Cable running state detection method, device and electronic equipment

Abstract

This invention discloses a method, apparatus, and electronic device for detecting the operating status of cables. The method includes: acquiring multimodal data of a target cable; determining multiple data fusion paths; performing data fusion processing on the multimodal data based on each of the multiple data fusion paths to obtain fused datasets corresponding to each data fusion path; selecting a target dataset from the fused datasets corresponding to each of the multiple data fusion paths; and detecting the operating status of the target cable based on the target dataset to obtain the cable detection result. This invention solves the technical problem of inaccurate cable operating status detection in related technologies.

Description

Technical Field

[0001] This invention relates to the field of power systems, and more specifically, to a method, apparatus, and electronic device for detecting the operating status of cables. Background Technology

[0002] In related technologies, cables are the lifeblood of power systems, and their operating status directly affects the safety and reliability of the power grid. Therefore, to ensure the long-term stable operation of cables, promptly detect potential hazards, and prevent faults, it is necessary to monitor the cable operating status. However, current technologies for monitoring cable operating status suffer from inaccuracies.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This invention provides a method, apparatus, and electronic device for detecting the operating status of cables, in order to at least solve the technical problem of inaccurate detection of cable operating status in related technologies.

[0005] According to one aspect of the present invention, a method for detecting the operating status of a cable is provided, comprising: acquiring multimodal data of a target cable; determining multiple data fusion paths, wherein the multiple data fusion paths respectively use data of different modes as reference data for data fusion; performing data fusion processing on the multimodal data based on the multiple data fusion paths respectively to obtain fused datasets corresponding to the multiple data fusion paths respectively; selecting a target dataset from the fused datasets corresponding to the multiple data fusion paths respectively; and performing operating status detection on the target cable based on the target dataset to obtain a cable detection result.

[0006] Optionally, when the multimodal data includes infrared spectrum data, operating parameter data, and environmental parameter data, the multiple data fusion paths include a first fusion path, a second fusion path, and a third fusion path, wherein the first fusion path is a fusion path that performs data fusion using infrared spectrum data as reference data, the second fusion path is a fusion path that performs data fusion using operating parameter data as reference data, and the third fusion path is a fusion path that performs data fusion using environmental parameter data as reference data.

[0007] Optionally, the multimodal data is fused based on the first fusion path using the following method to obtain a fused dataset corresponding to the first fusion path: Based on the comparison result between the current load current value and a first current threshold in the operating parameter data, the highest temperature value among all temperature values ​​corresponding to the infrared spectrum data is verified to obtain a first verification result, wherein the first verification result is used to characterize the accuracy of the highest temperature value under the current load current value; the temperature difference between the current ambient temperature value and the highest temperature value in the environmental parameter data is determined; based on the comparison result between the temperature difference and a temperature difference threshold, the highest temperature value is verified to obtain a second verification result, wherein the second verification result is used to characterize the accuracy of the highest temperature value under the current ambient temperature value; based on the first verification result and the second verification result, the multimodal data is fused to obtain a fused dataset corresponding to the first fusion path.

[0008] Optionally, the multimodal data is fused based on the second fusion path using the following method to obtain a fused dataset corresponding to the second fusion path: A first relationship table is determined corresponding to the current load current value in the operating parameter data, wherein the first relationship table characterizes the correlation between the load current value and the cable temperature value of the target cable; a temperature threshold corresponding to the current load current value is determined based on the first relationship table; the current load current value is verified based on a comparison between the highest temperature value among all temperature values ​​corresponding to the infrared spectrum data and the temperature threshold, resulting in a third verification result, wherein the third verification result characterizes the accuracy of the current load current value at the highest temperature value; the current load current value is verified based on a comparison between the environmental humidity value and the environmental humidity threshold in the environmental parameter data, resulting in a fourth verification result, wherein the fourth verification result characterizes the accuracy of the current load current value at the environmental humidity value; and the multimodal data is fused based on the third and fourth verification results to obtain a fused dataset corresponding to the second fusion path.

[0009] Optionally, the multimodal data is fused based on the third fusion path using the following method to obtain a fused dataset corresponding to the third fusion path: Based on the current ambient temperature value in the environmental parameter data and the highest temperature value among all temperature values ​​corresponding to the infrared spectrum data, the cable temperature rise value is determined; based on the cable temperature rise value and the temperature rise range, the current ambient temperature value is verified to obtain a fifth verification result, wherein the fifth verification result is used to characterize the accuracy of the current ambient temperature value at the highest temperature value; a second relationship table corresponding to the current load current value in the operating parameter data is determined, wherein… In this process, the second relationship table is used to characterize the correlation between the load current value and the ambient temperature value of the target cable; based on the second relationship table, a second current threshold corresponding to the current ambient temperature value is determined; based on the comparison result between the current load current value and the second current threshold, the current ambient temperature value is verified to obtain a sixth verification result, wherein the sixth verification result is used to characterize the accuracy of the current ambient temperature value under the current load current value; based on the fifth verification result and the sixth verification result, the multimodal data is subjected to data fusion processing to obtain a fused dataset corresponding to the third fusion path.

[0010] Optionally, the step of selecting the target dataset from the fused datasets corresponding to the plurality of data fusion paths includes: determining the data conflict index of the fused datasets corresponding to the plurality of data fusion paths; determining a plurality of candidate datasets from the fused datasets corresponding to the plurality of data fusion paths based on the data conflict index of the fused datasets corresponding to the plurality of data fusion paths, wherein the plurality of candidate datasets are fused datasets whose corresponding data conflict index is less than a data conflict threshold; and selecting the target dataset from the plurality of candidate datasets.

[0011] Optionally, the step of detecting the operating status of the target cable based on the target dataset to obtain the cable detection result includes: performing feature concatenation operation on each feature data in the target dataset to obtain a fused feature vector; and detecting the operating status of the target cable based on the fused feature vector to obtain the cable detection result.

[0012] According to one aspect of the present invention, a cable operation status detection device is provided, comprising: a first determining module for acquiring multimodal data of a target cable; a second determining module for determining multiple data fusion paths, wherein the multiple data fusion paths respectively use data of different modes as reference data for data fusion; a third determining module for performing data fusion processing on the multimodal data based on the multiple data fusion paths respectively to obtain fused datasets corresponding to the multiple data fusion paths respectively; a fourth determining module for filtering out a target dataset from the fused datasets corresponding to the multiple data fusion paths respectively; and a fifth determining module for performing operation status detection on the target cable based on the target dataset to obtain cable detection results.

[0013] According to one aspect of the present invention, an electronic device is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to execute the instructions to implement the cable operating status detection method described in any of the preceding embodiments.

[0014] According to one aspect of the present invention, a computer-readable storage medium is provided, wherein when the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the cable operation status detection method described in any of the preceding claims.

[0015] In this embodiment of the invention, by acquiring multimodal data of the target cable, a complete information foundation can be provided for subsequent multipath fusion, ensuring that the complementarity between different modal data can be utilized; multiple data fusion paths are determined, each using different modal data as a reference, allowing each path to examine the cable status from a unique perspective, avoiding the one-sidedness caused by a single reference perspective; multimodal data are fused based on these paths to obtain corresponding fused datasets, realizing parallel cross-validation of multi-source information, ensuring that reliable paths still support diagnosis when data is abnormal; the target dataset is selected from multiple fused datasets, eliminating unreliable fusion results caused by data conflicts or failures, avoiding abnormal data from contaminating the final detection conclusion; finally, the operational status is detected based on the selected target dataset, realizing cross-validation and optimization between fusion paths, avoiding interference from abnormal data on the detection results, thereby improving the accuracy of cable operational status detection, and thus solving the technical problem of inaccurate cable operational status detection in related technologies. Attached Figure Description

[0016] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0017] Figure 1 This is a flowchart of a cable operating status detection method according to an embodiment of the present invention;

[0018] Figure 2 This is a schematic diagram of the cable operation status detection system architecture in an optional embodiment of the present invention;

[0019] Figure 3 This is a structural block diagram of a cable operation status detection device according to an embodiment of the present invention. Detailed Implementation

[0020] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0021] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0022] Example 1

[0023] According to an embodiment of the present invention, an embodiment of a cable operating status detection method is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0024] Figure 1 This is a flowchart of a cable operating status detection method according to an embodiment of the present invention, such as... Figure 1 As shown, the method includes the following steps:

[0025] S102, acquire multimodal data of the target cable;

[0026] This involves the target cable, which is a power cable whose operational status needs to be monitored. For example, the target cable may be a high-voltage cable used to transmit electrical energy in the power system and is a key physical carrier for power grid transmission.

[0027] This involves multimodal data, which is a collection of various types of data reflecting the operating status of the target cable. For example, the multimodal data includes infrared spectrum data, operating parameter data, and environmental parameter data.

[0028] S104, determine multiple data fusion paths, wherein the multiple data fusion paths use data of different modalities as reference data for data fusion;

[0029] This involves multiple data fusion paths, which are processing paths that use corresponding modal data as a reference benchmark to perform consistency verification and feature integration on other modal data. For example, the data fusion path includes a first fusion path, a second fusion path, and a third fusion path, which respectively use infrared spectrum data, operational parameter data, and environmental parameter data as reference data for fusion. The main function is to perform cross-validation and parallel processing of multi-source information from the dominant perspective of different data modalities.

[0030] This involves data from different modalities, which are heterogeneous data types from different sensing methods or acquisition channels. For example, the data from different modalities includes infrared spectrum data (reflecting the temperature field distribution of the cable), operating parameter data (reflecting the electrical load status of the cable), and environmental parameter data (reflecting the external environmental conditions of the cable). Its main function is to provide diverse reference benchmarks for multi-path fusion, so that each fusion path can examine and verify other modal data from its own unique data perspective, avoiding the diagnostic bias caused by a single reference perspective.

