Direct current transmission data processing method, system and equipment based on multi-modal data fusion

Abstract

The invention discloses a direct current transmission data processing method, system and device based on multi-modal data fusion, and belongs to the technical field of direct current transmission data testing. The method comprises the following steps of performing quantitative judgment on data timeliness based on timeliness evaluation parameters of a long-distance direct-current cable data sensor to obtain a data acquisition timeliness judgment result; performing dynamic optimization on each data sensor based on a data acquisition timeliness judgment result, and performing cable data acquisition through the dynamically optimized data sensors; performing quantitative judgment on the data timeliness influence based on the timeliness influence parameter of the long-distance direct-current cable data fusion to obtain a data fusion timeliness influence judgment result; and performing dynamic optimization on each cable data based on a data fusion timeliness influence judgment result. According to the invention, data acquisition and data transmission timeliness are dynamically processed, so that high-timeliness fusion of multi-modal data of the long-distance direct-current cable is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of direct current transmission data testing, and in particular to a direct current transmission data processing method, system and device based on multimodal data fusion. Background Art

[0002] The application of multimodal data fusion in DC transmission data processing is primarily aimed at improving the monitoring, optimization, and forecasting capabilities of transmission systems. By combining data from diverse sources, a more comprehensive understanding of the DC transmission system's operating status and potential risks can be achieved. As modern DC transmission technology continues to evolve, data fusion technology provides strong support for the integration of smart grids and renewable energy.

[0003] Existing systems typically integrate data from multiple sensors and monitoring devices, using advanced data processing and fusion algorithms to monitor and analyze DC transmission systems in real time. Through data preprocessing, feature extraction, and fusion, they synthesize information from diverse data sources to identify potential faults, optimize operating parameters, and provide a basis for decision support.

[0004] For example, the invention patent announcement with announcement number: CN109270407B discloses a method for identifying the cause of a UHVDC transmission line fault based on multi-source information fusion, which includes: obtaining electrical quantity fault data and non-electrical quantity information of the UHVDC transmission line; extracting electrical characteristic quantities of the UHVDC transmission line and constructing an electrical characteristic input vector; extracting non-electrical characteristic quantities of the UHVDC transmission line and constructing a non-electrical characteristic input vector; constructing comprehensive neural network identification models for lightning strikes, wildfires, pollution, wind deviation, and bird damage, respectively, and selecting the number of hidden layers of the neural network model using the minimum error method; and using a self-learning method to identify specific fault causes.

[0005] For example, the invention patent with announcement number CN101915888B discloses a method for the extension fusion identification of lightning interference on ±800kV DC transmission lines. The method includes: extracting the characteristics of the time domain waveform by simulating electromagnetic transients under lightning faults, lightning interference, and non-lightning faults. For voltage sampling values ​​with a sampling rate of 10kHz and a time window of 5ms, the first 2ms of data are directly correlated, and the last 3ms of data are directly averaged. Finally, the correlation calculation results and the average calculation results are extensionally fused according to their respective weight coefficients, and their correlation function values ​​are calculated, forming a lightning interference identification criterion directly based on the sampling values. When the calculation result of the correlation function is greater than or equal to zero, it is identified as lightning interference and the protection is restored; when the calculation result of the correlation function is less than zero, it is identified as a line fault.

[0006] However, in the process of implementing the technical solutions of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems:

[0007] In the existing technology, when fusing the multimodal data of long-distance DC cables, the multimodal data of the DC cables needs to jump through relay stations multiple times, and the congestion levels of different relay stations vary, resulting in a chaotic timing of the multimodal data arriving at the fusion node. This leads to the problem of low timeliness of DC transmission data processing in multimodal data fusion. Summary of the Invention

[0008] The present invention provides a method, system and device for processing DC transmission data based on multimodal data fusion, thereby solving the problem in the prior art of low timeliness of DC transmission data processing due to the low timeliness of multimodal data fusion, caused by the need for multiple jumps of the multimodal data of the DC cable to the fusion node when fusing the multimodal data of the DC cable. This problem is solved.

[0009] The present invention provides a DC transmission data processing method based on multimodal data fusion, comprising the following steps: performing quantitative determination of data timeliness based on timeliness evaluation parameters of long-distance DC cable data sensors to obtain a data acquisition timeliness determination result; dynamically optimizing each data sensor based on the data acquisition timeliness determination result, and performing cable data acquisition through the dynamically optimized data sensor, wherein dynamic optimization of each data sensor means optimizing each data sensor based on the timeliness evaluation parameters of the long-distance DC cable data sensor to reduce the influence of sensor performance on the timeliness of DC cable data acquisition; performing quantitative determination of the influence of data timeliness based on timeliness influence parameters of long-distance DC cable data fusion to obtain a data fusion timeliness influence determination result; dynamically optimizing each cable data based on the data fusion timeliness influence determination result, wherein dynamic optimization of each cable data means optimizing the transmission state of a relay station based on the timeliness influence parameters of the long-distance DC cable data fusion to reduce the influence of the relay station state on the timeliness of DC cable data transmission.

[0010] The present invention provides a system for applying a DC transmission data processing method based on multimodal data fusion, comprising: an acquisition and determination module, an acquisition optimization module, a fusion determination module, a fusion optimization module and a DC transmission database; wherein the acquisition and determination module is used to perform a quantitative determination of data timeliness based on the timeliness evaluation parameters of the long-distance DC cable data sensor, and obtain a data acquisition timeliness determination result; the acquisition optimization module is used to dynamically optimize each data sensor based on the data acquisition timeliness determination result, and perform cable data acquisition through the dynamically optimized data sensor, and the dynamic optimization of each data sensor is expressed according to the long-distance DC cable data. According to the timeliness evaluation parameters of the sensor, each data sensor is optimized to reduce the impact of sensor performance on the timeliness of DC cable data acquisition; a fusion judgment module is used to quantitatively judge the impact of data timeliness based on the timeliness impact parameters of long-distance DC cable data fusion, and obtain the data fusion timeliness impact judgment result; a fusion optimization module is used to dynamically optimize each cable data based on the data fusion timeliness impact judgment result. Dynamic optimization of each cable data means optimizing the relay station transmission status according to the timeliness impact parameters of long-distance DC cable data fusion to reduce the impact of the relay station status on the timeliness of DC cable data transmission.

[0011] An embodiment of the present application further provides a DC power transmission data processing device for multimodal data fusion, comprising: a processor and a memory storing instructions executable by the processor; when the processor is configured to execute the instructions, the electronic device implements a DC power transmission data processing method based on multimodal data fusion.

[0012] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0013] 1. The present invention provides a DC transmission data processing method, system, and device based on multimodal data fusion, thereby quantitatively determining the timeliness evaluation parameters of long-distance DC cable data sensors and the parameters affecting data fusion timeliness. Based on the determination results, the sensor and data transmission processes are dynamically optimized, effectively reducing the negative impact of sensor performance and relay station status on the timeliness of data acquisition and transmission. This improves the timeliness of DC cable multimodal data fusion, ensures the real-time and reliability of data processing, and optimizes the efficiency of the entire data acquisition and transmission process.

