deviation of the DC transmission line.

[0073] The above matching yields the firing angle adjustment amount of the converter valve. Specifically, it involves matching the second deviation ratio of the operating health of the DC transmission line with the firing angle adjustment amount corresponding to each predefined second deviation ratio interval of the operating health, determining the specific interval of the second deviation ratio of the operating health of the DC transmission line, and allocating the firing angle adjustment amount corresponding to this interval to the converter valve to obtain the firing angle adjustment amount of the converter valve.

[0074] The above matching yields the firing angle adjustment rate, specifically by matching the operational health deviation of the DC transmission line with the firing angle adjustment rate corresponding to each predefined operational health deviation interval, determining the specific interval of the operational health deviation of the DC transmission line, and obtaining the firing angle adjustment rate corresponding to that interval.

[0075] In this embodiment, the adjustment step size of the firing angle is limited by the firing angle adjustment rate to avoid drastic fluctuations in DC voltage / power caused by sudden changes in the firing angle. At the same time, the firing angle adjustment rate is matched with the frequency regulation characteristics of the AC power grid (such as the inertial time constant of a 50Hz system of about 0.1-0.5s). The adjustment rate matches the dynamic response of the power grid, avoiding resonance between the firing angle adjustment and the frequency fluctuation of the power grid, reducing the risk of oscillation in the DC transmission system, and ensuring the stability of the AC side voltage.

[0076] The change in the firing angle of the converter valve directly affects the valve's conduction time and sequence, thereby regulating DC voltage and power. When the operating state of the DC transmission system changes, such as load fluctuations, adjusting the firing angle can ensure that the DC voltage and power meet actual needs, guaranteeing the normal operation of the system. Properly adjusting the firing angle can reduce harmonics generated during the converter process. Harmonics can interfere with other equipment in the power grid, affecting power quality. By optimizing the firing angle, harmonic content can be reduced, power quality improved, and other electrical equipment protected from harmonic damage.

[0077] If the operational health indicators of a DC transmission line are less than or equal to the operational health definition indicators, an early warning will be issued for the operational health of the DC transmission line, and the DC circuit breaker will be activated. At the same time, the duration of the operational health warning will be monitored. If the duration of the operational health warning is less than or equal to the duration defined for the operational health warning, the DC circuit breaker will perform pre-disconnection. If the duration of the operational health warning is longer than the duration defined for the operational health warning, the DC circuit breaker will execute the emergency operation and maintenance plan.

[0078] The aforementioned duration of the operational health warning specifically refers to the time from when the system detects the start of the operational health warning signal until the operational health problem is resolved and the warning signal ends.

[0079] The aforementioned DC circuit breaker implements an emergency operation and maintenance plan, specifically by using a magnetic field or air blowing technology to quickly extinguish the arc in the arc-extinguishing chamber of the DC circuit breaker to prevent overvoltage; disconnecting the isolating switches at both ends of the faulty DC transmission line section to form physical isolation, and sending a signal to the converter station to reduce the firing angle of the converter valve on the faulty side to the defined firing angle of the converter valve, quickly suppressing the rate of rise of the fault current; and simultaneously activating the temporary grounding switch of the faulty DC transmission line section to release residual charge and prevent back EMF hazards.

[0080] Furthermore, reducing the transmission voltage of the converter valve in a DC transmission line also includes: Based on the firing angle adjustment of the converter valve, the estimated monitoring cycle of the converter valve is obtained. Under the estimated monitoring cycle, the AC input terminal of the converter valve is monitored in advance, the operating data of the AC terminal of the converter valve is collected, the operating stability value of the AC terminal of the converter valve is determined, the estimated reactive power of the AC terminal of the converter valve is obtained, and the real-time reactive power of the AC terminal of the converter valve is collected. The real-time reactive power can be monitored by a power sensor.

[0081] The above matching yields the estimated monitoring cycle of the converter valve. Specifically, the firing angle adjustment of the converter valve is matched with the estimated monitoring cycle corresponding to each predefined firing angle adjustment interval to determine the specific interval of the firing angle adjustment of the converter valve and obtain the estimated monitoring cycle corresponding to that interval.