[0031] This involves reference data, which is core data used as a benchmark and basis in a certain data fusion path to verify the accuracy and consistency of other modal data. For example, infrared spectrum data is used as reference data in the first fusion path, operating parameter data is used as reference data in the second fusion path, and environmental parameter data is used as reference data in the third fusion path. This gives the fusion process a clear logical direction and improves the credibility of the fusion results.

[0032] S106, based on multiple data fusion paths, perform data fusion processing on multimodal data to obtain fused datasets corresponding to the multiple data fusion paths respectively;

[0033] This involves data fusion processing, which is a process of fusing data from different modalities. This includes cross-validation, consistency judgment, and feature integration according to preset verification logic, so as to achieve mutual verification and supplementation of cross-modal data, eliminate the one-sidedness and potential errors of a single data source, and generate more credible fused information.

[0034] This involves a fused dataset, which is a structured data set obtained by performing fusion processing on multimodal data in the corresponding data fusion path.

[0035] S108, Select the target dataset from the fused datasets corresponding to multiple data fusion paths respectively;

[0036] This involves a target dataset, which is a fusion dataset selected from multiple fusion datasets for the final cable operation status detection. For example, the target dataset includes fusion datasets with a data conflict index less than the data conflict threshold. Its main function is to serve as a reliable input for the final diagnostic process, eliminate unreliable fusion results caused by data conflicts or failures, and avoid abnormal data contaminating the final detection conclusion.

[0037] S110, based on the target dataset, performs operational status detection on the target cable to obtain cable detection results.

[0038] This includes operational status detection, which is an operation used to quantitatively assess the operational status of a target cable, such as thermal defect detection.

[0039] This includes cable testing results, which reflect the operating status of the target cable, including information such as whether there is poor contact or insulation deterioration.

[0040] Through the aforementioned steps S102-S110, by acquiring multimodal data of the target cable, a complete information foundation can be provided for subsequent multipath fusion, ensuring that the complementarity between different modal data can be utilized; multiple data fusion paths are determined, each using different modal data as a reference, allowing each path to examine the cable status from a unique perspective, avoiding the one-sidedness caused by a single reference perspective; multimodal data are fused based on these paths to obtain corresponding fused datasets, realizing parallel cross-validation of multi-source information, ensuring that reliable paths still support diagnosis when data is abnormal; the target dataset is selected from multiple fused datasets, eliminating unreliable fusion results caused by data conflicts or failures, avoiding abnormal data contaminating the final detection conclusion; finally, the operational status is detected based on the selected target dataset, realizing cross-validation and optimization between fusion paths, avoiding interference from abnormal data on the detection results, thereby improving the accuracy of cable operational status detection, and thus solving the technical problem of inaccurate cable operational status detection in related technologies.

[0041] As an optional embodiment, when the multimodal data includes infrared spectrum data, operating parameter data, and environmental parameter data, the multiple data fusion paths include a first fusion path, a second fusion path, and a third fusion path. The first fusion path is a fusion path that uses infrared spectrum data as reference data for data fusion, the second fusion path is a fusion path that uses operating parameter data as reference data for data fusion, and the third fusion path is a fusion path that uses environmental parameter data as reference data for data fusion.

[0042] This involves infrared image data, which is used to characterize the surface temperature distribution of the target cable and its corresponding temperature values. It can be acquired by an infrared thermal imager. For example, the infrared image data includes infrared thermal images of the cable body, joints and terminals, as well as the temperature values ​​corresponding to each pixel. It is used to reflect the heating state of the target cable from a thermal perspective and provides an intuitive basis for judging whether the cable has thermal defects such as poor contact or insulation deterioration.

[0043] This includes operational parameter data, which is the electrical parameter data generated by the target cable during actual operation. For example, the operational parameter data includes load current value, operating voltage value, power factor, etc., which are used to characterize the load-bearing status of the target cable from an electrical perspective.

[0044] This involves environmental parameter data, which are physical measurements of the external environment in which the target cable is located. For example, the environmental parameter data includes ambient temperature, ambient humidity, wind speed, etc., which are used to reflect the external heat dissipation conditions and climate conditions of the target cable from an environmental perspective.

[0045] This involves a first fusion path, which is a processing path that uses infrared spectral data as a reference standard to perform consistency verification and fusion.

[0046] This involves a second fusion path, which is a processing path that uses runtime parameter data as a reference for consistency verification and fusion.

[0047] This involves a third fusion path, which is a processing path that uses environmental parameter data as a reference benchmark to perform consistency verification and fusion.

[0048] In this embodiment, multimodal data is divided into infrared spectrum data, operational parameter data, and environmental parameter data. Correspondingly, a first fusion path is constructed with infrared spectrum data as a reference, a second fusion path with operational parameter data as a reference, and a third fusion path with environmental parameter data as a reference. This allows each fusion path to perform data verification and fusion processing from its own unique data modality perspective. Since different modal data have logical relationships with each other, these three parallel paths can achieve cross-verification of the same set of multimodal data from three dimensions: thermal, electrical, and environmental. When the reference data in one path is abnormal, other paths can still perform verification and fusion based on their own different reference benchmarks. This ensures that the final generated fusion dataset has multi-source mutual verification support, avoids the overall diagnostic results being contaminated due to distortion of a single data source, solves the reliability problem caused by single-perspective fusion, and achieves the effect of improving the fault tolerance and robustness of the fusion results.

[0049] As an optional embodiment, the multimodal data is fused based on a first fusion path in the following manner to obtain a fused dataset corresponding to the first fusion path: Based on the comparison between the current load current value and a first current threshold in the operating parameter data, the highest temperature value among all temperature values ​​corresponding to the infrared spectrum data is verified to obtain a first verification result, wherein the first verification result is used to characterize the accuracy of the highest temperature value under the current load current value; the temperature difference between the current ambient temperature value and the highest temperature value in the environmental parameter data is determined; based on the comparison between the temperature difference and a temperature difference threshold, the highest temperature value is verified to obtain a second verification result, wherein the second verification result is used to characterize the accuracy of the highest temperature value under the current ambient temperature value; based on the first verification result and the second verification result, the multimodal data is fused to obtain a fused dataset corresponding to the first fusion path.

[0050] This includes the current load current value, which is the real-time current value carried by the target cable at the current operating moment.

[0051] This involves a first current threshold, which is a current boundary value set based on the rated current carrying capacity of the target cable.

[0052] This includes the highest temperature value, which is the maximum value among all temperature values ​​corresponding to the infrared spectral data.

[0053] This involves a first verification result, which is an output flag that determines the accuracy of the highest temperature value based on the comparison between the current load current value and a first current threshold. For example, the first verification result includes two states: consistent or inconsistent, which are used to characterize whether the highest temperature value is within a reasonable range under the current load current conditions, and to provide electrical side verification information on the reliability of temperature data for the fused dataset.

[0054] This includes the current ambient temperature value, which is the real-time temperature of the external environment where the target cable is located at the current moment.

[0055] This involves the temperature difference, which is the difference between the highest temperature value and the current ambient temperature value.

[0056] This involves a temperature difference threshold, which is a temperature difference boundary value determined based on the maximum safe tolerance temperature of the target cable insulation material and the expected ambient temperature range. For example, the temperature difference threshold can be calculated by subtracting the highest expected ambient temperature from the maximum safe tolerance temperature of the target cable insulation material, and is used as a benchmark threshold to determine whether the cable temperature rise exceeds the safe range.

[0057] This involves a second verification result, which is an output flag that determines the accuracy of the highest temperature value based on the comparison between the temperature difference and the temperature difference threshold. For example, the second verification result includes two states: consistent and inconsistent. It is used to characterize whether the highest temperature value is within a reasonable range under the current ambient temperature conditions, and to provide environmental verification information on the credibility of temperature data for the fused dataset.

[0058] In this embodiment, based on the comparison between the current load current value and the first current threshold in the operating parameter data, the highest temperature value among all temperature values ​​corresponding to the infrared spectrum data is verified to obtain a first verification result. This allows for the determination of the accuracy of the highest temperature value under the current load current value from the perspective of electrical load. Then, the temperature difference between the current ambient temperature value and the highest temperature value in the environmental parameter data is determined, and based on the comparison between the temperature difference and the temperature difference threshold, the highest temperature value is verified to obtain a second verification result. This allows for the determination of the accuracy of the highest temperature value under the current ambient temperature value from the perspective of the external environment. Since the two sets of verifications are based on different physical logics (current-heating relationship and temperature difference-heat dissipation relationship), the multimodal data is fused based on the first and second verification results to obtain a fused dataset corresponding to the first fusion path. This ensures that the temperature features in the fused dataset under this path have undergone multi-logic cross-verification, avoiding misjudgments due to operating condition fluctuations or data distortion caused by a single verification basis. This solves the problem of insufficient reliability of single-dimensional verification and achieves the effect of improving the accuracy and reliability of the fused dataset of the first fusion path.

[0059] As an optional embodiment, the following method is used to perform data fusion processing on multimodal data based on the second fusion path to obtain a fused dataset corresponding to the second fusion path: A first relationship table is determined corresponding to the current load current value in the operating parameter data, wherein the first relationship table is used to characterize the correlation between the load current value and the cable temperature value of the target cable; Based on the first relationship table, a temperature threshold corresponding to the current load current value is determined; Based on the comparison result between the highest temperature value among all temperature values ​​corresponding to the infrared spectrum data and the temperature threshold, the current load current value is verified to obtain a third verification result, wherein the third verification result is used to characterize the accuracy of the current load current value at the highest temperature value; Based on the comparison result between the environmental humidity value and the environmental humidity threshold in the environmental parameter data, the current load current value is verified to obtain a fourth verification result, wherein the fourth verification result is used to characterize the accuracy of the current load current value at the environmental humidity value; Based on the third verification result and the fourth verification result, the multimodal data is fused to obtain a fused dataset corresponding to the second fusion path.