[0014] 2. The present invention compares the timeliness evaluation parameters of data sensors with preset thresholds, and dynamically optimizes the data acquisition process according to the status and timeliness requirements of different sensors, thereby adjusting the sampling frequency and dynamic sensor data buffer management, while ensuring that data sensors are processed in a timely manner, avoiding the timeliness of data fusion affected by data lag, thereby optimizing data acquisition timeliness and improving the real-time performance and processing efficiency of DC cable multimodal data fusion.

[0015] 3. The present invention compares the fusion timeliness influencing parameters of each cable data with the threshold, and dynamically optimizes the data transportation process according to the threshold comparison result, thereby optimizing the resource allocation of relay stations through which unqualified data passes, ensuring the timeliness of data transmission, and further realizing the precise optimization and real-time adjustment of the cable data transmission process. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 Flowchart of a method for processing DC power transmission data based on multimodal data fusion provided in an embodiment of the present application.

[0017] Figure 2 A flowchart illustrating the optimization of data collection timeliness provided in an embodiment of the present application.

[0018] Figure 3 A flowchart illustrating the timeliness optimization of data fusion provided in an embodiment of the present application.

[0019] Figure 4 Schematic diagram of a DC power transmission data processing system with multimodal data fusion provided in an embodiment of the present application. DETAILED DESCRIPTION

[0020] The embodiments of the present application provide a method, system and device for processing DC transmission data based on multimodal data fusion, which solves the problem in the prior art that when fusing multimodal data of long-distance DC cables, the multimodal data of the DC cable needs to jump through relay stations multiple times, and the congestion levels of different relay stations are different, resulting in a disordered timing of the multimodal data arriving at the fusion node, and the problem of low timeliness of DC transmission data processing of multimodal data fusion. The method quantitatively determines the timeliness of data based on the timeliness evaluation parameters of the long-distance DC cable data sensor to obtain the data acquisition timeliness determination result; dynamically optimizes each data sensor based on the data acquisition timeliness determination result, and performs cable data acquisition through the dynamically optimized data sensor. The dynamic optimization of each data sensor means that each data sensor is optimized according to the timeliness evaluation parameters of the long-distance DC cable data sensor to reduce the impact of sensor performance on the timeliness of DC cable data acquisition; the timeliness impact of data is quantitatively determined based on the timeliness impact parameters of long-distance DC cable data fusion, and the data fusion timeliness impact determination result is obtained; based on the data fusion timeliness impact determination result, each cable data is dynamically optimized, and the dynamic optimization of each cable data means that the relay station transmission status is optimized according to the timeliness impact parameters of long-distance DC cable data fusion to reduce the impact of the relay station status on the timeliness of DC cable data transmission, thereby reducing the impact on the timeliness of multi-modal DC cable data fusion.

[0021] The technical solution in the embodiments of the present application is to solve the problem of low timeliness of DC power transmission data processing after multimodal data fusion when fusing multimodal data of long-distance DC cables. Because the multimodal data of the DC cable needs to jump through relay stations multiple times, and the congestion levels of different relay stations vary, the timing of the multimodal data arriving at the fusion node is disordered. The overall concept is as follows:

[0022] By comprehensively evaluating and optimizing the timeliness of long-distance DC cable data sensors and the data fusion process, the timeliness of data acquisition is first quantitatively determined based on the timeliness evaluation parameters of the data sensors, and each sensor is dynamically optimized based on the determination results, thereby improving the timeliness of data acquisition. Then, based on the timeliness influencing parameters of data fusion, the timeliness impact of the data fusion process is quantitatively evaluated, and the cable data is dynamically optimized. The transmission status of the relay station is optimized to reduce the negative impact of the relay station status on the timeliness of DC cable data transmission, thus achieving high-timeliness fusion of long-distance DC cable multimodal data.

[0023] In order to better understand the above technical solution, the above technical solution will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0024] like Figure 1As shown, a flow chart of a DC transmission data processing method based on multimodal data fusion provided by an embodiment of the present application is provided. The method includes the following steps: performing a quantitative determination of data timeliness based on a timeliness evaluation parameter of a long-distance DC cable data sensor to obtain a data acquisition timeliness determination result; dynamically optimizing each data sensor based on the data acquisition timeliness determination result, and performing cable data acquisition using the dynamically optimized data sensor. Dynamically optimizing each data sensor means optimizing each data sensor based on the timeliness evaluation parameter of the long-distance DC cable data sensor to reduce the impact of sensor performance on the timeliness of DC cable data acquisition; quantitatively determining the impact of data timeliness based on the timeliness influence parameter of long-distance DC cable data fusion to obtain a data fusion timeliness influence determination result; dynamically optimizing each cable data based on the data fusion timeliness influence determination result. Dynamically optimizing each cable data means optimizing the transmission status of the relay station based on the timeliness influence parameter of the long-distance DC cable data fusion to reduce the impact of the relay station status on the timeliness of DC cable data transmission, thereby reducing the impact on the timeliness of multimodal DC cable data fusion.

[0025] In this embodiment, cable data refers to multi-dimensional physical quantity data collected by various sensors and monitoring equipment to reflect the operating status of the DC cable, including but not limited to cable body status data such as voltage and partial discharge, environmental and auxiliary equipment data such as wind speed and step voltage, etc. By dynamically optimizing various timeliness parameters in the process of long-distance DC cable data acquisition and fusion, the negative impact of relay station congestion and sensor performance on data timeliness can be effectively reduced. By quantitatively evaluating the timeliness of data acquisition and data fusion and optimizing each link, it can be ensured that the timing confusion problem of multi-modal DC cable data is alleviated when it reaches the fusion node, thereby improving the real-time and accuracy of DC transmission data processing. This optimization process improves the overall performance of the system, making the data transmission and processing of long-distance DC cables more efficient and timely.

[0026] In addition, the DC transmission database is used to store relevant data of the DC transmission data processing method based on multimodal data fusion, including: reference sampling frequency, allowable deviation sampling frequency, critical response time, critical clock synchronization error, first threshold for timeliness assessment, second threshold for timeliness assessment and sensor status threshold, etc. The data in the DC transmission database can be obtained through cooperation with power equipment manufacturers or power grid companies and other departments, or can be directly queried through industry-related databases such as the Wukong dataset and the Muge evaluation benchmark.

[0027] Furthermore, the timeliness of data is quantitatively determined based on the timeliness evaluation parameters of the long-distance DC cable data sensor, and the steps for obtaining the data acquisition timeliness determination result include: first, quantifying the influence of the timeliness performance of the data sensor on the timeliness of DC cable data acquisition according to the timeliness evaluation parameters of each data sensor, and obtaining the timeliness evaluation parameters of each data sensor, the timeliness evaluation parameters including sampling frequency, response time, clock synchronization error and sensor state parameters, the timeliness evaluation parameters representing the quantitative data of the degree of influence of the timeliness evaluation parameters on the timeliness of DC cable data acquisition; quantifying the influence of the data sensor state on the timeliness of DC cable data acquisition according to the sensor state parameters, and obtaining the sensor state parameters, the sensor state parameters including output impedance, insulation resistance and voltage drift, the sensor state parameters representing the quantitative data of the degree of influence of the sensor state parameters on the working state of the sensor.