[0082] The aforementioned predictive monitoring refers to the collection and predictive analysis of operating data at the AC input terminal of the converter valve during the predicted monitoring period. Real-time monitoring of the AC input terminal allows for the acquisition of operating status information for the AC side of the converter valve. The operating status of the converter valve directly affects the stability of the AC side voltage. Predictive monitoring of the AC input terminal of the converter valve allows for real-time monitoring of parameters such as voltage angular velocity fluctuation rate. Once voltage fluctuations are detected, synchronous adjustment of reactive power can promptly compensate for system reactive power deficits or absorb excess reactive power, preventing damage to equipment from large voltage fluctuations, enhancing the system's anti-interference capability under complex operating conditions, and ensuring the continuous and stable operation of the DC transmission system.

[0083] Based on the firing angle adjustment rate, the power adjustment rate is matched and obtained. While increasing the initial firing angle of the converter valve to the corrected firing angle of the converter valve at the firing angle adjustment rate, the power supplementation device is simultaneously started to increase the real-time reactive power of the AC end of the converter valve to the estimated reactive power of the AC end of the converter valve at the power adjustment rate.

[0084] The above matching yields the power regulation rate, specifically by matching the firing angle regulation rate with the power regulation rate corresponding to each predefined firing angle regulation rate interval, determining the specific interval of the firing angle regulation rate, and obtaining the power regulation rate corresponding to that interval.

[0085] When the firing angle of the converter valve increases, its conduction changes, affecting the reactive power demand on the AC side. If the reactive power is not adjusted accordingly, it may lead to AC voltage instability, impacting the overall system stability. Synchronously activating the power supplementation device to increase the real-time reactive power to the estimated reactive power maintains AC voltage stability and ensures stable system operation. Proper reactive power adjustment helps protect the converter valve and other related equipment. Insufficient reactive power may subject the converter valve to excessive voltage and current stress during operation, shortening equipment lifespan or even causing damage. Synchronous adjustment to ensure reactive power meets demand reduces the risk of equipment failure and improves equipment reliability and lifespan.

[0086] Specifically, the determination of the operational stability value of the AC terminal of the converter valve is carried out through the following process: The operating data of the AC terminal of the converter valve includes the control signal-current phase difference entropy, the voltage angular velocity fluctuation rate, and the overvoltage rise rate of the AC terminal. The specific operating data of the AC terminal of the converter valve can be extracted from the user report of the converter valve.

[0087] The control signal-current phase difference entropy, voltage angular velocity fluctuation rate, and overvoltage rise rate at the AC terminal of the converter valve are normalized by removing units, and then weighted and aggregated to obtain the operating stability value of the AC terminal of the converter valve. The specific analysis is as follows: ; In the formula, ST represents the operational stability value of the AC terminal of the converter valve, HΦ is the normalized value of the phase difference entropy between the control signal and the current at the AC terminal of the converter valve, VAR is the normalized value of the voltage angular velocity fluctuation rate at the AC terminal of the converter valve, and UX is the normalized value of the overvoltage rise rate at the AC terminal of the converter valve. The weight element corresponding to the predefined control signal-current phase difference entropy in the power transmission operation and maintenance database. For the predefined weight elements corresponding to voltage angular velocity fluctuations in the power transmission operation and maintenance database, The weight element corresponding to the predefined overvoltage rise rate in the power transmission operation and maintenance database.

[0088] It should be explained that the aforementioned control signal-current phase difference entropy refers to the disorder of the phase difference distribution between the converter valve trigger pulse signal and the actual AC current, reflecting the matching between the converter valve's own control strategy and the valve group's dynamic response; the voltage angular velocity fluctuation rate refers to the degree of fluctuation of the rotational angular velocity of the composite vector of the three-phase voltages at the AC end of the converter valve within the fundamental period, reflecting the symmetry of the grid voltage and the synchronization of the converter valve triggering; the overvoltage rise rate refers to the degree of change of the transient overvoltage at the AC end when the converter valve is turned on / off.