[0060] This involves a first relationship table, which is a relationship table used to characterize the correlation between the load current value and the cable temperature value of the target cable. For example, the first relationship table can be constructed by calculating the theoretical conductor temperature under different load currents based on the thermoelectric model of the target cable and then correcting it with the rated allowable temperature.

[0061] This involves a temperature threshold, which is a temperature boundary value corresponding to the current load current value determined based on a first relational table. For example, the temperature threshold can be obtained by querying the first relational table under the current load current value to obtain the corresponding upper limit of cable temperature reference.

[0062] This involves a third verification result, which is an output flag that judges the accuracy of the current load current value based on the comparison between the highest temperature value and the temperature threshold. For example, the third verification result includes two states: consistent and inconsistent. It is used to characterize whether the current load current value is within a reasonable range under the current highest temperature condition, and to provide thermal verification information on the reliability of current data for the fused dataset.

[0063] This includes the ambient humidity value, which is the real-time humidity value of the external environment where the target cable is located at the current moment. For example, this ambient humidity value can be obtained by an integrated environmental sensor deployed near the target cable.

[0064] This involves an environmental humidity threshold, which is a humidity boundary value set based on the operating environment requirements of the target cable. For example, the environmental humidity threshold can be set according to the upper limit of humidity corresponding to the risk of insulation moisture in the cable operation standard.

[0065] This involves a fourth verification result, which is an output flag that determines the accuracy of the current load current value based on the comparison between the ambient humidity value and the ambient humidity threshold. For example, the fourth verification result includes two states: consistent and inconsistent. It is used to characterize whether the current load current value is within a reasonable range under the current ambient humidity conditions, and to provide the fused dataset with verification information on the reliability of the current data from the environmental side.

[0066] In this embodiment, by determining the first relationship table corresponding to the current load current value and querying the corresponding temperature threshold, and then comparing the highest infrared temperature value with the temperature threshold, the physical correlation between the load current and the cable temperature can be used to reversely verify whether the current load current value matches the measured temperature from the thermal side. At the same time, the ambient humidity value is compared with the ambient humidity threshold, and the current value is verified a second time from the environmental side by utilizing the objective constraint that the insulation performance will decrease under high humidity conditions, which will limit the cable current carrying capacity. Since the thermal side verification (temperature-current relationship) and the environmental side verification (humidity-insulation relationship) are based on different physical mechanisms, the abnormality of one verification link will not affect the other verification. Therefore, by integrating the third verification result and the fourth verification result to comprehensively evaluate the current load current value, it can be ensured that the current characteristics in the fused dataset under this path have been confirmed by multiple physical logic cross-validation, avoiding misjudgment caused by interference from local factors based on a single verification basis. This solves the problem of insufficient reliability of single-dimensional verification of electrical data in the second fusion path, and achieves the effect of improving the accuracy and credibility of the fused dataset of this path.

[0067] As an optional embodiment, the following method is used to perform data fusion processing on multimodal data based on a third fusion path to obtain a fusion dataset corresponding to the third fusion path: Based on the current ambient temperature value in the environmental parameter data and the highest temperature value among all temperature values ​​corresponding to the infrared spectrum data, the cable temperature rise value is determined; based on the cable temperature rise value and the temperature rise range, the current ambient temperature value is verified to obtain a fifth verification result, wherein the fifth verification result is used to characterize the accuracy of the current ambient temperature value at the highest temperature value; a second relationship table corresponding to the current load current value in the operating parameter data is determined, wherein the second relationship table is used to characterize the correlation between the target cable's load current value and the ambient temperature value; based on the second relationship table, a second current threshold corresponding to the current ambient temperature value is determined; based on the comparison result between the current load current value and the second current threshold, the current ambient temperature value is verified to obtain a sixth verification result, wherein the sixth verification result is used to characterize the accuracy of the current ambient temperature value at the current load current value; based on the fifth and sixth verification results, the multimodal data is fused to obtain a fusion dataset corresponding to the third fusion path.

[0068] This involves the cable temperature rise value, which is the difference between the highest temperature value and the current ambient temperature value. For example, the cable temperature rise value can be calculated by subtracting the current ambient temperature value from the highest temperature value to quantify the actual temperature rise of the target cable relative to the environment.

[0069] This involves a temperature rise range, which is the range of temperature rise fluctuations of the cable determined based on historical statistical data of the target cable under normal operating conditions. For example, this temperature rise range can be constructed based on historical temperature rise data of the target cable under different ambient temperature conditions.

[0070] This involves a fifth verification result, which is an output flag that determines the accuracy of the current ambient temperature value based on whether the cable temperature rise value is within the temperature rise range. For example, the fifth verification result includes two states: consistent and inconsistent. It is used to characterize whether the current ambient temperature value is within a reasonable range under the current highest temperature condition, and to provide thermal verification information on the credibility of the ambient temperature data for the fused dataset.

[0071] This involves a second relationship table, which is used to characterize the relationship between the load current value of the target cable and the ambient temperature value. For example, the second relationship table can be constructed based on the influence of ambient temperature on the current carrying capacity of the target cable, reflecting the upper limit of the safe current that the cable can carry under different ambient temperatures.

[0072] This involves a second current threshold, which is a current boundary value corresponding to the current ambient temperature value determined based on a second relation table. For example, the second current threshold can be obtained by querying the second relation table under the current ambient temperature value to obtain the corresponding safe current upper limit.

[0073] This includes a sixth verification result, which is an output flag that determines the accuracy of the current ambient temperature value based on the comparison between the current load current value and the second current threshold. For example, the sixth verification result includes two states: consistent and inconsistent. It is used to characterize whether the current ambient temperature value is within a reasonable range under the current load current condition, and to provide the electrical side with verification information on the credibility of the ambient temperature data for the fused dataset.

[0074] In this embodiment, the cable temperature rise value is obtained by calculating the difference between the highest temperature value and the current ambient temperature value, and it is determined whether the temperature rise value is within the temperature rise range. The historical statistical law of cable temperature rise relative to ambient temperature can be used to verify whether the current ambient temperature value matches the measured infrared temperature from the thermal side. Then, the second relationship table corresponding to the current load current value is determined and the second current threshold is obtained by querying. The current load current value is compared with the second current threshold, and the ambient temperature value is verified a second time from the electrical side using the physical constraint relationship between ambient temperature and cable current carrying capacity. Since the thermal side verification (temperature rise-ambient temperature relationship) and the electrical side verification (temperature-current carrying capacity relationship) are based on different physical logics, the current ambient temperature value is comprehensively evaluated by integrating the fifth verification result and the sixth verification result. This can avoid misjudgment caused by interference from local factors on a single verification basis, solve the problem of insufficient reliability of single-dimensional verification of ambient temperature in the third fusion path, and achieve the effect of improving the accuracy and credibility of the fusion dataset of this path.

[0075] As an optional embodiment, the target dataset is selected from the fused datasets corresponding to the multiple data fusion paths, including: determining the data conflict index of the fused datasets corresponding to the multiple data fusion paths; based on the data conflict index of the fused datasets corresponding to the multiple data fusion paths, determining multiple candidate datasets from the fused datasets corresponding to the multiple data fusion paths, wherein the multiple candidate datasets are fused datasets whose corresponding data conflict index is less than the data conflict threshold; and selecting the target dataset from the multiple candidate datasets.

[0076] This includes a data conflict index, which is used to quantify the total number of times each data fusion path is judged to be inconsistent among all consistency verification flags within a preset fusion period. For example, the data conflict index can be obtained by accumulating the number of conflicts for each data fusion path through a conflict counter, serving as a quantitative basis for measuring the overall data consistency and reliability of the fusion path.

[0077] This involves a data conflict threshold, which is a boundary value of the conflict index used to determine whether a fused dataset can be adopted. For example, the data conflict threshold can be calculated by multiplying the total number of consistency checks within a preset fusion cycle by the allowable conflict ratio. As a boundary standard for screening candidate datasets, fused datasets with conflict levels exceeding the allowable range are eliminated.

[0078] This involves multiple candidate datasets, which are fusion datasets selected from multiple fusion datasets whose data conflict index is less than the data conflict threshold. For example, these multiple candidate datasets include fusion datasets corresponding to each effective fusion path retained after comparing conflict indices, which are used as candidate sets for further screening of the final target dataset, ensuring that the data participating in subsequent screening have basic reliability guarantees.

[0079] In this embodiment, the total number of inconsistencies in the consistency verification of each fused dataset within a preset period is used as the data conflict index, which quantifies the overall reliability of each data fusion path. This data conflict index is then compared with a data conflict threshold. Since a lower conflict index indicates a higher degree of mutual corroboration between the modal data within the path and a lower probability of data distortion or anomalies, multiple candidate datasets with conflict indices below the threshold are selected, eliminating fusion results with insufficient reliability. Finally, the target dataset is selected from the multiple candidate datasets, ensuring that the data ultimately used for detection has undergone dual quality checks of the conflict index and the threshold. This prevents highly conflicting and anomalous data from entering the diagnostic stage, solving the problem of unreliable fused data contaminating the final detection results and improving the reliability of the target dataset.

[0080] As an optional embodiment, based on the target dataset, the target cable is subjected to operation status detection to obtain cable detection results, including: performing feature concatenation operation on each feature data in the target dataset to obtain a fused feature vector; and performing operation status detection on the target cable based on the fused feature vector to obtain cable detection results.

[0081] This involves various feature data, which are feature information from different modalities contained in the target dataset. For example, these feature data include the highest temperature value in the infrared spectrum data, the load current value in the operating parameter data, and the ambient temperature and humidity values ​​in the environmental parameter data, providing multi-dimensional basic data elements for constructing the fused feature vector.