[0028] Among them, the step of quantifying the influence of the data sensor state on the timeliness of DC cable data acquisition according to the sensor state parameters and obtaining the sensor state parameters includes: obtaining sensor state parameter reference data from a preset DC transmission database, specifically including: critical output impedance, critical insulation resistance and critical voltage drift; performing a proportion approach calculation on the output impedance, critical insulation resistance and voltage drift of each data sensor with the critical output impedance, insulation resistance and critical voltage drift respectively, to obtain a proportion approach calculation result; using the sensor state parameter weight ratio to weight the proportion approach calculation results respectively, to obtain a weighted processing result, the sensor state parameter weight ratio includes the output impedance weight ratio, the insulation resistance weight ratio and the voltage drift weight ratio; coupling the weighted processing result and then performing an inverse proportional operation to obtain the sensor state parameter of each data sensor.

[0029] The sensor state parameters of each data sensor are obtained as follows:

[0030]

[0031] Where, SS i represents the sensor state parameter of the i-th data sensor, α4 represents the output impedance weight ratio, α6 represents the insulation resistance weight ratio, α7 represents the voltage drift weight ratio, OI 1i Represents the output impedance of the i-th data sensor. Output impedance is the electrical characteristic of the sensor in the working state. It represents the impedance of the sensor to the current. It can be obtained by measuring the voltage and current relationship between the output terminal of the sensor and the reference terminal using an oscilloscope or impedance analyzer, and using the impedance analyzer. OI0 represents the critical output impedance, IR 1iIndicates the insulation resistance of the i-th data sensor. Insulation resistance refers to the electrical insulation performance between the sensor and the external environment, reflecting the effectiveness of the sensor's electrical insulation during operation. It can be measured using an insulation resistance tester. IR0 represents the critical insulation resistance, and VD 1i = represents the voltage drift of the i-th data sensor, which can be recorded using a precision voltmeter, and VD0 represents the critical voltage drift. The sensor state parameters are all averaged over multiple (e.g., 5) measurements.

[0032] α5, α6, and α7 are the weight ratios corresponding to the output impedance, insulation resistance, and voltage drift preset in the DC transmission database, respectively. These weight ratios are numerical indicators that measure the impact of the above sensor state parameters on the sensor state parameters. Specifically, there is a mapping table for each possible sensor state parameter value, insulation resistance, and voltage drift, which records each possible sensor state parameter value and its corresponding weight ratio. These mapping relationships can be one-to-one or many-to-one. For example, in actual applications, when it is necessary to evaluate the sensor state parameters of a data sensor, the measured output impedance, insulation resistance, and voltage drift can be entered into their respective mapping tables, and the weight ratios corresponding to these values ​​can be quickly found. The weight ratios range from 0 to 1.

[0033] Output impedance, insulation resistance, and voltage drift are the primary factors affecting sensor accuracy and stability, and they are interrelated. For example, low insulation resistance can cause internal current leakage in the sensor, affecting output impedance and voltage drift, leading to data errors. Output impedance can also be related to voltage drift, as high impedance makes the signal susceptible to external interference, which can exacerbate voltage drift. Comprehensive analysis of sensor status parameters can accurately determine whether the sensor is operating normally and whether maintenance or calibration is required, thereby ensuring system data accuracy and stability.

[0034] Then, reference data of timeliness evaluation parameters are obtained from a preset direct current transmission database, specifically including: reference sampling frequency, allowable deviation sampling frequency, critical response time, critical clock synchronization error and critical sensor state parameter; relative deviation conformity calculation is performed on the sampling frequency, reference sampling frequency and allowable deviation sampling frequency of each data sensor to obtain a sampling frequency calculation result; the critical response time, critical clock synchronization error and sensor state parameter of each data sensor are respectively subjected to a proportion approach calculation with the response time, clock synchronization error and critical sensor state parameter to obtain a response time calculation result, a clock synchronization error calculation result and a sensor state parameter calculation result; the sampling frequency calculation result, response time calculation result, clock synchronization error calculation result and sensor state parameter calculation result are respectively weighted using the timeliness evaluation parameter weight ratio, and the weighted processing results are then coupled to obtain the timeliness evaluation parameters of each data sensor, where the timeliness evaluation parameter weight ratio includes the sampling frequency weight ratio, the response time weight ratio, the clock synchronization error weight ratio and the sensor state parameter weight ratio.

[0035] The timeliness evaluation parameters of each data sensor are obtained as follows:

[0036]

[0037] In the formula, TE i represents the timeliness evaluation parameter of the i-th data sensor, α1 represents the sampling frequency weight ratio, α2 represents the response time weight ratio, α3 represents the clock synchronization error weight ratio, α4 represents the sensor state parameter weight ratio, SF 1i Indicates the sampling frequency of the i-th data sensor, which refers to the number of times the data sensor collects data within a single time monitoring point. It can be obtained by the frequency parameters set by the data acquisition system. SF0 represents the reference sampling frequency, SF2 represents the allowable deviation sampling frequency, and RT 1i It represents the response time of the i-th data sensor. The response time refers to the time it takes for the sensor to respond to changes in the input signal. It can be measured through experimental tests to measure the time it takes for the sensor to receive input changes and output stable data. RT0 represents the critical response time, and CE 1i It represents the clock synchronization error of the i-th data sensor, which can be obtained by collecting the clock time of each data sensor at a certain moment and calculating the difference between the clock time of a data sensor and the average clock time of each data sensor at that moment. CE0 represents the critical clock synchronization error, and SS 1i represents the sensor state parameter of the i-th data sensor, SS0 represents the critical sensor state parameter, where i is the number of each data sensor, i=1, 2, 3, ..., N, and N is the total number of data sensors.

[0038] α1, α2, α3, and α4 are the weight ratios corresponding to the sampling frequency, response time, clock synchronization error, and sensor state parameters preset in the DC transmission database. These weight ratios are numerical indicators that measure the impact of these parameters on the timeliness evaluation parameters. Specifically, a mapping table exists for each of the sampling frequency, response time, clock synchronization error, and sensor state parameters, recording each possible parameter value and its corresponding weight ratio. These mappings can be one-to-one or many-to-one. For example, in practical applications, when evaluating the timeliness evaluation parameters of a data sensor, the measured sampling frequency, response time, clock synchronization error, and sensor state parameters can be entered into their respective mapping tables. The corresponding weight ratios can then be quickly found. The weight ratios range from 0 to 1.

[0039] The four are interrelated. For example, sampling frequency and response time directly affect the sensor's reaction speed to input signals and the accuracy of data acquisition. If the response time is longer and the sampling frequency is lower, the sensor may not be able to obtain or accurately reflect the changing signal in a timely manner. Clock synchronization error affects the time alignment of data between multiple sensors. Inconsistent clocks of different sensors can lead to timeliness issues during data fusion. Sensor state parameters reflect the health of the sensor. Smaller sensor state parameters may lead to increased response time, decreased sampling frequency, or increased clock synchronization error, which in turn affects the accuracy and timeliness of the data. The timeliness evaluation parameters obtained through comprehensive analysis can accurately evaluate the timeliness performance of each data sensor. At the same time, when the sensors are in good health, the timeliness and accuracy of data acquisition can be maximized, thereby ensuring the quality of data analysis.

