Fault detection method for cable branch box

Abstract

The invention relates to the technical field of power equipment fault detection, in particular to a fault detection method of a cable branch box, and aims to solve the problems that fault detection of the cable branch box in the prior art mainly depends on operation parameter monitoring of single equipment, and an independent environment sensor is adopted to collect peripheral data of the equipment, so that the fault detection efficiency is low. And the general rule that multiple devices in the same area are influenced by the environment is difficult to reflect, so that the correlation analysis of environmental factors and faults is lack of statistical support. According to the invention, multi-dimensional operation parameters such as current, voltage, temperature, humidity and the like are collected at the same time, the characteristic data of the environment area where the cable branch box is located are combined, comprehensive state evaluation is realized by using the fault detection model, and the method has the advantage of realizing accurate fault early warning through integrating the multi-dimensional operation parameters and the environment characteristic data.

Description

Technical Field

[0001] This application relates to the field of power equipment fault detection technology, and more specifically, to a fault detection method for cable branch boxes. Background Technology

[0002] In power systems, cable distribution boxes are crucial power distribution equipment, and their operating status directly affects the reliability and security of power supply. Existing technologies for fault detection in cable distribution boxes mainly suffer from the following problems: First, traditional detection methods typically focus only on a single operating parameter, such as current or voltage, lacking comprehensive analysis of multiple parameters, including temperature and humidity, resulting in an inability to fully assess the equipment's condition. Secondly, existing technologies often overlook the impact of environmental factors on equipment operation, while in actual operation, the performance of cable branch boxes can vary significantly depending on the environmental region they are located in. Furthermore, conventional detection methods struggle to identify potential faults. For instance, when equipment is in a critical state but no obvious abnormalities have yet appeared, existing methods cannot provide timely warnings. These problems are particularly pronounced in complex power grid environments and harsh weather conditions. Statistics show that approximately 40% of cable branch box faults originate from the combined effects of environmental factors and equipment condition, and nearly half of these faults develop into serious accidents due to the lack of timely warnings.

[0003] Furthermore, existing detection systems typically rely on fixed-location sensor networks, which cannot adapt to scenarios involving equipment movement or environmental changes, resulting in monitoring blind spots. To address these issues, existing technologies urgently need improvement. Summary of the Invention

[0004] (a) Technical problems to be solved To address the aforementioned issues, this invention proposes a fault detection method for cable branch boxes. This method aims to solve the problem that existing technologies for fault detection in cable branch boxes primarily rely on monitoring the operating parameters of a single device and using independent environmental sensors to collect data from the surrounding environment. This makes it difficult to reflect the common patterns of environmental influences on multiple devices within the same area, resulting in a lack of statistical support for the correlation analysis between environmental factors and faults.

[0005] (II) Technical Solution This application of the present invention provides a fault detection method for cable branch boxes, the technical solution of which is as follows: acquiring operating parameter data of the cable branch box, the operating parameter data including at least one of current, voltage, temperature, and humidity; acquiring environmental characteristic data of the cable branch box, the environmental characteristic data including environmental characteristic data of at least one environmental area in which the cable branch box is located during a specified time period, the environmental characteristic data of each environmental area being characterized by the operating parameter data of all cable branch boxes located in the environmental area during the specified time period; and providing the acquired operating parameter data and environmental characteristic data of the cable branch box to a cable branch box fault detection model for fault identification and early warning.

[0006] Furthermore, this application also proposes that obtaining environmental characteristic data of cable branch boxes includes: obtaining the environmental change trajectory of the cable branch box within a specified time period; determining at least one environmental area where the cable branch box is located based on the environmental change trajectory, wherein the environmental area is a preset environmental zone; for each determined environmental area, obtaining the operating parameter data of all cable branch boxes located in the environmental area within the specified time period to determine the operating parameter characteristics of the environmental area; and determining the environmental characteristic data of the cable branch box according to the operating parameter characteristics corresponding to each environmental area involved in the environmental change trajectory.