[0082] This involves a feature concatenation operation, which is a data integration process that combines the feature data in the target dataset into a fused feature vector in a predetermined order. For example, the feature concatenation operation can use vector concatenation to link the feature data in the feature dimension to form a feature vector containing all the key information.

[0083] This involves a fused feature vector, which is a structured vector obtained by concatenating the features of each feature data.

[0084] In this embodiment, feature concatenation is performed on the feature data of each feature data in the target dataset to obtain a fused feature vector. This integrates heterogeneous features from multiple modalities such as infrared spectra, operating parameters, and environmental parameters into a structured vector representation in a unified format, enabling information from different dimensions to form a complete data expression at the feature level. Then, the target cable's operating status is detected based on this fused feature vector. Since the fused feature vector contains multimodal comprehensive information that has been filtered and verified, subsequent detection can simultaneously utilize multiple dimensions such as thermal, electrical, and environmental features for comprehensive judgment, avoiding the one-sidedness caused by relying solely on single-modal features for diagnosis, thereby improving the accuracy and reliability of the cable operating status detection results.

[0085] Based on the above embodiments and optional embodiments, an optional implementation method is provided, which is described in detail below.

[0086] In related technologies, cables are the lifeblood of power systems, and their operational status directly affects the safety and reliability of the power grid. Therefore, to ensure the long-term stable operation of cables, promptly identify potential hazards, and prevent faults, it is necessary to monitor the cable's operational status. However, current technologies for monitoring cable operational status suffer from inaccuracies.

[0087] For example, in related technologies, infrared thermal imaging technology is used as a non-contact detection method to detect the surface temperature of cables and their accessories (such as joints and terminals) and to provide fault warnings. However, the following problems still exist:

[0088] First, the reliability of diagnostics based on a single data source is insufficient. The surface temperature of a cable is not only affected by internal faults (such as poor contact or insulation degradation), but also strongly correlated with operating parameters (such as load current) and environmental parameters (such as ambient temperature, sunlight, and wind speed). Related technologies often rely solely on the absolute value of infrared temperature or simple temperature rise for judgment, which can easily lead to false alarms (such as misjudging a fault due to a sudden increase in load) or missed alarms (such as good ambient heat dissipation masking potential defects).

[0089] Secondly, there is a lack of effective multi-source data fusion mechanisms. Some existing detection systems attempt to introduce multimodal data, but they mostly adopt simple data stacking or serial processing modes. This mode lacks the ability to cross-validate data, and any anomaly or distortion from any data source will directly "contaminate" the final diagnostic results, resulting in poor fault tolerance and robustness of the system.

[0090] Finally, the adaptability and intelligence of the diagnostic logic need to be improved. Most existing detection systems use fixed threshold criteria, which cannot adapt to the dynamic operating conditions of cables and complex environmental interference. They also lack the intelligent decision-making ability to provide reliable conclusions even when some data is unreliable or contradictory.

[0091] There is currently no effective solution to the above problems.

[0092] In view of this, an optional embodiment of the present invention provides a cable operation status detection method, which can be implemented by a cable operation status detection system to effectively solve the above-mentioned technical problems.

[0093] Taking high-voltage cables (cables used to transmit high voltage, ranging from 1kV to 1000kV) as an example, Figure 2 This is a schematic diagram of the cable operation status detection system architecture in an optional embodiment of the present invention, such as... Figure 2 As shown, the system includes a multimodal data acquisition module, a data preprocessing module, a multimodal data fusion module, an intelligent diagnosis module, and a result output module. Among them:

[0094] The multimodal data acquisition module is used to collect multimodal data of high-voltage cables, including infrared spectrum data, operating parameter data, and environmental parameter data.

[0095] The data preprocessing module, connected to the multimodal data acquisition module, is used to preprocess the acquired multimodal data (including data cleaning, format unification and timestamp alignment), and output the preprocessed multimodal data, including preprocessed infrared spectrum data, operating parameter data and environmental parameter data.

[0096] The multimodal data fusion module, connected to the data preprocessing module, is used to integrate the preprocessed multimodal data in parallel, generate multiple data fusion paths, and output the fused data set that passes the path filtering.

[0097] The intelligent diagnostic module, connected to the multimodal data fusion module, is used to analyze all filtered fusion data sets to determine the condition of high-voltage cables and generate diagnostic conclusions.

[0098] The results output module, connected to the intelligent diagnosis module, is used to display diagnostic conclusions.

[0099] S1, acquire multimodal data of high-voltage cables;

[0100] For example, the multimodal data is acquired from the high-voltage cable through the multimodal data acquisition module. After that, the multimodal data can be further preprocessed by the data preprocessing module (including data cleaning, format unification and timestamp alignment). Subsequent processing is based on the preprocessed multimodal data.

[0101] S2, determine multiple data fusion paths, wherein the multiple data fusion paths use data of different modalities as reference data for data fusion;

[0102] Specifically, when the multimodal data includes infrared spectrum data, operational parameter data, and environmental parameter data, the multiple data fusion paths include a first fusion path, a second fusion path, and a third fusion path. The first fusion path is a fusion path that uses infrared spectrum data as reference data for data fusion, the second fusion path is a fusion path that uses operational parameter data as reference data for data fusion, and the third fusion path is a fusion path that uses environmental parameter data as reference data for data fusion.

[0103] S3, based on multiple data fusion paths, performs data fusion processing on multimodal data to obtain fused datasets corresponding to the multiple data fusion paths respectively;

[0104] For example, the multimodal data fusion module performs data fusion processing on multimodal data based on multiple data fusion paths, resulting in fused datasets corresponding to each of the multiple data fusion paths.

[0105] The multimodal data fusion module includes:

[0106] The parallel fusion unit is used to simultaneously generate a first fusion path based on infrared spectral data as the guiding data, a second fusion path based on operational parameter data as the guiding data, and a third fusion path based on environmental parameter data as the guiding data.

[0107] The path management unit, connected to the parallel fusion unit, is used to perform consistency checks, data overlay, and path filtering on the first fusion path, the second fusion path, and the third fusion path. The path management unit includes a consistency check submodule, a conflict counter, a data overlay submodule, and a path filtering submodule. The consistency check submodule includes a first path check unit, a second path check unit, and a third path check unit.

[0108] Specifically, through the first path verification unit, based on the first fusion path, data fusion processing is performed on the multimodal data to obtain a fusion dataset corresponding to the first fusion path; based on the comparison result between the current load current value and the first current threshold in the operating parameter data, the highest temperature value among all temperature values ​​corresponding to the infrared spectrum data is verified to obtain a first verification result, wherein the first verification result is used to characterize the accuracy of the highest temperature value under the current load current value; the temperature difference between the current ambient temperature value and the highest temperature value in the environmental parameter data is determined; based on the comparison result between the temperature difference and the temperature difference threshold, the highest temperature value is verified to obtain a second verification result, wherein the second verification result is used to characterize the accuracy of the highest temperature value under the current ambient temperature value; based on the first verification result and the second verification result, data fusion processing is performed on the multimodal data to obtain a fusion dataset corresponding to the first fusion path.

[0109] For example, the first path verification unit is connected to the parallel fusion unit to receive infrared spectral data from the first data fusion path and perform the following operations:

[0110] Extract the highest temperature feature value (i.e., the highest temperature value) from the infrared spectrum data; receive the operating parameter data for fusion verification with the basic guidance data of the first fusion path, and read the load current value (i.e., the current load current value) from it; compare the load current value with the preset first current threshold to generate a first current consistency flag (i.e., the first verification result), wherein the first current threshold is set based on the rated current carrying capacity of the high-voltage cable;

[0111] The system receives environmental parameter data for fusion verification with the basic guidance data of the first fusion path and reads the ambient temperature value (i.e., the current ambient temperature value). It calculates the difference between the highest temperature feature value and the ambient temperature value and compares the difference with the pre-stored first temperature difference threshold to generate a first temperature difference consistency flag (i.e., the second verification result). The first temperature difference threshold is a fixed threshold determined based on the maximum safe tolerance temperature of the high-voltage cable insulation material.

[0112] Specifically, through the second path verification unit, multimodal data is fused along the second fusion path to obtain a fused dataset corresponding to the second fusion path: a first relationship table is determined corresponding to the current load current value in the operating parameter data, wherein the first relationship table is used to characterize the correlation between the load current value and the temperature value of the high-voltage cable; based on the first relationship table, a temperature threshold corresponding to the current load current value is determined; based on the comparison result between the highest temperature value among all temperature values ​​corresponding to the infrared spectrum data and the temperature threshold, the current load current value is verified to obtain a third verification result, wherein the third verification result is used to characterize the accuracy of the current load current value at the highest temperature value; based on the comparison result between the environmental humidity value and the environmental humidity threshold in the environmental parameter data, the current load current value is verified to obtain a fourth verification result, wherein the fourth verification result is used to characterize the accuracy of the current load current value at the environmental humidity value; based on the third verification result and the fourth verification result, multimodal data is fused to obtain a fused dataset corresponding to the second fusion path.

[0113] For example, the second path verification unit is connected to the parallel fusion unit to receive the running parameter data of the second data fusion path and perform the following operations:

[0114] The load current value is extracted from the operating parameter data; infrared spectrum data for fusion verification with the basic guidance data of the second fusion path is received, and the highest temperature characteristic value is read from it; the load current value is queried from the preset first correspondence table (i.e., the first relationship table) to obtain the corresponding reference temperature threshold (i.e., the temperature threshold), and the highest temperature characteristic value is compared with the reference temperature threshold to generate a second temperature consistency flag (i.e., the third verification result); wherein, the first correspondence table is based on the thermoelectric model of the high-voltage cable and is constructed with reference to the nominal rated allowable temperature of the high-voltage cable, and is used to describe the theoretical upper limit of the conductor or key component of the high-voltage cable under different load currents;

[0115] The system receives environmental parameter data for fusion verification with the basic guidance data of the proposed fusion path, and reads the environmental humidity value from it. The environmental humidity value is compared with the preset humidity threshold (i.e., the environmental humidity threshold) to generate a first humidity consistency flag (i.e., the fourth verification result). The humidity threshold is set based on the operating environment requirements of the high-voltage cable.