[0040] A first timeliness evaluation threshold, a second timeliness evaluation threshold, and a sensor status threshold are obtained from a preset direct current transmission database; the timeliness evaluation parameter of each data sensor is compared with the first timeliness evaluation threshold and the second timeliness evaluation threshold respectively; if the timeliness evaluation parameter of a data sensor is less than the first timeliness evaluation threshold, the corresponding data acquisition timeliness determination result is recorded as the first acquisition timeliness; if the timeliness evaluation parameter of a data sensor is greater than or equal to the first timeliness evaluation threshold and less than the second timeliness evaluation threshold, the corresponding data acquisition timeliness determination result is recorded as the second acquisition timeliness; if the timeliness evaluation parameter of a data sensor is greater than or equal to the second timeliness evaluation threshold, the corresponding data acquisition timeliness determination result is recorded as the third acquisition timeliness; the data acquisition timeliness determination result includes the first acquisition timeliness, the second acquisition timeliness, and the third acquisition timeliness.

[0041] like Figure 2As shown, it is a flow chart of data acquisition timeliness optimization provided by an embodiment of the present application. The present invention first performs a data acquisition timeliness determination, and the data acquisition timeliness determination result includes a first acquisition timeliness, a second acquisition timeliness, and a third acquisition timeliness. When the data acquisition timeliness determination result of a certain data sensor is the first acquisition timeliness, the data sensor is marked as a timeliness abnormal sensor, and further adjusted according to its sensor state parameter. If the sensor state parameter exceeds the threshold, the sampling frequency is dynamically adjusted. If the sensor state parameter does not exceed the threshold, a timeliness abnormality warning is performed; when the data acquisition timeliness determination result of a certain data sensor is the second acquisition timeliness, it is further determined whether the sum of the sampling frequencies of each data sensor exceeds its threshold. If so, the sampling frequency is calculated and adjusted, otherwise the dynamic data buffer is opened and the data acquisition time is dynamically corrected; when the data acquisition timeliness determination result of a certain data sensor is the third acquisition timeliness, no additional processing is performed. The present invention adopts a three-level judgment system to accurately identify sensor performance. Through sampling frequency sum monitoring and dynamic buffer management, it avoids system overload while ensuring the accuracy of key data collection and reduces unnecessary adjustment losses, providing a highly timely and reliable data foundation for DC cable status monitoring.

[0042] Specifically, the step of dynamically optimizing each data sensor based on the data acquisition timeliness determination result includes: if the data acquisition timeliness determination result of a data sensor is the first acquisition timeliness, then marking the data sensor as a timeliness abnormal sensor, and processing each timeliness abnormal sensor according to the sensor state parameter and sensor state threshold of each timeliness abnormal sensor.

[0043] The step of processing each time-sensor abnormality according to the sensor state parameter and the sensor state threshold of each time-sensor abnormality includes: comparing the sensor state parameter of each time-sensor abnormality with the sensor state threshold; if the sensor state parameter of a time-sensor abnormality is less than the sensor state threshold, issuing a sensor time-sensor abnormality warning and notifying a preset personnel to replace or repair the sensor; if the sensor state parameter of a time-sensor abnormality is greater than or equal to the sensor state threshold, determining whether the cable voltage fluctuation in the current monitoring time period of the time-sensor abnormality exceeds a preset voltage fluctuation threshold; if so, gradually adjusting the sampling frequency of the time-sensor abnormality according to a preset abnormal sensor sampling frequency adjustment value until the cable voltage fluctuation in the next monitoring time period does not exceed the voltage fluctuation threshold; if the sampling frequency of the time-sensor abnormality is adjusted to the preset maximum sampling frequency, but the cable voltage fluctuation in the next monitoring time period still exceeds the voltage fluctuation threshold, issuing a sensor abnormality warning and notifying a preset personnel to inspect and repair the sensor; otherwise, maintaining the current sampling frequency and reducing the first time-sensitivity assessment threshold of the time-sensor abnormality to the time-sensitivity assessment parameter of the time-sensor abnormality.

[0044] If the data collection timeliness judgment result of a data sensor is the second collection timeliness, then determine whether the sum of the sampling frequencies of each data sensor exceeds the preset sampling frequency threshold. If so, mark the difference between the sum of the sampling frequencies of each data sensor and the sampling frequency threshold as the deviation sampling frequency, and match the deviation sampling frequency with the sampling frequency adjustment value corresponding to each deviation sampling frequency preset in the DC power transmission database. In the DC power transmission database, each deviation sampling frequency and the sampling frequency adjustment value correspond one-to-one to form a mapping relationship table, which records each deviation sampling frequency and its corresponding sampling frequency adjustment value. These relationships can be one-to-one or many-to-one. When obtaining the sampling frequency adjustment value, it is only necessary to input the deviation sampling frequency into the mapping relationship table. The DC power transmission database can quickly locate and return the sampling frequency adjustment value corresponding to the deviation sampling frequency, and gradually adjust the sampling frequency of each data sensor according to the sampling frequency adjustment value until the sum of the sampling frequencies of each data sensor does not exceed the preset sampling frequency. Frequency threshold, otherwise the dynamic sensor data buffer is opened, and the dynamic memory management mechanism is used to expand the cache during congestion and quickly clear and release memory space when idle, which can help the sensor process data without delay due to the previous data not being processed completely, and mark the difference between the second threshold of acquisition timeliness evaluation and the timeliness evaluation parameter of the data sensor as the deviation acquisition timeliness evaluation parameter, and use the deviation acquisition timeliness evaluation parameter to match the data adjustment time corresponding to each deviation acquisition timeliness evaluation parameter preset in the database to obtain the data adjustment time corresponding to the timeliness evaluation parameter of the data sensor. When the acquisition data of a certain time monitoring point of the data sensor is obtained again, the acquisition data of the data sensor at the data adjustment time point before the time monitoring point can be obtained until the timeliness evaluation parameter of the data sensor reaches the second threshold of timeliness evaluation; if the data acquisition timeliness judgment result of a data sensor is the third acquisition timeliness, no additional processing is performed.

[0045] In this embodiment, the present invention significantly improves the data timeliness and reliability of the long-distance DC cable monitoring system through an intelligent sensor dynamic optimization mechanism. Its core advantages are reflected in three aspects: First, a hierarchical processing mechanism is established to perform intelligent diagnosis and adaptive adjustment on sensors with timeliness anomalies, which not only avoids invalid data interference but also maximizes the effective monitoring capability; second, a global coordination strategy is designed to optimize resource allocation at the system level and prevent data congestion through sampling frequency sum threshold control and dynamic buffer management; third, a data compensation mechanism is implemented to use historical data for timeliness compensation to ensure data continuity. These innovations work together to enable the system to automatically identify and process sensor anomalies, intelligently balance sampling loads, and effectively maintain data timeliness, ultimately achieving high-quality fusion of multimodal data and providing accurate and timely monitoring data support for the stable operation of the DC transmission system.