[0007] Furthermore, this application also proposes that determining at least one environmental area where a cable branch box is located based on an environmental change trajectory includes: using a geographic grid partitioning algorithm to determine at least one environmental area where the cable branch box is located based on the environmental change trajectory.

[0008] Furthermore, this application also proposes that obtaining environmental characteristic data of cable branch boxes further includes: determining the weights of the operating parameter characteristics corresponding to each environmental area, wherein determining the environmental characteristic data of cable branch boxes based on the obtained operating parameter characteristics corresponding to each environmental area includes: performing a weighted summation of the obtained operating parameter characteristics of each environmental area to obtain the environmental characteristic data of cable branch boxes.

[0009] Furthermore, this application also proposes that determining the weights of the operating parameter characteristics corresponding to each environmental area includes: for each environmental area, determining the weights of the operating parameter characteristics corresponding to the environmental area based on the frequency of occurrence of the environmental area in the environmental change trajectory, the total number of environmental areas involved in the environmental change trajectory, and the proportion of the number of cable branch boxes in the environmental area to the total number of monitoring within a specified time period.

[0010] Furthermore, this application also proposes to obtain the operating parameter data of all cable branch boxes in the environmental area within a specified time period for each identified environmental area, in order to determine the operating parameter characteristics of the environmental area, including: calculating the average value of the operating parameter data of each cable branch box in the environmental area as the operating parameter characteristics of the environmental area.

[0011] Furthermore, this application also proposes that each operating parameter data has a corresponding weight, and for each determined environmental area, the operating parameter data of all cable branch boxes in the environmental area within a specified time period is obtained to determine the operating parameter characteristics of the environmental area, including: weighted summation of the operating parameter data of each cable branch box in the environmental area to obtain the operating parameter characteristics of the environmental area.

[0012] Furthermore, this application also proposes a computing device comprising: at least one processor; and a memory storing instructions that, when executed by the at least one processor, cause the at least one processor to perform the method described above.

[0013] Furthermore, this application also proposes a non-transitory machine-readable storage medium storing executable instructions that, when executed, cause the machine to perform the above-described method.

[0014] (III) Beneficial Effects Compared with the prior art, the beneficial effects of the present invention are as follows: This invention simultaneously collects multi-dimensional operating parameters such as current, voltage, temperature, and humidity, and combines them with characteristic data of the environment where the cable branch box is located. It utilizes a fault detection model to achieve comprehensive status assessment, which has the advantage of achieving accurate fault early warning by integrating multi-dimensional operating parameters and environmental characteristic data. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of the logical structure of a fault detection method for cable branch boxes. Detailed Implementation

[0017] The following will refer to the appendix to this application. Figure 1The technical solutions in this application are clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application. It should be noted that similar reference numerals and letters in the following drawings indicate similar items; therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0018] In existing technologies, fault detection in cable distribution boxes primarily relies on monitoring the operating parameters of individual devices, such as detecting abnormal current or voltage. This method cannot effectively identify latent faults caused by environmental factors, such as insulation aging due to prolonged exposure to high humidity or localized overheating caused by regional power grid load fluctuations. While traditional technologies attempt to incorporate environmental parameter monitoring, they typically use independent environmental sensors to collect data from the surrounding environment, making it difficult to reflect the common patterns of environmental influences on multiple devices within the same area. This results in a lack of statistical support for the correlation analysis between environmental factors and faults.

[0019] To address the aforementioned issues, the inventors discovered that identifying latent faults requires considering both the equipment's own condition and environmental interference factors. Traditional single-device parameter monitoring cannot capture common environmental influences, while data from independent environmental sensors is difficult to directly correlate with equipment operating status. By analyzing the common characteristics of operating parameters of multiple devices in the same environmental area, it was found that the operating data of a cluster of devices can indirectly reflect the patterns of environmental influence on the devices. Based on this, a method is proposed to construct environmental characteristic data using the operating parameters of all devices within the environmental area, and to integrate this data with the parameters of individual devices for analysis, thereby establishing a dual assessment mechanism for equipment status and environmental factors.