[0116] Specifically, through the third path verification unit, based on the third fusion path, the multimodal data is fused to obtain a fused dataset corresponding to the third fusion path: Based on the current ambient temperature value in the environmental parameter data and the highest temperature value among all temperature values ​​in the infrared spectrum data, the high-voltage cable temperature rise value is determined; based on the high-voltage cable temperature rise value and the temperature rise range, the current ambient temperature value is verified to obtain a fifth verification result, which characterizes the accuracy of the current ambient temperature value at the highest temperature value; a second relationship table corresponding to the current load current value in the operating parameter data is determined, which characterizes the correlation between the high-voltage cable load current value and the ambient temperature value; based on the second relationship table, a second current threshold corresponding to the current ambient temperature value is determined; based on the comparison result between the current load current value and the second current threshold, the current ambient temperature value is verified to obtain a sixth verification result, which characterizes the accuracy of the current ambient temperature value at the current load current value; based on the fifth and sixth verification results, the multimodal data is fused to obtain a fused dataset corresponding to the third fusion path.

[0117] For example, the third path verification unit is connected to the parallel fusion unit to receive environmental parameter data from the third data fusion path and perform the following operations:

[0118] The system extracts the ambient temperature value from the environmental parameter data; receives the infrared spectrum data for fusion verification with the basic guidance data of the third fusion path, and reads the highest temperature characteristic value from it; calculates the difference between the highest temperature characteristic value and the ambient temperature value to obtain the high-voltage cable temperature rise value; determines whether the high-voltage cable temperature rise value is within the preset normal temperature rise range, and generates a second temperature difference consistency flag (i.e., the fifth verification result), where the normal temperature rise range is a dynamic range obtained based on the statistical analysis of historical normal operation data of the high-voltage cable under a specific ambient temperature;

[0119] The system receives the operational parameter data for fusion verification with the basic guidance data of the proposed third fusion path and reads the load current value from it. Based on the ambient temperature value, it queries the pre-stored second correspondence table (i.e., the second relationship table) to obtain the corresponding safe current threshold (i.e., the second current threshold). The system then compares the load current value with the safe current threshold to generate a second current consistency flag (i.e., the sixth verification result). The second correspondence table is constructed based on the influence of ambient temperature on the current carrying capacity of high-voltage cables.

[0120] S4, select the target dataset from the fusion datasets corresponding to multiple data fusion paths respectively;

[0121] Specifically, S4 includes: determining the data conflict index of the fusion datasets corresponding to the multiple data fusion paths through a conflict counter; determining multiple candidate datasets from the fusion datasets corresponding to the multiple data fusion paths through a path filtering submodule based on the data conflict index of the fusion datasets corresponding to the multiple data fusion paths, wherein the multiple candidate datasets are fusion datasets whose corresponding data conflict index is less than the data conflict threshold; and filtering the target dataset from the multiple candidate datasets.

[0122] For example, the conflict counter is connected to the consistency check submodule for:

[0123] Receive all consistency flags output by the first path verification unit, the second path verification unit, and the third path verification unit; establish conflict counts for the first data fusion path, the second data fusion path, and the third data fusion path respectively; whenever a consistency flag is determined to be inconsistent, increment the conflict count of its corresponding data fusion path by one.

[0124] The path filtering submodule is connected to the conflict counter and is used for:

[0125] At the end of the preset fusion cycle, the final conflict count value of each data fusion path is read; the final conflict count value is compared with the preset path conflict threshold; if the final conflict count value of a data fusion path exceeds the path conflict threshold, a failure flag for that path is generated.

[0126] S5, based on the target dataset, performs operational status detection on the high-voltage cable to obtain the high-voltage cable detection results.

[0127] Specifically, S5 includes: performing feature splicing operations on the feature data of each feature data in the target dataset through the data overlay submodule to obtain a fused feature vector; and using the intelligent diagnosis module to detect the operating status of the high-voltage cable based on the fused feature vector to obtain the high-voltage cable detection result.

[0128] For example, the data overlay submodule is connected to both the consistency check submodule and the path filtering submodule, and is used for:

[0129] For data fusion paths that have not been marked as invalid, when their consistency flag is determined to be consistent, a data overlay operation is performed. The data overlay operation is as follows: the key features of the data to be fused that triggered the consistency flag are combined with the key features of the basic guiding data of the current path to form the fused data set of the path. The combination includes concatenating the key features into a feature vector.

[0130] Subsequently, the multimodal data fusion module marks the data fusion paths that were not marked as invalid by the path filtering submodule and their corresponding fusion data sets, specifically fusion feature vectors, and outputs them to the intelligent diagnosis module.

[0131] The intelligent diagnostic module includes multiple diagnostic models and a decision-maker, wherein:

[0132] Multiple diagnostic models, each corresponding to a different type of data fusion path, are used to: receive the fused data set of a data fusion path and output a preliminary diagnostic result;

[0133] The decision-maker connects to all diagnostic models and receives all valid preliminary diagnostic results. It generates a final diagnostic conclusion based on a preset majority voting rule. If all data fusion paths are marked as invalid, a diagnostic failure alarm is output.

[0134] The following description, with specific examples, will further illustrate this point.

[0135] Considering that single infrared spectrum data is prone to false alarms / missed alarms due to load and environmental interference, a multimodal data acquisition module is used to collect infrared spectrum data, operating parameter data, and environmental parameter data of high-voltage cables, as detailed below:

[0136] Infrared image data acquisition: A high-resolution infrared thermal imager (resolution 640×512, temperature measurement range -20℃~150℃, temperature measurement accuracy ±2%) is used and installed on the poles or inspection robots along the high-voltage cable line. The focus is on photographing the high-voltage cable body, joints and terminal heads. One frame of image is acquired every 10 seconds, and the acquisition timestamp is recorded at the same time.

[0137] Operating parameter data acquisition: The power system's data acquisition and monitoring control system (SCADA) is accessed via industrial Ethernet. The operating parameters of the high-voltage cable are retrieved in real time from the SCADA database, including load current (unit: A), operating voltage (unit: kV), and power factor (unitless). The acquisition frequency is synchronized with the infrared spectrum data (once every 10 seconds).

[0138] Environmental parameter data acquisition: An integrated environmental sensor (temperature and humidity sensor + wind speed sensor) is deployed next to the infrared thermal imager to collect ambient temperature (unit: ℃, accuracy ±0.5℃), ambient humidity (unit: %RH, accuracy ±2%), and wind speed (unit: m / s, accuracy ±0.1m / s). The acquisition frequency is also once every 10 seconds.

[0139] Through the above operations, the synchronous acquisition of three types of data—infrared, operational, and environmental—is achieved, covering all influencing factors of the high-voltage cable's condition, including its own heating, operating load, and external environment. This lays the data foundation for subsequent multimodal fusion and avoids the one-sidedness of single infrared data.

[0140] Considering the inconsistent formats, asynchronous timing, and noise of multi-source data, which can lead to data mismatch during fusion, data standardization is necessary through preprocessing. Therefore, the data preprocessing module is connected to the multimodal data acquisition module to perform data cleaning, format unification, and timestamp alignment on the acquired raw data, and outputs standard infrared spectrum data, standard operating parameter data, and standard environmental parameter data. The specific techniques and operations are as follows:

[0141] 1) Data cleaning:

[0142] The processing of different noise types for the three types of data includes:

[0143] Infrared spectral data: Median filtering algorithm is used to remove salt-and-pepper noise (such as isolated bright spots caused by dust obstruction) from the image. If a frame of the image is missing (such as due to transmission interruption), it is filled in by the average temperature of the two frames before and after the image.

[0144] Operating parameter data: Outliers (such as a sudden increase in load current to 3 times the rated value due to sensor failure) are removed using the 3σ principle. Outliers are replaced with the average of the first 5 data collections. The 3σ principle means that in a normal distribution, approximately 99.73% of the data falls within the range of the mean μ plus or minus three standard deviations σ. Data outside this range can be considered outliers.

[0145] Environmental parameter data: If the humidity value exceeds 100% or is negative, it is considered invalid and replaced with the average value of three adjacent sensors in the same time period;

[0146] 2) Standardized format:

[0147] All data should be converted to a unified format. Infrared spectral data should have temperature features extracted (such as the highest and average temperatures in the high-voltage cable joint area) and stored in a unified format file in key-value pairs. Operating parameters and environmental parameters should retain the original fields directly, with only the units being unified (such as voltage being unified to "kV" and wind speed being unified to "m / s").

[0148] 3) Timestamp alignment:

[0149] The time of all acquisition devices is synchronized using NTP (Network Time Protocol), with the acquisition timestamp of infrared spectrum data as the benchmark (because infrared data is most sensitive to temperature changes). If the timestamp of operating parameters or environmental parameters differs from the benchmark by more than 500ms, the data values ​​are adjusted by linear interpolation to ensure that the three types of data correspond one-to-one at the same timestamp.

[0150] Through the above operations, standard, clean, and synchronized multimodal data is output, eliminating fusion barriers caused by data format and time differences, providing a reliable data foundation for subsequent consistency verification, and avoiding misjudgments caused by data chaos.

[0151] Considering that traditional multi-source fusion uses simple stacking and lacks cross-validation, a single data anomaly can contaminate the result. Therefore, the multimodal data fusion module is connected to the data preprocessing module to integrate different types of preprocessed data in parallel, generate multiple data fusion paths, and output the fused data set that passes the path filtering.