[0046] Furthermore, the timeliness impact parameters of the long-distance DC cable data fusion are used to quantitatively determine the timeliness impact of the data, and the steps for obtaining the data fusion timeliness impact determination result include: quantifying the impact of the relay station status on the timeliness of data transmission according to the timeliness impact parameters of each cable data, and obtaining the fusion timeliness impact parameters of each cable data, the timeliness impact parameters include the delay time of each relay station, the flow rate of each relay station and the timeliness evaluation parameter, and the fusion timeliness impact parameter represents the quantitative data of the degree of influence of the delay time of each relay station, the flow rate of each relay station and the timeliness evaluation parameter on the timeliness of data transmission; obtaining the timeliness impact parameter reference data from the preset DC transmission database, specifically including: the critical relay station delay time, the critical relay station flow rate and the critical timeliness evaluation parameter; performing a proportion approach calculation on the delay time of each relay station of each cable data and the flow rate of each relay station of each cable data, respectively, with the critical relay station delay time and the critical relay station flow rate, and then using the relay station delay time weight ratio and the relay station flow weight ratio to assign the relay station influence proportion approach calculation result respectively. The relay station delay time and relay station flow weighted processing results are coupled and averaged to obtain the relay station influence calculation result; the critical timeliness evaluation parameter and the timeliness evaluation parameter of each cable data are calculated with a proportion approach, and then the timeliness evaluation parameter proportion approach calculation result is weighted using the timeliness evaluation parameter weight ratio, and then the timeliness evaluation parameter weighted processing result is coupled with the relay station influence calculation result to obtain the fusion timeliness influence parameter of each cable data; the timeliness evaluation parameter weighted processing result is obtained from the preset DC transmission database. Take a timeliness impact threshold; compare the fusion timeliness impact parameter of each cable data with the timeliness impact threshold; if the fusion timeliness impact parameter of a certain cable data is greater than the timeliness impact threshold, then the corresponding data fusion timeliness impact determination result is recorded as the first fusion timeliness; if the fusion timeliness impact parameter of a certain cable data is less than or equal to the timeliness impact threshold, then the corresponding data fusion timeliness impact determination result is recorded as the second fusion timeliness; the data fusion timeliness impact determination result includes the first fusion timeliness and the second fusion timeliness.

[0047] The acquisition method of the influencing parameters of the fusion timeliness of each cable data is as follows:

[0048]

[0049] Where TI j represents the timeliness influencing parameter of the fusion of the j-th cable data, β1 represents the weight ratio of the relay station delay time, β2 represents the weight ratio of the relay station flow, β3 represents the weight ratio of the timeliness evaluation parameter, DT 1jkIndicates the delay time of the kth relay station of the jth type cable data, which means the time it takes for a certain type of cable data to be transmitted from the relay station to the next relay station. It can be obtained through network monitoring tools or the logs of the device itself. DT0 represents the critical relay station delay time, ST 1jk It represents the traffic of the kth relay station of the jth category cable data, which can be obtained in real time through network traffic analysis tools or monitoring equipment of the relay station. ST0 represents the traffic of the critical relay station, TE 1j It represents the timeliness evaluation parameter of the data sensor corresponding to the j-th type of cable data, TE0 represents the critical timeliness evaluation parameter, where j is the number of each type of cable data, j = 1, 2, 3, ..., J, J is the total number of cable data, k is the number of each relay station, k = 1, 2, 3, ..., M, M is the total number of relay stations.

[0050] β1, β2, and β3 are the weight ratios corresponding to the relay station delay time, relay station flow, and timeliness evaluation parameters preset in the DC transmission database. These weight ratios are numerical indicators that measure the influence of the above-mentioned timeliness influencing parameters on the fusion timeliness influencing parameters. Specifically, there is a mapping relationship table for each relay station delay time, relay station flow, and timeliness evaluation parameter. The table records each possible timeliness influencing parameter value and its corresponding weight ratio. These mapping relationships can be one-to-one or many-to-one. For example, in actual applications, when it is necessary to evaluate the fusion timeliness influencing parameters of a certain cable data, the measured relay station delay time, relay station flow, and timeliness evaluation parameters can be input into their respective corresponding mapping relationship tables, and the weight ratios corresponding to these values ​​can be quickly found. The weight ratio ranges from 0 to 1.

[0051] In this embodiment, the relay station delay time, relay station traffic, and timeliness evaluation parameters are interrelated. For example, the relay station delay time and traffic directly affect timeliness. At the same time, there is a certain positive correlation between traffic and delay time. When the traffic exceeds the carrying threshold of the relay station, the greater the traffic, the heavier the load of the relay station, and the longer the delay time may be. Conversely, the smaller the traffic, the shorter the delay time. If the delay time of some relay stations is longer or the traffic is larger, the timeliness evaluation parameter will be smaller. The data fusion timeliness influencing parameter obtained by comprehensive analysis reflects the timeliness of the data transmission process, and can detect abnormal transmission of the relay station. At the same time, according to the fusion timeliness influencing parameter, the traffic distribution is adjusted and the relay station configuration is optimized, which can achieve accurate timeliness management and optimization, ensure that the data can reach the destination within a reasonable time, and effectively improve the timeliness of the system and the stability of data transmission.

[0052] like Figure 3As shown, it is a flow chart of data fusion timeliness optimization provided by an embodiment of the present application. The present invention first performs a fusion timeliness judgment, and the fusion timeliness judgment result includes a first fusion timeliness and a second fusion timeliness. When the data fusion timeliness judgment result of a certain cable data is the first fusion timeliness, the cable data is marked as unqualified data, and the relay station resources are adjusted. When the data fusion timeliness judgment result of a certain cable data is the second fusion timeliness, the cable data is updated and stored. The present invention quickly identifies data quality problems through a hierarchical judgment mechanism, immediately triggers dynamic adjustment of relay station resources for unqualified data, and saves qualified data, effectively solving the timing confusion problem caused by relay station congestion.

[0053] Specifically, the steps of dynamically optimizing each cable data based on the data fusion timeliness impact judgment result include: if the data fusion timeliness judgment result of a certain cable data is the first fusion timeliness, the cable data is marked as unqualified data, a data abnormality prompt is sent to the unqualified data, and a data re-acquisition notification is sent at the same time, and the relay station resource scheduling parameters are dynamically adjusted according to the difference between the fusion timeliness impact parameter of the cable data and the timeliness impact threshold.