[0020] Example 1 Therefore, this application proposes a fault detection method for cable branch boxes, comprising the following steps: S100. Obtain the operating parameter data of the cable branch box, including at least one of current, voltage, temperature, and humidity; S200. Obtain environmental characteristic data of the cable branch box. The environmental characteristic data includes environmental characteristic data of at least one environmental area of ​​the environment in which the cable branch box is located during a specified time period. The environmental characteristic data of each environmental area is characterized by the operating parameter data of all cable branch boxes in the environmental area during the specified time period. S300 provides the acquired operating parameter data and environmental characteristic data of the cable branch box to the cable branch box fault detection model for fault identification and early warning.

[0021] Among them, the operating parameter data refers to the physical quantities that reflect the operating status of the cable branch box. Specifically, it can be collected by current sensors, voltage sensors, temperature sensors or humidity sensors to characterize the real-time status of the equipment itself.

[0022] Environmental characteristic data refers to characteristic quantities that reflect the impact of the environmental area where the cable branch box is located on the equipment operation. Specifically, it can be calculated by statistically analyzing the operating parameter data of all cable branch boxes in the same environmental area within a specified time period. For example, the average or weighted value of the equipment operating parameters in the calculation area can be used to quantify the potential impact of environmental factors on the equipment.

[0023] The cable branch box fault detection model refers to an algorithmic model used to analyze operating parameters and environmental characteristic data. Specifically, it can be trained using machine learning models or deep learning models. By integrating the equipment's own status and environmental area characteristic data, it can identify equipment abnormalities and output early warning signals.

[0024] Specifically, during the operation of the cable distribution box, sensors first collect parameters such as current and voltage to construct equipment status monitoring data. Simultaneously, the environmental area involved is determined based on the equipment's location change trajectory within a specified time period, for example, by using a geographic grid partitioning algorithm to divide the area.

[0025] For each environmental area, operational parameter data for all cable branch boxes within that area are collected, such as calculating average current or temperature-weighted values, to form characteristic data representing the environmental impact of that area. The operational parameters of individual devices, along with the characteristic data of their respective environmental areas, are input into a fault detection model. The model identifies faults caused by equipment anomalies or environmental factors by comparing the differences between the device's own parameters and the regional environmental characteristics. For example, if the temperature of a device is significantly higher than the average temperature of other devices in the same area, the model can determine it as a localized overheating fault; if multiple devices in the area exhibit abnormal humidity parameters, it may indicate a decline in the insulation performance of the devices due to abnormal environmental humidity.

[0026] Traditional methods rely solely on individual device parameters or data from independent environmental sensors, failing to establish a correlation between device operating status and environmental influences. This solution characterizes environmental features through the operating parameters of a cluster of devices within an environmental area, eliminating the need for additional environmental sensors and reducing hardware costs. Furthermore, by fusing device parameters with environmental characteristics, it can simultaneously identify both device-specific faults and latent faults caused by environmental factors, such as voltage fluctuations due to regional electromagnetic interference or temperature anomalies caused by insufficient heat dissipation.

[0027] This application can comprehensively identify both overt faults and latent faults related to environmental factors in cable branch boxes, such as insulation aging caused by long-term environmental humidity or overheating caused by regional power grid load fluctuations. By constructing environmental characteristic data through equipment cluster operating parameters, it achieves quantitative analysis of the impact of environmental factors, solving the problem of missed latent faults caused by traditional single-parameter monitoring and independent environmental data collection, and improving the accuracy and reliability of fault detection.

[0028] This application further proposes to obtain the environmental change trajectory of a cable branch box within a specified time period; based on the environmental change trajectory, determine at least one environmental area where the cable branch box is located, wherein the environmental area is a preset environmental zone; for each determined environmental area, obtain the operating parameter data of all cable branch boxes located in the environmental area within the specified time period to determine the operating parameter characteristics of the environmental area; and determine the environmental characteristic data of the cable branch box according to the operating parameter characteristics corresponding to each environmental area involved in the environmental change trajectory.