[0152] The multimodal data fusion module includes: a parallel fusion unit, used to simultaneously generate a first data fusion path based on standard infrared spectrum data as the guiding data, a second data fusion path based on standard operating parameter data as the guiding data, and a third data fusion path based on standard environmental parameter data as the guiding data;

[0153] The path management unit, connected to the parallel fusion unit, is used to perform consistency checks, data overlay, and path filtering on the first data fusion path, the second data fusion path, and the third data fusion path. The path management unit includes a consistency check submodule, a conflict counter, a data overlay submodule, and a path filtering submodule. The consistency check submodule includes a first path check unit, a second path check unit, and a third path check unit.

[0154] The first path verification unit is connected to the parallel fusion unit and is used to receive standard infrared spectral data from the first data fusion path and perform the following operations:

[0155] Extract the highest temperature feature value from the standard infrared spectrum data; receive the standard operating parameter data for fusion verification with the basic guidance data of the first fusion path, and read the load current value from it; compare the load current value with the preset first current threshold to generate a first current consistency flag, wherein the first current threshold is set based on the rated current carrying capacity of the high-voltage cable;

[0156] Receive standard environmental parameter data for fusion verification with the basic guidance data of the first fusion path, and read the ambient temperature value from it; calculate the difference between the highest temperature feature value and the ambient temperature value, and compare the difference with the pre-stored first temperature difference threshold to generate a first temperature difference consistency flag, wherein the first temperature difference threshold is a fixed threshold determined based on the maximum safe tolerance temperature of the high-voltage cable insulation material;

[0157] The second path verification unit is connected to the parallel fusion unit and is used to receive standard operating parameter data of the second data fusion path and perform the following operations:

[0158] The load current value is extracted from the standard operating parameter data; the standard infrared spectrum data for fusion verification with the basic guidance data of the second fusion path is received, and the highest temperature characteristic value is read from it; the load current value is queried from the preset first correspondence table to obtain the corresponding reference temperature threshold, and the highest temperature characteristic value is compared with the reference temperature threshold to generate a second temperature consistency flag; wherein, the first correspondence table is based on the thermoelectric model of the high-voltage cable and is constructed with reference to the nominal rated allowable temperature of the high-voltage cable, and is used to describe the theoretical upper limit of the conductor or key component of the high-voltage cable under different load currents;

[0159] The system receives standard environmental parameter data for fusion verification with the basic guidance data of the proposed second fusion path, and reads the environmental humidity value from it; it compares the environmental humidity value with a preset humidity threshold to generate a first humidity consistency flag, wherein the humidity threshold is set based on the operating environment requirements of the high-voltage cable;

[0160] The third path verification unit is connected to the parallel fusion unit and is used to receive standard environmental parameter data of the third data fusion path and perform the following operations:

[0161] The system extracts the ambient temperature value from the environmental parameter data; receives the standard infrared spectrum data for fusion verification with the basic guidance data of the third fusion path, and reads the highest temperature characteristic value from it; calculates the difference between the highest temperature characteristic value and the ambient temperature value to obtain the high-voltage cable temperature rise value; determines whether the high-voltage cable temperature rise value is within the preset normal temperature rise range, and generates a second temperature difference consistency flag, wherein the normal temperature rise range is a dynamic range obtained based on the statistical analysis of historical normal operation data of the high-voltage cable under a specific ambient temperature;

[0162] The system receives the standard operating parameter data for fusion verification with the basic guidance data of the proposed fusion path and reads the load current value from it; it queries the pre-stored second correspondence table according to the ambient temperature value to obtain the corresponding safe current threshold, and compares the load current value with the safe current threshold to generate a second current consistency flag. The second correspondence table is constructed based on the influence of ambient temperature on the current carrying capacity of high-voltage cables.

[0163] The conflict counter is connected to the consistency verification submodule and is used to: receive all consistency flags output by the first path verification unit, the second path verification unit, and the third path verification unit; establish conflict counts for the first data fusion path, the second data fusion path, and the third data fusion path respectively; and increment the conflict count of its data fusion path whenever a consistency flag is determined to be inconsistent.

[0164] The path filtering submodule is connected to the conflict counter and is used to: read the final conflict count value of each data fusion path at the end of the preset fusion cycle; compare the final conflict count value with the preset path conflict threshold; if the final conflict count value of a data fusion path exceeds the path conflict threshold, generate a failure flag for that path.

[0165] The data overlay submodule is connected to the consistency check submodule and the path filtering submodule, and is used to: for data fusion paths that have not been marked as invalid, when their consistency flag is determined to be consistent, perform a data overlay operation; wherein, the data overlay operation is: to combine the key features of the data to be fused that has triggered the consistency flag with the key features of the basic guiding data of the current path to form the fused data set of the path, wherein the combination includes concatenating the key features into a feature vector;

[0166] The multimodal data fusion module is used to output only the data fusion paths that were not marked as invalid by the path filtering submodule and their corresponding fusion data sets to the intelligent diagnostic module; the specific operation process is as follows:

[0167] Parallel Fusion Unit: This unit generates three data fusion paths. Using multi-threading, it simultaneously processes the three types of preprocessed data, generating three independent data fusion paths. Each path uses one type of data as the "basic guiding data" and the other two types as the "data to be fused," as detailed below:

[0168] The first data fusion path: the basic guiding data is "standard infrared spectrum data" (core feature: maximum temperature max_temp), and the data to be fused is "standard operating parameter data" (core feature: load current I) and "standard environmental parameter data" (core feature: ambient temperature T_env);

[0169] The second data fusion path: the basic guiding data is "standard operating parameter data" (core feature: load current I), and the data to be fused is "standard infrared spectrum data" (core feature: maximum temperature max_temp) and "standard environmental parameter data" (core feature: ambient humidity RH, where RH specifically represents relative humidity);

[0170] The third data fusion path: the basic guiding data is "standard environmental parameter data" (core feature: ambient temperature T_env), and the data to be fused is "standard infrared spectrum data" (core feature: maximum temperature max_temp) and "standard operating parameter data" (core feature: load current I);

[0171] The path management unit performs data verification, conflict counting, and filtering. It interacts with the parallel fusion unit via shared memory, processing the three paths in four steps.

[0172] Sm1, Consistency Verification Submodule:

[0173] Cross-modal data validation: This submodule contains three path validation units, which perform consistency validation on the basic guidance data and the data to be fused for each of the three paths, generating a consistency flag ("consistent" or "inconsistent"). The specific operations are as follows:

[0174] 1) First path verification unit:

[0175] For the first data fusion path, the consistency between infrared data and operational and environmental data is verified. The steps are as follows:

[0176] Extracting key features from basic guidance data: Read "max_temp" (e.g., 55.2℃) from the data field of standard infrared spectrum data; this is the maximum temperature characteristic value of the high-voltage cable.

[0177] Verification of consistency with operating parameters: Read the "load current I" (e.g., 400A) from the operating parameter data to be merged, and compare I with the preset "first current threshold". The first current threshold is set based on the rated current carrying capacity of the high-voltage cable: Assume the high-voltage cable model is YJV22-8.7 / 15kV-1×2500 (indicating that the high-voltage cable is a copper core cross-linked polyethylene insulated steel tape armored PVC sheathed power cable, rated voltage is 8.7 / 15kV, single core, conductor cross-sectional area is 2500). The rated current carrying capacity is 500A. Considering that the temperature will rise rapidly when the current carrying capacity exceeds 80%, the first current threshold is set to 500A × 80% = 400A. If I ≤ 400A, the "first current consistency flag = consistent" is generated. If I > 400A, "inconsistent" is generated.

[0178] Verify consistency with environmental parameters: Read the "ambient temperature T_env" (e.g., 30℃) from the environmental parameter data to be fused, calculate the temperature difference ΔT = max_temp - T_env (55.2℃ - 30℃ = 25.2℃), and compare ΔT with the preset "first temperature difference threshold"—the first temperature difference threshold is determined based on the maximum safe withstand temperature of the high-voltage cable insulation material: the high-voltage cable insulation material is cross-linked polyethylene (XLPE), whose maximum safe withstand temperature is 90℃. Considering that the highest ambient temperature is 40℃, the first temperature difference threshold is set to 90℃ - 40℃ = 50℃; if ΔT ≤ 50℃, generate "first temperature difference consistency flag = consistent"; if ΔT > 50℃, generate "inconsistent".

[0179] 2) Second Path Verification Unit: For the second data fusion path, verify the consistency between the running data and the infrared and environmental data. The steps are as follows:

[0180] Extracting key features from basic guidance data: Read "load current I" (e.g., 300A) from standard operating parameter data;

[0181] Verification of consistency with infrared data: Read "max_temp" (e.g., 42℃) from the infrared spectrum data to be fused. Query the "First Correspondence Table" based on I to obtain the reference temperature threshold. The First Correspondence Table is constructed based on the high-voltage cable thermoelectric model and nominal values ​​(rated parameter values ​​that ensure long-term safe operation under standard conditions): First, calculate the theoretical conductor temperature under different I values ​​using the thermoelectric model (formula: T_conductor=T_env+I²×R×t, where T_conductor is the cable conductor temperature, i.e., the core temperature; R is the high-voltage cable conductor resistance; t is the energizing time). Then, combine this with the nominal value "allowable temperature 70℃ at rated current carrying capacity 500A" to correct the theoretical temperature and form a table (e.g., reference temperature threshold = 35℃ at I=200A, = 45℃ at I=300A, = 55℃ at I=400A). If max_temp ≤ reference temperature threshold (42℃ ≤ 45℃), generate "Second Temperature Consistency Flag = Consistent"; otherwise, "Inconsistent".