[0054] Among them, the step of dynamically adjusting the relay station resource scheduling parameters according to the difference between the fusion timeliness influencing parameter of the cable data and the timeliness influencing threshold includes: the relay station resource scheduling parameters include the relay station flow threshold and the relay station monitoring frequency; the difference between the fusion timeliness influencing parameter of the cable data and the timeliness influencing threshold is marked as the deviation fusion timeliness influencing parameter; the first deviation timeliness influencing threshold and the second deviation timeliness influencing threshold are obtained from the preset direct current transmission database; the deviation fusion timeliness influencing parameter is compared with the first deviation timeliness influencing threshold and the second deviation timeliness influencing threshold respectively, if the deviation fusion timeliness influencing parameter is greater than the second deviation timeliness influencing threshold, the relay station resource scheduling parameters of the relay station through which the cable data passes are adjusted to the preset maximum value of the relay station resource scheduling parameters, that is, the maximum flow limit allowed by the relay station (such as bandwidth, data packet rate, etc.) and the highest monitoring frequency of indicators such as flow and delay are turned on until the deviation fusion The timeliness impact parameter is less than the first threshold of the deviation timeliness impact; if the deviation fusion timeliness impact parameter is within the interval corresponding to the first threshold of the deviation timeliness impact and the second threshold of the deviation timeliness impact, the relay station flow threshold and the relay station monitoring frequency are adjusted step by step according to the preset relay station resource scheduling parameter adjustment value, and the relay station flow threshold adjustment value and the relay station monitoring frequency adjustment value are respectively used to sum with the relay station flow threshold and the relay station monitoring frequency to obtain the adjusted relay station flow threshold and the adjusted relay station monitoring frequency, and it is determined whether the deviation fusion timeliness impact parameter is less than the first threshold of the deviation timeliness impact under the adjusted relay station flow threshold and the adjusted relay station monitoring frequency. If so, no additional processing is performed; otherwise, the relay station flow threshold adjustment value and the relay station monitoring frequency adjustment value are continued to be summed with the adjusted relay station flow threshold and the adjusted relay station monitoring frequency until the deviation fusion timeliness impact parameter is less than the first threshold of the deviation timeliness impact;If the deviation fusion timeliness influencing parameter is less than the first deviation timeliness influencing threshold, the current relay station resource scheduling parameter is maintained, and it is determined whether the relay station traffic exceeds the relay station traffic threshold after adjustment. If so, the relay station bandwidth is dynamically adjusted according to the difference between the relay station traffic after adjustment and the relay station traffic threshold. The difference between the relay station traffic after adjustment and the relay station traffic threshold is marked as the deviation traffic value, and the deviation traffic value is compared with the first deviation traffic threshold and the second deviation traffic threshold preset in the DC transmission database respectively. If the deviation traffic value is less than the first deviation traffic threshold, the relay station bandwidth is increased by 5%. If the deviation traffic value is within the interval corresponding to the first deviation traffic threshold and the second deviation traffic threshold, the relay station bandwidth is increased by 10%. If the deviation traffic value is greater than the second deviation traffic threshold, the relay station bandwidth is increased by 20%. If the adjusted relay station bandwidth exceeds the preset bandwidth threshold, the adjusted relay station bandwidth is changed to the bandwidth threshold, and the service quality priority scheduling traffic shaping mechanism is enabled. Otherwise, the difference between the deviation timeliness influencing first threshold and the deviation fusion timeliness influencing parameter is changed to The value is matched with the relay station monitoring frequency adjustment value corresponding to the difference range between each first threshold value of deviation timeliness impact and the deviation fusion timeliness impact parameter preset in the DC transmission database. In the DC transmission database, the difference range between each first threshold value of deviation timeliness impact and the deviation fusion timeliness impact parameter and the relay station monitoring frequency adjustment value are mapped one-to-one to form a mapping relationship table. The table records the difference range between each first threshold value of deviation timeliness impact and the deviation fusion timeliness impact parameter and its corresponding relay station monitoring frequency adjustment value. These relationships can be one-to-one or many-to-one. When obtaining the relay station monitoring frequency adjustment value, it is only necessary to input the difference between the first threshold value of deviation timeliness impact and the deviation fusion timeliness impact parameter into the mapping relationship table. The DC transmission database can quickly locate and return the relay station monitoring frequency adjustment value corresponding to the difference between the first threshold value of deviation timeliness impact and the deviation fusion timeliness impact parameter. The relay station monitoring frequency is gradually adjusted according to the relay station monitoring frequency adjustment value until the fusion timeliness impact parameter of the cable data is less than or equal to the timeliness impact threshold.

[0055] If the data fusion timeliness determination result of a certain cable data is the second fusion timeliness, the cable data is marked as qualified data, and the qualified data is updated and stored.

[0056] In this embodiment, the present invention effectively solves the timeliness problem in the multimodal data transmission of long-distance DC cables through an intelligent data fusion dynamic optimization mechanism. The core value of this solution lies in: first, a data quality grading system is established, which divides data into two categories: unqualified and qualified based on the timeliness judgment results. For unqualified data, an abnormal alarm and retransmission mechanism are immediately triggered to ensure data reliability; second, a dynamic adjustment strategy based on deviation value is designed. According to the difference between the fusion timeliness influencing parameter and the threshold, the flow threshold, relay station monitoring frequency and bandwidth allocation of the relay station are intelligently adjusted to achieve optimal resource configuration; finally, fast parameter matching is achieved through a preset mapping relationship table, which greatly improves the system response speed. These technical means work together to ensure the transmission priority of key data and optimize the overall network resource utilization. Ultimately, the timeliness and accuracy of multimodal data fusion are significantly improved, providing a strong guarantee for the stable operation of the DC transmission system. This solution is particularly suitable for long-distance DC cable monitoring scenarios with multiple relay jumps and complex network conditions. It has significant advantages such as fast response, high resource utilization, and good system stability.

[0057] like Figure 4 As shown, it is a schematic diagram of a DC transmission data processing system for multimodal data fusion provided by an embodiment of the present application. The present application provides a system for applying a DC transmission data processing method based on multimodal data fusion, comprising: an acquisition and determination module, an acquisition optimization module, a fusion determination module, a fusion optimization module, and a DC transmission database; wherein the acquisition and determination module is used to perform a quantitative determination of the timeliness of data based on the timeliness evaluation parameters of the long-distance DC cable data sensor to obtain a data acquisition timeliness determination result; the acquisition optimization module is used to dynamically optimize each data sensor based on the data acquisition timeliness determination result, and perform cable data acquisition through the dynamically optimized data sensor, and perform a quantitative determination of each data sensor. Dynamic optimization means optimizing each data sensor according to the timeliness evaluation parameters of the long-distance DC cable data sensor to reduce the impact of sensor performance on the timeliness of DC cable data acquisition; a fusion judgment module is used to quantitatively judge the impact of data timeliness based on the timeliness impact parameters of long-distance DC cable data fusion, and obtain the data fusion timeliness impact judgment result; a fusion optimization module is used to dynamically optimize each cable data based on the data fusion timeliness impact judgment result. Dynamic optimization of each cable data means optimizing the relay station transmission status according to the timeliness impact parameters of long-distance DC cable data fusion to reduce the impact of the relay station status on the timeliness of DC cable data transmission.

[0058] An embodiment of the present application further provides a DC power transmission data processing device for multimodal data fusion, comprising: a processor and a memory storing instructions executable by the processor; when the processor is configured to execute the instructions, the electronic device implements a DC power transmission data processing method based on multimodal data fusion.

[0059] In summary, the embodiments of the present application comprehensively evaluate and optimize the timeliness of long-distance DC cable data sensors and data fusion processes. First, the timeliness of data acquisition is quantitatively determined based on the timeliness evaluation parameters of the data sensors, and each sensor is dynamically optimized based on the determination results, thereby improving the timeliness of data acquisition. Then, based on the timeliness influencing parameters of data fusion, the timeliness impact of the data fusion process is quantitatively evaluated, and the cable data is dynamically optimized to optimize the transmission status of the relay station to reduce the negative impact of the relay station status on the timeliness of DC cable data transmission, thereby achieving high-timeliness fusion of long-distance DC cable multimodal data.