[0029] The environmental change trajectory refers to the spatial movement path of the cable branch box within a specified time period due to adjustments in its installation location or changes in environmental conditions. This can be achieved using geographic grid partitioning algorithms or GPS positioning technology to dynamically capture changes in the environmental area experienced by the equipment. The environmental area refers to an independent spatial partition pre-defined based on temperature and humidity ranges, power grid load levels, or electromagnetic interference intensity. This can be defined using a geographic information system or regional coding rules to distinguish the differences in the impact of different environments on equipment operation. Operating parameter characteristics refer to the statistical representation of operating parameter data from all cable branch boxes within the same environmental area. This can be achieved using average calculation or weighted summation algorithms to eliminate random errors in data from individual devices and improve the representativeness of the regional environmental characteristics.

[0030] Specifically, a geographic grid division algorithm is used to track the movement path of cable distribution boxes within a specified time period, identifying multiple preset environmental zones they pass through. For each environmental zone, current, voltage, temperature, and humidity data for all cable distribution boxes within that zone are collected, and their average values ​​are calculated as the operational parameter characteristics of that zone. For example, when equipment moves from a low-load zone to a high-load zone, the average current value of multiple devices in the high-load zone is higher than that in the low-load zone, and this difference is precisely quantified.

[0031] Subsequently, the operational parameter characteristics of all environmental areas involved in the equipment trajectory are integrated to form environmental characteristic data covering all time periods and multiple regions. Therefore, the risk of insulation aging in high-temperature and high-humidity areas or the potential for joint overheating in high-load areas can be accurately identified through correlation analysis of the characteristic data of the corresponding areas.

[0032] Traditional methods only collect static environmental parameters at fixed installation locations, failing to reflect the multi-regional environmental impact caused by dynamic equipment movement. For example, in existing technologies, if a cable branch box is relocated for construction, the high humidity characteristics of its new environmental area may be missed, leading to unforeseen insulation aging faults. This solution, however, uses dynamic trajectory tracking to pinpoint all traversed environmental areas and combines this with statistical analysis of the operating parameters of clustered equipment within those areas. This avoids single-point data bias and achieves quantitative superposition of multi-regional environmental impacts, significantly improving the comprehensiveness of latent fault monitoring.

[0033] This application addresses the problem of inaccurate characteristic data collection for cable distribution boxes due to dynamic changes in the environment. By statistically analyzing the parameters of clustered devices, it eliminates the random interference from individual device data and accurately distinguishes the differences in the impact of different environmental areas. For example, the declining insulation performance of equipment in high-humidity areas can be promptly identified by an abnormal increase in the average humidity data of multiple devices in that area, avoiding missed fault detection due to overlooked regional characteristics.

[0034] Meanwhile, the design that integrates dynamic trajectory with multi-regional features enables environmental feature data to fully reflect the equipment's operating environment throughout the entire time period, providing more comprehensive input parameters for the fault detection model and enhancing the ability to identify latent faults.

[0035] This application further proposes using a geographic grid partitioning algorithm to determine at least one environmental area where a cable branch box is located based on environmental change trajectories.

[0036] The geographic grid partitioning algorithm divides physical space into standardized grid units with fixed latitude and longitude ranges. Each grid unit corresponds to a preset environmental parameter statistical dimension, such as storing historical temperature and humidity ranges or grid load benchmark values ​​for the region. This algorithm uses a coordinate matching mechanism to associate device trajectory coordinates with grid units, solving the problem of ambiguous boundaries in traditional regional partitioning.

[0037] Environmental change trajectory refers to the coordinate sequence formed by the location of the cable distribution box within a specified time period. This could be latitude and longitude data collected based on GPS or IoT positioning technology, reflecting the movement path of the equipment in physical space. This trajectory data serves as the input source for grid matching, ensuring that the area division matches the actual operating path of the equipment.

[0038] Specifically, the geographic grid division algorithm divides the target area into multiple standardized units using a preset grid size, such as dividing a city area into 1km × 1km grids. After the environmental change trajectory of the cable distribution box is generated, its coordinate sequence is input into the algorithm. The grid unit where the equipment is located is determined by comparing the coordinates with the grid boundary. For example, when the equipment trajectory coordinates fall within the range of 116.3°–116.4°E longitude and 39.9°–40.0°N latitude, it is matched to the corresponding grid unit.