[0182] Verify the consistency with environmental data: Read the "ambient humidity RH" (e.g., 75%) from the environmental parameter data to be fused, and compare it with the preset "humidity threshold" - the humidity threshold is set based on the operating environment requirements of high-voltage cables: according to cable operation standards, high-voltage cables are prone to insulation dampness when the humidity exceeds 85%, so the threshold is set to 85%; if RH≤85%, generate "first humidity consistency flag = consistent", otherwise "inconsistent";

[0183] 3) Third Path Verification Unit: For the third data fusion path, verify the consistency between environmental data and infrared and operational data. The steps are as follows:

[0184] Extract key features from basic guidance data: Read "ambient temperature T_env" (e.g., 25℃) from standard environmental parameter data;

[0185] Verify consistency with infrared data: Read "max_temp" (e.g., 40℃) from the infrared spectrum data to be fused, calculate the high-voltage cable temperature rise value ΔT_rise = max_temp - T_env (15℃), and determine whether ΔT_rise is within the "normal temperature rise range"—the normal temperature rise range is based on historical data statistics: collect the normal operation data of this type of high-voltage cable over the past 3 years (excluding fault periods), and statistically find that when T_env = 20℃, the normal temperature rise is 10-20℃, when T_env = 25℃ it is 8-18℃, and when T_env = 30℃ it is 5-15℃, forming a dynamic range; if 15℃ is within 8-18℃, generate "second temperature difference consistency flag = consistent", otherwise "inconsistent";

[0186] Verify the consistency of the operating data: Read the "load current I" (e.g., 450A) from the operating parameter data to be integrated, and obtain the safe current threshold by querying the "second correspondence table" based on T_env. The second correspondence table is constructed based on the influence of ambient temperature on current carrying capacity: an increase in ambient temperature will reduce the current carrying capacity of high-voltage cables, so when T_env=20℃, the safe current threshold = 500A (rated 100%), when T_env=25℃ = 480A (96%), and when T_env=30℃ = 450A (90%). If I ≤ safe current threshold (450A ≤ 480A), generate "second current consistency flag = consistent", otherwise "inconsistent".

[0187] Sm2, Conflict Counter: Counts the number of inconsistent paths. The conflict counter is connected to the consistency verification submodule via a serial port. After receiving all consistency flags, it performs the following operations:

[0188] For the first fusion path, the second fusion path and the third fusion path, independent conflict counters C1, C2 and C3 are established respectively, and the initial value of each is 0;

[0189] Traverse the consistency flags of each path: if any consistency flag of a path is "inconsistent", then the counter of that path is incremented by 1. For example, if the "first current consistency flag = inconsistent" of the first fusion path, then C1 = 0 + 1 = 1; if both "first current" and "first temperature difference" are inconsistent, then C1 = 2.

[0190] The counter value is updated every 10 seconds (synchronized with the data acquisition frequency) to record the real-time number of collisions for each path.

[0191] Sm3, Path Filtering Submodule: Eliminates invalid paths. The path filtering submodule communicates with the conflict counter via Ethernet and sets a "fusion period" (i.e., the time window for path validity judgment). The specific operation is as follows:

[0192] Determine the fusion cycle: Considering the lag in temperature changes of high-voltage cables (temperature stabilizes approximately 5 minutes after load change), the fusion cycle is set to 5 minutes (300 seconds), meaning that the path failure is checked every 5 minutes.

[0193] Set a path conflict threshold: A total of 30 consistency checks (300 seconds / 10 seconds) are performed within the fusion cycle. If the number of conflicts on a certain path is too high, it indicates that the data reliability of that path is low. Therefore, the threshold is set to 5 (i.e., the number of conflicts ≤ 5 is acceptable, and more than 5 is considered failure).

[0194] Execution filtering: When the fusion cycle ends, read the final values ​​of C1, C2, and C3. If C1=6 (exceeds the threshold of 5), generate "failure flag = yes" for the first fusion path; if C2=3 (≤5), generate "failure flag = no"; similarly, judge C3.

[0195] Sm4, Data Overlay Submodule: Generates the fused data set. The Data Overlay Submodule is connected to the Consistency Check Submodule and the Path Filtering Submodule, respectively. Data overlay is performed only on "non-failed paths". The steps are as follows:

[0196] Filter non-invalid paths: Read the results from the path filtering submodule and keep only paths with "invalid flag = no";

[0197] Consistency assessment: For non-failed paths, check if all their consistency flags are "consistent"—if a path is not failed but has an "inconsistent" flag, do not perform the stacking; if all flags are "consistent", then perform the stacking.

[0198] Data overlay operation: Concatenate the "key features of the basic guiding data" and the "key features of the data to be fused" of the path into a feature vector (i.e., the fused data set) — for example: if the second fusion path is not invalid and the consistency flags are all "consistent", the basic guiding data feature is "I=300A", and the data to be fused features are "max_temp=42℃" and "RH=75%", then the concatenation is into a feature vector: [300A,42℃,75%], and stored in a unified format (e.g., "fusion_data_2":{"features":[300,42,75],"path":"second"});

[0199] Output valid fusion data: Only the fusion data set of non-failed paths is output to the intelligent diagnosis module, and the fusion data of failed paths is not output to avoid invalid data interfering with the diagnosis;

[0200] Through the above operations, cross-validation of multimodal data is achieved using three parallel paths. Unreliable data is eliminated through conflict counting and path filtering, solving the problem of "data anomalies contaminating results" in traditional fusion. Data superposition forms a structured feature vector, providing high-quality input for intelligent diagnosis, and significantly improving the system's fault tolerance and robustness.

[0201] Considering that traditional diagnosis uses fixed thresholds, has poor adaptability, and lacks intelligent decision-making capabilities, the intelligent diagnosis module is connected to the multimodal data fusion module to analyze all filtered fusion data sets to determine the status of high-voltage cables and generate diagnostic conclusions. The intelligent diagnosis module includes multiple diagnostic models and a decision maker.

[0202] Multiple diagnostic models, each corresponding to a different type of data fusion path, are used to: receive the fused data set of a data fusion path and output a preliminary diagnostic result;

[0203] The decision-maker, connected to all diagnostic models, receives all valid preliminary diagnostic results and generates a final diagnostic conclusion based on a preset majority voting rule. If all data fusion paths are marked as invalid, a diagnostic failure alarm is output. The specific techniques and operations are as follows:

[0204] First, multiple diagnostic models: path-specific diagnostics, designing different types of diagnostic models (all trained on historical fault data) based on the feature differences of the first, second, and third fusion paths. Each model only receives the fusion data set of the corresponding path and outputs preliminary diagnostic results.

[0205] The first fusion path diagnostic model corresponds to the first fusion data set (infrared + operational + environmental features). It adopts a convolutional neural network (CNN) model. Since it needs to process the temperature spatial features of the infrared spectrum, the CNN can extract the spatial distribution features of the heat-generating area through the convolutional layer. Combined with operational and environmental features, it can determine the fault type (such as poor contact of the joint can cause local point heating, and excessive load can cause overall heating). The model outputs preliminary diagnostic results ("normal", "slight overheating", "severe overheating", "insulation degradation").

[0206] The second fusion path diagnostic model corresponds to the second fusion data set (operational + infrared + environmental features). It adopts a decision tree model. Since the operating parameters (load current) are the core causes of the fault, the decision tree can quickly determine the state through branch logic (such as I > 400A and max_temp > 55℃ → severe overheating) and output preliminary results.

[0207] The third fusion path diagnostic model corresponds to the third fusion data set (environmental + infrared + operational characteristics). It adopts the support vector machine (SVM) model. Since the influence of ambient temperature on temperature rise is non-linear, SVM can fit the non-linear relationship through the kernel function (e.g., when T_env=30℃, ΔT_rise>15℃→slight anomaly) and output preliminary results.

[0208] Decision Maker: Generates the final diagnostic conclusion. The decision maker communicates with the three diagnostic models through an application programming interface (API) and integrates the preliminary results according to the "majority voting rule." The specific operation is as follows:

[0209] Receive valid preliminary results: Only receive the diagnostic model output corresponding to the non-failed path. For example, if the first fusion path fails, only receive the preliminary results of the second and third models.

[0210] Execute majority vote:

[0211] If there are two or more valid paths: count the votes of the preliminary results, and the result with the most votes is the final conclusion. For example, if the second model outputs "normal" and the third model outputs "normal", then the final conclusion is "normal". If the second model outputs "mild overheating" and the third model outputs "normal", then "normal" is taken (a conservative approach is used to avoid false alarms).

[0212] If only one valid path exists: take the preliminary result of that path as the final conclusion.

[0213] Diagnostic failure handling: If all three paths are marked as failed (no valid preliminary results), the decision-maker outputs a "diagnostic failure alarm". The alarm information includes the "data anomaly type" (such as "continuous infrared data conflict, possible sensor failure"), reminding maintenance personnel to check the equipment.

[0214] Through the above steps, a path-specific model is adopted to adapt to different data characteristics, avoiding the limitations of a single model; majority voting ensures the reliability of decision-making; and diagnostic failure alerts avoid giving ambiguous conclusions, solving the problems of poor adaptability and unintelligent decision-making in traditional diagnosis and improving diagnostic accuracy.

[0215] Considering that the diagnostic results are not displayed intuitively and that maintenance personnel cannot quickly obtain key information, the results output module is connected to the intelligent diagnostic module and adopts a dual output method of "visual interface + report generation" to display the diagnostic conclusions. The specific technology and operation are as follows:

[0216] A visual interface is displayed, such as on the monitoring screen of a power dispatch center:

[0217] Final diagnosis (color-coded: green = normal, yellow = mild abnormality, red = severe abnormality, gray = diagnosis failed);

[0218] Effectively integrate key features of the dataset (such as "load current 300A, maximum temperature 42℃, ambient humidity 75%));

[0219] The failure status of each path (e.g., "First fusion path: Failed, number of conflicts 6").

[0220] Real-time infrared spectral image (with the location of the highest temperature marked).