[0060] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0061] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0062] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0063] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0064] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0065] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A method for processing DC transmission data based on multimodal data fusion, characterized in that: The following steps are involved: The timeliness evaluation parameters of the long-distance DC cable data sensor are used to quantitatively determine the timeliness of the data and obtain the result of the data acquisition timeliness determination. Dynamically optimizing each data sensor based on the data acquisition timeliness determination result, and performing cable data acquisition using the dynamically optimized data sensor, wherein the dynamic optimization of each data sensor means optimizing each data sensor based on the timeliness evaluation parameters of the long-distance DC cable data sensor to reduce the impact of sensor performance on the timeliness of DC cable data acquisition; The timeliness impact parameters of long-distance DC cable data fusion are used to quantitatively determine the impact of data timeliness, and the results of the data fusion timeliness impact determination are obtained. Based on the results of the data fusion timeliness impact judgment, each cable data is dynamically optimized. The dynamic optimization of each cable data means optimizing the relay station transmission status according to the timeliness impact parameters of the long-distance DC cable data fusion to reduce the impact of the relay station status on the timeliness of DC cable data transmission.

2. The method for processing DC power transmission data based on multimodal data fusion according to claim 1, characterized in that: The step of performing quantitative determination of data timeliness based on the timeliness evaluation parameters of the long-distance DC cable data sensor to obtain a data acquisition timeliness determination result includes: Quantifying the influence of the timeliness performance of the data sensor on the timeliness of DC cable data acquisition according to the timeliness evaluation parameters of each data sensor, and obtaining the timeliness evaluation parameters of each data sensor, wherein the timeliness evaluation parameters include sampling frequency, response time, clock synchronization error and sensor state parameters, and the timeliness evaluation parameters represent quantitative data of the degree of influence of the timeliness evaluation parameters on the timeliness of DC cable data acquisition; The sensor state parameters are obtained based on the influence of the sensor state on the timeliness of DC cable data acquisition, wherein the sensor state parameters include output impedance, insulation resistance and voltage drift, and the sensor state parameters represent the quantitative data of the degree of influence of the sensor state parameters on the working state of the sensor; Obtaining timeliness evaluation parameter reference data from a preset DC transmission database, specifically including: reference sampling frequency, allowable deviation sampling frequency, critical response time, critical clock synchronization error, and critical sensor state parameters; Perform relative deviation compliance calculation on the sampling frequency of each data sensor, the reference sampling frequency and the allowable deviation sampling frequency to obtain the sampling frequency calculation result; Performing a proportional approximation operation on the critical response time, critical clock synchronization error, and sensor state parameter of each data sensor, respectively, with the response time, clock synchronization error, and critical sensor state parameter to obtain a response time operation result, a clock synchronization error operation result, and a sensor state parameter operation result; The sampling frequency calculation result, the response time calculation result, the clock synchronization error calculation result and the sensor state parameter calculation result are weighted respectively using the timeliness evaluation parameter weight ratio, and then the weighted processing results are coupled to obtain the timeliness evaluation parameters of each data sensor, wherein the timeliness evaluation parameter weight ratio includes the sampling frequency weight ratio, the response time weight ratio, the clock synchronization error weight ratio and the sensor state parameter weight ratio; Obtaining a first timeliness evaluation threshold, a second timeliness evaluation threshold, and a sensor state threshold from a preset direct current transmission database; The timeliness evaluation parameter of each data sensor is compared with the first timeliness evaluation threshold and the second timeliness evaluation threshold respectively. If the timeliness evaluation parameter of a data sensor is less than the first timeliness evaluation threshold, the corresponding data collection timeliness determination result is recorded as the first collection timeliness; If the timeliness evaluation parameter of a data sensor is greater than or equal to the first timeliness evaluation threshold and less than the second timeliness evaluation threshold, the corresponding data collection timeliness determination result is recorded as the second collection timeliness; If the timeliness evaluation parameter of a data sensor is greater than or equal to the second timeliness evaluation threshold, the corresponding data collection timeliness determination result is recorded as the third collection timeliness; The data collection timeliness determination result includes a first collection timeliness, a second collection timeliness and a third collection timeliness.

3. The method for processing DC power transmission data based on multimodal data fusion according to claim 2, characterized in that: The step of quantifying the influence of the sensor state on the timeliness of DC cable data acquisition according to the sensor state parameter to obtain the sensor state parameter comprises: Obtain sensor status parameter reference data from a preset DC transmission database, specifically including: critical output impedance, critical insulation resistance, and critical voltage drift; Perform a proportion approximation operation on the output impedance, critical insulation resistance and voltage drift of each data sensor and the critical output impedance, insulation resistance and critical voltage drift respectively to obtain a proportion approximation operation result; The weighted ratios of the sensor state parameters are used to perform weighted processing on the results of the proportion approach calculation to obtain weighted processing results, wherein the weighted ratios of the sensor state parameters include the output impedance weighted ratio, the insulation resistance weighted ratio, and the voltage drift weighted ratio; The weighted processing results are coupled and then inversely proportionally calculated to obtain the sensor state parameters of each data sensor.

4. The method for processing DC power transmission data based on multimodal data fusion according to claim 1, characterized in that: The step of dynamically optimizing each data sensor based on the data collection timeliness determination result includes: If the data collection timeliness determination result of a data sensor is the first collection timeliness, the data sensor is marked as a timeliness abnormal sensor, and each timeliness abnormal sensor is processed according to the sensor state parameter and sensor state threshold of each timeliness abnormal sensor; If the data collection timeliness determination result of a data sensor is the second collection timeliness, then determine whether the sum of the sampling frequencies of the data sensors exceeds a preset sampling frequency threshold. If so, gradually adjust the sampling frequencies of the data sensors according to a sampling frequency adjustment value obtained by the difference between the sum of the sampling frequencies of the data sensors and the sampling frequency threshold, until the sum of the sampling frequencies of the data sensors does not exceed the preset sampling frequency threshold. Otherwise, open the dynamic sensor data buffer, and dynamically correct the data collection time according to the difference between the timeliness evaluation parameter of the data sensor and the second timeliness evaluation threshold, until the timeliness evaluation parameter of the data sensor reaches the second timeliness evaluation threshold. If the data collection timeliness determination result of a data sensor is the third collection timeliness, no additional processing is performed.

5. The method for processing DC power transmission data based on multimodal data fusion according to claim 4, characterized in that: The step of processing each timeliness abnormal sensor according to the sensor state parameter and sensor state threshold of each timeliness abnormal sensor includes: Compare the sensor state parameter of each timeliness abnormal sensor with the sensor state threshold value. If the sensor state parameter of a timeliness abnormal sensor is less than the sensor state threshold value, a sensor timeliness abnormality warning is issued. If the sensor state parameter of a certain time-sensitive abnormality sensor is greater than or equal to the sensor state threshold, then determine whether the cable voltage fluctuation in the current monitoring time period of the time-sensitive abnormality sensor exceeds the preset voltage fluctuation threshold. If so, gradually adjust the sampling frequency of the time-sensitive abnormality sensor according to the preset abnormality sensor sampling frequency adjustment value until the cable voltage fluctuation in the next monitoring time period does not exceed the voltage fluctuation threshold. If the sampling frequency of the time-sensitive abnormality sensor is adjusted to reach the preset sampling frequency maximum value, but the cable voltage fluctuation in the next monitoring time period still exceeds the voltage fluctuation threshold, then issue a sensor abnormality warning and notify the preset personnel to check and repair. Otherwise, the current sampling frequency is maintained, and the first threshold value of the timeliness evaluation of the timeliness abnormality sensor is reduced to the timeliness evaluation parameter of the timeliness abnormality sensor.