[0039] Environmental characteristic data for each grid cell is obtained by statistically analyzing the operating parameters of all cable branch boxes within that grid, such as calculating the average temperature or voltage fluctuation range of the equipment within that grid. This division method provides clear quantitative boundaries for environmental areas, avoiding the bias in environmental characteristic data caused by ambiguous attribution when equipment is located at traditional administrative boundaries.

[0040] Traditional methods rely on static administrative divisions or single environmental sensors for location, resulting in coarse-grained regional divisions and unclear boundaries. For example, administrative divisions might group a humid, low-lying area where the equipment is located with a dry, high-altitude area, leading to mixed environmental characteristic data. This solution, however, uses a geographic grid division algorithm to achieve fine-grained regional division using standardized grid units, ensuring uniform environmental conditions within each grid without requiring additional sensor hardware. For instance, when the equipment trajectory crosses multiple grids, environmental characteristic data can be weighted based on the dwell time in each grid, overcoming the data fusion difficulties caused by uneven regional divisions in traditional methods.

[0041] This application can improve the accuracy of environmental area division and avoid distortion of environmental feature data caused by blurred boundaries or differences in the internal environment of the area. For example, when a cable branch box is located near a road boundary, it can be accurately matched to the corresponding grid cell, accurately reflecting the impact of vehicle electromagnetic interference on the equipment, thereby providing reliable environmental correlation data for the fault detection model and improving the accuracy of hidden fault identification.

[0042] This application further proposes that obtaining environmental characteristic data of cable branch boxes also includes determining the weights of the operating parameter characteristics corresponding to each environmental area. Based on the obtained operating parameter characteristics corresponding to each environmental area, determining the environmental characteristic data of cable branch boxes includes weighted summation of the obtained operating parameter characteristics of each environment to obtain the environmental characteristic data of cable branch boxes.

[0043] The environmental region refers to the spatial range formed by a pre-defined geographic grid. Specifically, a geographic grid partitioning algorithm can be used to divide the environment where the cable branch box is located into multiple independent regions, each corresponding to different environmental characteristic attributes. The weighted summation refers to the linear superposition of the operating parameter characteristics of different environmental regions according to weight coefficients. Specifically, this can be achieved by calculating the sum of the products of the operating parameter characteristics of each region and their corresponding weights. The weight coefficients reflect the degree of difference in the impact of each region on equipment failure.

[0044] Specifically, in the multiple environmental areas covered by the cable distribution box's movement trajectory, the weight is first determined based on the proportion of the device's dwell time in each area to the total monitoring time. For example, if a certain area has a frequency of 30% of the total monitoring time, it is assigned a weight of 0.3. Furthermore, the weight is adjusted based on the proportion of the number of cable distribution boxes in that area to the total number of monitored devices. For example, if the number of devices in a certain area reaches 20%, the weight is increased to 0.4.

[0045] The operating parameters of each region are then normalized, for example, temperature and current parameters are converted to standard values. Finally, the normalized feature data of each region are multiplied by their respective weights and summed to form comprehensive feature data characterizing the overall environmental impact. This process ensures that anomalous parameters in areas with high dwell frequency or dense equipment occupy a higher proportion in the comprehensive data. For example, in a high-humidity region with a weighting coefficient of 0.6, the anomalous feature of 85% of the average humidity value contributes 51% of the humidity feature value after weighting, significantly higher than the 17% contribution of the region with a weight of 0.2.

[0046] Traditional methods use arithmetic averaging or direct aggregation to process environmental data, resulting in the dilution of key features from high-impact areas by low-impact data. For example, if a cable distribution box spends 60% of its time in a high-temperature zone, traditional methods, by averaging its high-temperature features with low-temperature data, weaken the high-temperature anomaly by 50%. This proposed solution, however, uses weighted allocation to increase the proportion of high-temperature features to 60%, fully preserving the indicative role of anomaly parameters in fault detection. This data processing method based on quantified weights effectively solves the problem of environmental feature distortion caused by equalization.