[0221] Diagnostic report generation: Automatically generates diagnostic reports, including: report generation time, high-voltage cable number, key data statistics during the fusion period (such as average load current, highest ambient temperature), final diagnostic conclusion, and recommended measures (such as "minor anomaly: it is recommended to retest the load current within 24 hours" and "diagnosis failure: it is recommended to check the power supply of the infrared thermal imager"). The report is automatically sent to the operation and maintenance personnel via email.

[0222] Through the above operations, the diagnostic results become intuitive and easy to understand, and the key data and suggestions are clear. Maintenance personnel can quickly determine the status of high-voltage cables and take measures to shorten the fault response time.

[0223] The above optional implementation methods can achieve at least the following beneficial effects:

[0224] (1) Compared with related technologies, this invention realizes efficient and reliable fusion and intelligent diagnosis of infrared spectrum, operating parameters and environmental parameters by constructing a parallel multi-path fusion and consistency verification architecture, which greatly improves the accuracy and reliability of high-voltage cable status diagnosis, significantly reduces false alarm and false alarm rates, and enhances fault tolerance and robustness when facing single or multiple data anomalies.

[0225] (2) Compared with related technologies, this invention establishes three fusion paths based on different data and performs cross-modal consistency verification on the data to be fused (such as using load current to verify infrared temperature and using ambient temperature to verify cable temperature rise), thereby realizing mutual verification and supplementation of multi-source information, overcoming the one-sidedness of single data diagnosis, and significantly reducing misjudgments caused by working conditions or environmental interference.

[0226] (3) Compared with related technologies, the present invention adopts a parallel fusion engine to process three fusion paths at the same time, which greatly improves the efficiency of data fusion and the response speed of the system compared with the traditional serial processing mode, and meets the needs of real-time status monitoring of high-voltage cables.

[0227] (4) Compared with related technologies, this invention improves the transparency of the decision-making process by integrating the preliminary diagnostic results of multiple effective paths and forming a final conclusion based on preset voting rules. When all paths fail, a diagnostic failure alarm will be clearly output to remind maintenance personnel to intervene, thus avoiding giving incorrect or ambiguous conclusions.

[0228] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0229] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0230] Example 2

[0231] According to an embodiment of the present invention, an apparatus for implementing the above-described cable operating status detection method is also provided. Figure 3 This is a structural block diagram of a cable operating status detection device according to an embodiment of the present invention, such as... Figure 3 As shown, the device includes: a first determining module 302, a second determining module 304, a third determining module 306, a fourth determining module 308, and a fifth determining module 310. The device will be described in detail below.

[0232] The first determining module 302 is used to acquire multimode data of the target cable;

[0233] The second determining module 304 is connected to the first determining module 302 and is used to determine multiple data fusion paths, wherein the multiple data fusion paths use data of different modalities as reference data for data fusion.

[0234] The third determining module 306 is connected to the second determining module 304 and is used to perform data fusion processing on multimodal data based on multiple data fusion paths to obtain fused datasets corresponding to the multiple data fusion paths respectively.

[0235] The fourth determining module 308, connected to the third determining module 306, is used to filter out the target dataset from the fused datasets corresponding to the multiple data fusion paths respectively.

[0236] The fifth determining module 310, connected to the fourth determining module 308, is used to detect the operating status of the target cable based on the target dataset and obtain the cable detection result.

[0237] It should be noted that the first determining module 302, the second determining module 304, the third determining module 306, the fourth determining module 308, and the fifth determining module 310 mentioned above correspond to steps S102 to S110 in the cable operation status detection method. The multiple modules and the corresponding steps implement the same instances and application scenarios, but are not limited to the content disclosed in the above embodiment 1.

[0238] Example 3

[0239] According to another aspect of the present invention, an electronic device is also provided, comprising: a processor; and a memory for storing processor-executable instructions, wherein the processor is configured to execute instructions to implement the cable operation status detection method of any of the above embodiments.

[0240] Example 4

[0241] According to another aspect of the present invention, a computer-readable storage medium is also provided, which, when the instructions in the computer-readable storage medium are executed by a processor of an electronic device, enables the electronic device to perform the cable operation status detection method described above.

[0242] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0243] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0244] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0245] The units described 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 units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0246] Furthermore, the functional units in the various embodiments of the present invention 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.

[0247] 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 the present invention, 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 several 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 described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0248] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for detecting the operating status of a cable, characterized in that, include: Acquire multimodal data of the target cable; Multiple data fusion paths are determined, wherein the multiple data fusion paths use data of different modalities as reference data for data fusion. The multimodal data is fused based on the multiple data fusion paths to obtain fused datasets corresponding to the multiple data fusion paths respectively. The target dataset is selected from the fused datasets corresponding to the multiple data fusion paths. Based on the target dataset, the target cable is subjected to operational status detection to obtain cable detection results.

2. The method according to claim 1, characterized in that, When the multimodal data includes infrared spectrum data, operating parameter data, and environmental parameter data, the multiple data fusion paths include a first fusion path, a second fusion path, and a third fusion path. The first fusion path is a fusion path that uses infrared spectrum data as reference data for data fusion, the second fusion path is a fusion path that uses operating parameter data as reference data for data fusion, and the third fusion path is a fusion path that uses environmental parameter data as reference data for data fusion.

3. The method according to claim 2, characterized in that, The multimodal data is fused based on the first fusion path using the following method to obtain a fused dataset corresponding to the first fusion path: Based on the comparison result between the current load current value and the first current threshold in the operating parameter data, the highest temperature value among all temperature values ​​corresponding to the infrared spectrum data is verified to obtain a first verification result, wherein the first verification result is used to characterize the accuracy of the highest temperature value under the current load current value. Determine the temperature difference between the current ambient temperature value and the highest temperature value in the environmental parameter data; Based on the comparison result between the temperature difference and the temperature difference threshold, the highest temperature value is verified to obtain a second verification result, wherein the second verification result is used to characterize the accuracy of the highest temperature value under the current ambient temperature value. Based on the first verification result and the second verification result, the multimodal data is subjected to data fusion processing to obtain a fused dataset corresponding to the first fusion path.

4. The method according to claim 2, characterized in that, The multimodal data is fused using the second fusion path in the following manner to obtain a fused dataset corresponding to the second fusion path: A first relationship table is determined corresponding to the current load current value in the operating parameter data, wherein the first relationship table is used to characterize the correlation between the load current value and the cable temperature value of the target cable; Based on the first relationship table, determine the temperature threshold corresponding to the current load current value; Based on the comparison between the highest temperature value among all temperature values ​​corresponding to the infrared spectrum data and the temperature threshold, the current load current value is verified to obtain a third verification result, wherein the third verification result is used to characterize the accuracy of the current load current value at the highest temperature value. Based on the comparison result between the environmental humidity value and the environmental humidity threshold in the environmental parameter data, the current load current value is verified to obtain a fourth verification result, wherein the fourth verification result is used to characterize the accuracy of the current load current value under the environmental humidity value. Based on the third and fourth verification results, the multimodal data is subjected to data fusion processing to obtain a fused dataset corresponding to the second fusion path.

5. The method according to claim 2, characterized in that, The multimodal data is fused using the third fusion path in the following manner to obtain a fused dataset corresponding to the third fusion path: The cable temperature rise value is determined based on the current ambient temperature value in the environmental parameter data and the highest temperature value among all temperature values ​​corresponding to the infrared spectrum data. Based on the cable temperature rise value and the temperature rise range, the current ambient temperature value is verified to obtain a fifth verification result, wherein the fifth verification result is used to characterize the accuracy of the current ambient temperature value at the highest temperature value. A second relationship table is determined corresponding to the current load current value in the operating parameter data, wherein the second relationship table is used to characterize the correlation between the load current value of the target cable and the ambient temperature value; Based on the second relationship table, a second current threshold corresponding to the current ambient temperature value is determined; Based on the comparison result between the current load current value and the second current threshold, the current ambient temperature value is verified to obtain a sixth verification result, wherein the sixth verification result is used to characterize the accuracy of the current ambient temperature value under the current load current value; Based on the fifth and sixth verification results, the multimodal data is subjected to data fusion processing to obtain a fused dataset corresponding to the third fusion path.

6. The method according to claim 1, characterized in that, The step of selecting the target dataset from the fused datasets corresponding to the multiple data fusion paths includes: Determine the data conflict index of the fusion dataset corresponding to each of the multiple data fusion paths; Based on the data conflict index of the fusion datasets corresponding to the multiple data fusion paths, multiple candidate datasets are determined from the fusion datasets corresponding to the multiple data fusion paths, wherein the multiple candidate datasets are fusion datasets whose corresponding data conflict index is less than the data conflict threshold. The target dataset is selected from the multiple candidate datasets.

7. The method according to any one of claims 1 to 6, characterized in that, The step of performing operational status detection on the target cable based on the target dataset to obtain cable detection results includes: The feature data in the target dataset are concatenated to obtain a fused feature vector. Based on the fused feature vector, the operating status of the target cable is detected to obtain the cable detection result.

8. A cable operating status detection device, characterized in that, include: The first determining module is used to acquire multimodal data of the target cable; The second determining module is used to determine multiple data fusion paths, wherein the multiple data fusion paths respectively use data of different modalities as reference data for data fusion; The third determining module is used to perform data fusion processing on the multimodal data based on the multiple data fusion paths respectively, to obtain fused datasets corresponding to the multiple data fusion paths respectively; The fourth determining module is used to filter out the target dataset from the fused datasets corresponding to the multiple data fusion paths respectively; The fifth determining module is used to perform operational status detection on the target cable based on the target dataset and obtain cable detection results.

9. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the cable operating status detection method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by the processor of the electronic device, the electronic device is able to perform the cable operation status detection method as described in any one of claims 1 to 7.

Images