[0089] The weight elements corresponding to the control signal-current phase difference entropy, the voltage angular velocity fluctuation rate, and the overvoltage rise rate are all extracted from the power transmission operation and maintenance database. The mapping relationship can be one-to-one or many-to-one. For example, the control signal-current phase difference entropy, voltage angular velocity fluctuation rate, and overvoltage rise rate are respectively mapped to the weight elements corresponding to the control signal-current phase difference entropy, voltage angular velocity fluctuation rate, and overvoltage rise rate preset in the power transmission operation and maintenance database to form a mapping set. The real-time control signal-current phase difference entropy, voltage angular velocity fluctuation rate, and overvoltage rise rate are then substituted into the mapping set to obtain the weight elements corresponding to the control signal-current phase difference entropy, voltage angular velocity fluctuation rate, and overvoltage rise rate.

[0090] In this embodiment, multivariate analysis of the control signal-current phase difference entropy, voltage angular velocity fluctuation rate, and overvoltage rise rate is used to specifically consider the correlation between these parameters. When the control signal-current phase difference entropy is high, it means that the phase matching between the trigger pulse signal and the actual AC current is poor, and the triggering time of the converter valve is inaccurate. This will cause deviations in the timing of the converter valve's opening and closing, which in turn increases the fluctuation of the rotational angular velocity of the three-phase voltage composite vector at the AC terminal of the converter valve within the fundamental period, i.e., the voltage angular velocity fluctuation rate increases. The deviation in the timing of the converter valve's opening and closing will also generate a large transient overvoltage at the AC terminal of the converter valve, increasing the overvoltage rise rate, and thus greatly reducing the operational stability of the AC terminal of the converter valve.

[0091] This invention determines the operational stability of the AC terminal of a converter valve by collecting operational data from the AC terminal. The converter valve is a key component of a DC transmission system, and the operational stability of its AC terminal directly affects the performance of the entire system. By collecting operational data and calculating the operational stability value, the operating status of the AC terminal of the converter valve can be monitored in real time. If the stability value is low, it indicates that the converter valve may have problems such as inaccurate triggering timing, large voltage fluctuations, or a high risk of overvoltage, requiring timely adjustment and maintenance to ensure the safe and stable operation of the converter valve. When adjusting the converter valve's trigger angle to reduce the transmission voltage, the estimated reactive power is obtained based on the operational stability value of the AC terminal of the converter valve, and a power compensation device is simultaneously activated to adjust the real-time reactive power. This ensures that the reactive power at the AC terminal of the converter valve is reasonably compensated under different operating conditions, maintaining AC side voltage stability, improving the system's power factor, reducing the impact of reactive power on the system, and thus improving the overall operating efficiency and stability of the DC transmission system.

[0092] refer to Figure 4As shown in the embodiment of this application, a DC transmission line operation and maintenance system is also provided, characterized in that the system includes: The data acquisition module is used to acquire feature datasets and multimodal data in DC transmission lines, and to calculate the discrete values ​​of the feature datasets. The data fusion module is used to perform a fusion operation on the discrete data values ​​of the multimodal data to generate fused data. The feature extraction module is used to input the fused data into the spatiotemporal feature fusion model and output the running feature data; The operation and maintenance execution module is used to generate operation and health indicators for DC transmission lines based on the operation characteristic data, and to perform operation and maintenance on DC transmission lines according to the operation and health indicators.

[0093] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0094] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0095] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various concepts, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to a determination” as used herein may be interpreted as “when…” or “when…” or “in response to a determination.”

[0096] As used in this application, the terms "at least one", "multiple", "each", "any", etc., "at least one" includes one, two or more, "multiple" includes two or more, "each" refers to each of the corresponding multiples, and "any" refers to any one of the multiples.

[0097] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0098] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0099] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0100] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0101] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0102] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application 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 this application 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 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.

[0103] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0104] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0105] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0106] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0107] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0108] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for the operation and maintenance of a DC transmission line, characterized in that, The method includes: Collect feature datasets and multimodal data from DC transmission lines, and calculate the discrete values ​​of the feature datasets; Based on the discrete values ​​of the data, a fusion operation is performed on the multimodal data to generate fused data; The fused data is input into the spatiotemporal feature fusion model, and the running feature data is output. Based on the operational characteristic data, operational health indicators for DC transmission lines are generated, and the DC transmission lines are maintained and operated according to these operational health indicators.