6. The method for processing DC power transmission data based on multimodal data fusion according to claim 1, characterized in that: The step of performing quantitative determination of the impact of data timeliness based on the timeliness impact parameter of the long-distance DC cable data fusion to obtain the data fusion timeliness impact determination result includes: Quantifying the influence of the relay station status on the timeliness of data transmission according to the timeliness influencing parameters of each cable data, and obtaining the fused timeliness influencing parameters of each cable data, wherein the timeliness influencing parameters include the delay time of each relay station, the flow rate of each relay station and the timeliness evaluation parameter, and the fused timeliness influencing parameter represents the quantitative data of the degree of influence of the delay time of each relay station, the flow rate of each relay station and the timeliness evaluation parameter on the timeliness of data transmission; Obtain reference data of timeliness influencing parameters from a preset HVDC database, specifically including: critical relay station delay time, critical relay station flow, and critical timeliness evaluation parameters; The delay time of each relay station of each cable data and the flow rate of each relay station of each cable data are respectively calculated with the delay time of the critical relay station and the flow rate of the critical relay station. Then, the weight ratio of the relay station delay time and the weight ratio of the relay station flow rate are used to respectively weight the results of the relay station influence proportion approach degree calculation. The weighted processing results of the relay station delay time and the relay station flow rate are coupled and averaged to obtain the relay station influence calculation result. The critical timeliness evaluation parameter and the timeliness evaluation parameter of each cable data are subjected to a proportion approximation operation, and the timeliness evaluation parameter proportion approximation operation result is weighted using the timeliness evaluation parameter weight ratio. The timeliness evaluation parameter weighting operation result is then coupled with the relay station influence operation result to obtain the fused timeliness influence parameter of each cable data. Obtaining a timeliness impact threshold from a preset DC transmission database; Compare the fusion timeliness impact parameter of each cable data with the timeliness impact threshold. If the fusion timeliness impact parameter of a certain cable data is greater than the timeliness impact threshold, the corresponding data fusion timeliness impact judgment result is recorded as the first fusion timeliness; If the fusion timeliness impact parameter of a certain cable data is less than or equal to the timeliness impact threshold, the corresponding data fusion timeliness impact determination result is recorded as the second fusion timeliness; The data fusion timeliness impact determination result includes a first fusion timeliness and a second fusion timeliness.

7. The method for processing DC power transmission data based on multimodal data fusion according to claim 1, characterized in that: The step of dynamically optimizing each cable data based on the data fusion timeliness impact determination result includes: If the data fusion timeliness determination result of a certain cable data is the first fusion timeliness, the cable data is marked as unqualified data, a data abnormality prompt is sent to the unqualified data, and the relay station resource scheduling parameters are dynamically adjusted according to the difference between the fusion timeliness influencing parameter of the cable data and the timeliness influencing threshold; If the data fusion timeliness determination result of a certain cable data is the second fusion timeliness, the cable data is marked as qualified data, and the qualified data is updated and stored.

8. The method for processing DC power transmission data based on multimodal data fusion according to claim 7, characterized in that: The step of dynamically adjusting the relay station resource scheduling parameters according to the difference between the fusion timeliness influencing parameter of the cable data and the timeliness influencing threshold comprises: The relay station resource scheduling parameters include a relay station flow threshold and a relay station monitoring frequency; The difference between the fusion timeliness influencing parameter of the cable data and the timeliness influencing threshold is marked as the deviation fusion timeliness influencing parameter; Obtaining a first threshold value of timeliness impact of deviation and a second threshold value of timeliness impact of deviation from a preset direct current transmission database; The deviation fusion timeliness impact parameter is compared with the first deviation timeliness impact threshold and the second deviation timeliness impact threshold respectively. If the deviation fusion timeliness impact parameter is greater than the second deviation timeliness impact threshold, the relay station resource scheduling parameter is adjusted to the preset maximum value of the relay station resource scheduling parameter until the deviation fusion timeliness impact parameter is less than the first deviation timeliness impact threshold; If the deviation fusion timeliness impact parameter is within the range of the deviation timeliness impact first threshold and the deviation timeliness impact second threshold, the relay station flow threshold and the relay station monitoring frequency are adjusted step by step according to the preset relay station resource scheduling parameter adjustment value until the deviation fusion timeliness impact parameter is less than the deviation timeliness impact first threshold; If the deviation fusion timeliness impact parameter is less than the deviation timeliness impact first threshold, the current relay station resource scheduling parameters are maintained, and it is determined whether the relay station traffic exceeds the adjusted relay station traffic threshold. If so, the relay station bandwidth is dynamically adjusted according to the difference between the adjusted relay station traffic and the relay station traffic threshold. Otherwise, the relay station monitoring frequency is gradually adjusted according to the relay station monitoring frequency adjustment value obtained based on the difference between the deviation timeliness impact first threshold and the deviation fusion timeliness impact parameter until the fusion timeliness impact parameter of the cable data is less than or equal to the timeliness impact threshold.

9. A system using the method for processing DC power transmission data based on multimodal data fusion according to any one of claims 1 to 8, characterized in that: It includes an acquisition and determination module, an acquisition and optimization module, a fusion and determination module, a fusion and optimization module, and a DC transmission database; The acquisition and determination module is configured to perform a quantitative determination of the timeliness of the data based on the timeliness evaluation parameters of the long-distance DC cable data sensor, and obtain a determination result of the timeliness of the data acquisition; The acquisition optimization module is used to dynamically optimize each data sensor based on the data acquisition timeliness determination result, and to perform cable data acquisition through the dynamically optimized data sensor. The dynamic optimization of each data sensor means optimizing each data sensor according to the timeliness evaluation parameters of the long-distance DC cable data sensor to reduce the impact of sensor performance on the timeliness of DC cable data acquisition; The fusion determination module is used to perform quantitative determination of the impact of data timeliness based on the timeliness impact parameter of the long-distance DC cable data fusion, and obtain a data fusion timeliness impact determination result; The fusion optimization module is used to dynamically optimize the cable data based on the data fusion timeliness impact judgment result. The dynamic optimization of the cable data means optimizing the relay station transmission status according to the timeliness impact parameters of the long-distance DC cable data fusion to reduce the impact of the relay station status on the timeliness of DC cable data transmission.

10. A DC transmission data processing device for multi-modal data fusion, characterized in that: include: a processor, a memory storing instructions executable by the processor; When the processor is configured to execute the instruction, the electronic device implements the direct current transmission data processing method based on multimodal data fusion according to any one of claims 1 to 8.

Images