[0047] This application achieves a quantitative distinction of the degree of influence of different environmental regions, avoiding the dilution of data in key areas by low-relevance areas, and enabling environmental feature data to accurately reflect the true state of the environment in which the cable branch box is located. Furthermore, the weighted summation of comprehensive feature data can highlight abnormal parameters in high-weight areas, providing more accurate input for the fault detection model, thereby improving the reliability of identifying latent faults such as valve core jamming and insulation aging.

[0048] This application further proposes to determine the weights of the operating parameter characteristics corresponding to each environmental region, including, for each environmental region, determining the weights of the operating parameter characteristics corresponding to the environmental region based on the frequency of occurrence of the environmental region in the environmental change trajectory, the total number of environmental regions involved in the environmental change trajectory, and the proportion of the number of cable branch boxes in the environmental region to the total number of monitoring during a specified period.

[0049] The frequency of occurrence of an environmental area in the environmental change trajectory refers to the proportion of time that the cable branch box is in that environmental area within a specified period. This can be achieved by statistically analyzing the timestamp distribution of equipment location data. This parameter is used to quantify the cumulative effect of environmental factors on equipment.

[0050] The total number of environmental areas involved in the environmental change trajectory refers to the number of different preset zones covered by the cable branch box during a specified time period. This can be achieved by counting the number of area numbers output by the geographic grid partitioning algorithm. This parameter is used to eliminate weight bias caused by differences in trajectory coverage. The ratio of the number of cable branch boxes in an environmental area to the total number of monitored devices during a specified time period refers to the ratio of the number of monitored devices in that environmental area to the total number of devices in the entire monitoring network. This can be achieved by matching device location data with the monitoring device list. This parameter is used to ensure the statistical reliability of operational parameter characteristics.

[0051] Specifically, by calculating the ratio of the frequency of occurrence of an environmental area to the total number of environmental areas, the inconsistency in weight calculation benchmarks caused by different equipment movement ranges can be eliminated. For example, if an equipment is in a high-humidity area for 15 days within a 30-day monitoring period and its trajectory covers a total of 3 areas, the frequency weight of that area is 15 / 30, and the total number of areas has a weight factor of 1 / 3. Multiplying these two factors yields the initial weight value. Further considering the proportion of equipment in that area, if the number of monitored equipment in that area accounts for 20% of the total, the final weight is determined by both the initial weight and the proportion value. This multi-dimensional calculation method assigns higher weights to environmental characteristics such as long exposure time, concentrated area coverage, and high sample density, thereby more accurately reflecting the actual impact of the environment on equipment failure during the weighted summation process.

[0052] Traditional methods often rely on manual experience to set fixed weights or consider only a single-dimensional parameter, such as assigning weights based solely on the historical frequency of regional faults, ignoring the impact of equipment exposure duration, regional coverage, and sample size on data validity. This solution constructs a dynamic weight allocation mechanism by integrating three quantitative dimensions: time cumulative effect, regional distribution balance, and sample statistical reliability. This makes the generation process of environmental characteristic data more closely reflect the multi-factor coupling effects in the actual operation scenario of cable branch boxes.

[0053] This application addresses the problem of insufficient fault identification accuracy caused by unreasonable weight allocation, enabling environmental feature data to accurately characterize the impact of different regions on equipment faults. For example, in the detection of insulation aging faults in cable branch boxes, environmental features with high frequency of occurrence and a large proportion of equipment in high-humidity areas are given higher weights. The corresponding humidity parameter features dominate the model input, thereby improving the model's accuracy in identifying humidity-related latent faults and avoiding missed or false detections caused by interference from low-weighted regional parameters.

[0054] This application further proposes to obtain the operating parameter data of all cable branch boxes in the identified environmental areas within a specified time period, in order to determine the operating parameter characteristics of the environmental areas, including: calculating the average value of the operating parameter data of each cable branch box in the environmental area as the operating parameter characteristics of the environmental area.