2. The DC transmission line operation and maintenance method according to claim 1, characterized in that, The calculation of the discrete values ​​of the feature dataset specifically includes: The feature dataset is averaged to generate a feature comparison dataset, where the feature comparison dataset values ​​include the collection frequency, collection duration, collection data volume, and collection success rate. Calculate the degree of deviation between the feature dataset and the feature comparison dataset, perform weighted aggregation on the degree of deviation, and generate discrete data values.

3. The DC transmission line operation and maintenance method according to claim 1, characterized in that, Before performing the fusion operation on the multimodal data based on the discrete data values ​​to generate the fused data, the process further includes: The interval classification of the discrete data values ​​is determined by a predefined interval of discrete data values. If the interval of the discrete data values ​​is classified as the first discrete value interval, then no correction is needed for the fusion operation; If the interval of the discrete values ​​of the data is classified as a second discrete value interval or a third discrete value interval, then the fusion operation is modified.

4. The DC transmission line operation and maintenance method according to claim 3, characterized in that, If the interval of the discrete data values ​​is classified as a second discrete value interval or a third discrete value interval, the fusion operation is modified, specifically including: If the interval of the discrete data value is classified as the second discrete value interval, the minimum value of the second discrete value interval is extracted, and the difference between the discrete data value and the minimum value of the second discrete value interval is processed to generate the discrete data deviation. Based on the discrete data deviation, the time axis span calibration value and the coordinate axis span calibration value are obtained. The initial values ​​of the axis span and coordinate axis span of the fusion operation are collected and added to the time axis span calibration value and coordinate axis span calibration value, respectively, to generate the time axis span correction value and coordinate axis span correction value. If the interval of the data discrete values ​​is classified as a third discrete value interval, the data discrete values ​​are proportionally processed to the third discrete value interval to obtain a first deviation ratio. Based on the first deviation ratio, an interval calibration value is obtained. The initial set of extreme values ​​of the fusion data operation is collected and processed with the interval calibration value to obtain the data normalization correction interval.

5. The DC transmission line operation and maintenance method according to claim 4, characterized in that, The step of performing a fusion operation on the multimodal data based on the discrete values ​​of the data to generate fused data specifically includes: If the interval of the discrete values ​​of the data is classified as the first discrete value interval, then the multimodal data of the DC transmission line is extracted, and the fusion operation is performed according to the various initial values ​​of the multimodal data fusion operation to generate fused data; If the interval of the discrete values ​​of the data is classified as the second discrete value interval, then the multimodal data of the DC transmission line is extracted, and a multimodal data fusion operation is performed with the time axis span correction value and the coordinate axis span correction value to generate fused data; If the interval of the discrete data value is classified as the third discrete value interval, then the multimodal data of the DC transmission line is extracted, and a multimodal data fusion operation is performed using the time axis span correction value, the coordinate axis span correction value, and the data normalization correction interval.

6. The DC transmission line operation and maintenance method according to claim 1, characterized in that, The generation of operational health indicators for DC transmission lines based on the operational characteristic data specifically includes: The operational characteristic data are normalized and weighted, and then aggregated to generate operational health indicators for DC transmission lines.

7. The DC transmission line operation and maintenance method according to claim 6, characterized in that, Before performing operation and maintenance on the DC transmission line based on the aforementioned operational health indicators, the following steps are also included: The operational health indicators are compared with predefined operational health indicators at thresholds; Based on the threshold comparison results, the necessity of maintaining the DC transmission line is determined.

8. The DC transmission line operation and maintenance method according to claim 6, characterized in that, The operational characteristic data include the partial discharge phase asymmetry of the DC transmission line, the cross-sensor correlation coefficient of the DC transmission line, the coupling entropy of the environmental and electrical parameters of the DC transmission line, and the proportion of leakage current harmonics of the DC transmission line.

9. A DC transmission line operation and maintenance system, characterized in that, The system includes: The data acquisition module is used to acquire feature datasets and multimodal data in DC transmission lines, and to calculate the discrete values ​​of the feature datasets. The data fusion module is used to perform a fusion operation on the multimodal data based on the discrete values ​​of the data to generate fused data; The feature extraction module is used to input the fused data into the spatiotemporal feature fusion model and output the running feature data; The operation and maintenance execution module is used to generate operation and health indicators for DC transmission lines based on the operation characteristic data, and to perform operation and maintenance on DC transmission lines according to the operation and health indicators.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 8.

Images