[0055] The environmental region refers to the geographical area where the cable distribution box is located. Specifically, a geographic grid partitioning algorithm can be used to create a pre-defined environmental partition, which is used to divide spatial areas with similar environmental conditions. This feature, through geographic spatial partitioning, ensures that equipment within the same area is affected by uniform environmental factors, providing a spatial benchmark for subsequent data aggregation.

[0056] Operating parameter data includes at least one of current, voltage, temperature, and humidity, which can be collected in real time by sensors to reflect the operating status of the cable branch box. This data is used to characterize the correlation between equipment operating status and the environment, providing raw input for environmental feature extraction. The average value refers to the arithmetic mean of the operating parameter data of all cable branch boxes within the same environmental area, which can be achieved by summing and dividing by the number of devices. This calculation method eliminates abnormal fluctuations in individual devices and extracts feature values ​​reflecting common environmental impacts.

[0057] Specifically, when determining the operational parameter characteristics of a specific environmental area, the first step is to obtain the operational parameter data of all cable distribution boxes within that area over a specified period. For example, within a geographic grid, there are five cable distribution boxes with temperature monitoring values ​​of 28℃, 29℃, 30℃, 27℃, and 200℃, respectively. The 200℃ value of the fourth device is an anomaly caused by a sensor false alarm. By calculating the average, the sum of the temperature data from the five devices is divided by the number of devices, resulting in 62.8℃. At this point, the anomaly significantly raises the average value, potentially indicating the need to incorporate a data cleaning mechanism, such as removing data exceeding a preset threshold before recalculating the average. After correction, the average temperature of the four normal devices is 28.5℃, which accurately reflects the trend of the environmental temperature's impact on the devices. This type of calculation avoids misjudgments of environmental characteristics caused by single device failures or short-term disturbances, enabling the model to identify latent faults caused by long-term environmental factors (such as overheating of joints due to excessive regional power grid load).

[0058] Traditional methods typically use data from a single device or simply list device parameters within a region. For example, existing technologies might select the operating parameters of a specific cable distribution box within a region as environmental characteristics. When that device experiences an individual fault, the environmental characteristics can deviate significantly from the true values. This solution, however, decouples environmental characteristics from individual device states by calculating the average value of data from a cluster of devices. This reduces reliance on the quality of data from a single device and enhances the robustness of the characteristic data against interference. Furthermore, compared to solutions that introduce complex filtering algorithms or increase the number of sensors, this method requires only basic arithmetic operations, meeting the low-cost requirements of large-scale cable distribution box deployment scenarios.

[0059] This application eliminates interference from abnormal operating parameters caused by individual equipment differences or occasional sensor malfunctions, making the characteristic data of environmental areas more stable and reliable. For example, in areas with high humidity, when the insulation resistance values ​​of multiple cable branch boxes all show a slow downward trend, the gradual impact of environmental humidity on insulation materials can be captured by calculating the average value, thus triggering an early warning before the insulation layer completely ages and causes a short circuit. Simultaneously, this solution establishes a unified calculation benchmark for characteristic data from different environmental areas, enabling the model to quickly compare the degree of difference in environmental influence between areas and accurately identify latent faults caused by environmental factors such as uneven regional power grid load and differences in electromagnetic interference.

[0060] This application further proposes that, when determining the operating parameter characteristics of an environmental area, the operating parameter data of each cable branch box in the environmental area be weighted and summed to obtain the operating parameter characteristics of the environmental area.

[0061] The weights of operating parameter data refer to the numerical values ​​assigned based on the varying degrees of importance of different operating parameters to cable branch box faults. This can be achieved using expert experience or historical data analysis. For example, a higher weight can be assigned to temperature based on its correlation with insulation aging faults, while a second-highest weight can be assigned to current based on its correlation with overload faults. The weighted summation of operating parameter data involves multiplying the data of a specific operating parameter from all cable branch boxes within the same environmental area by its corresponding weight and then summing the results. This can be achieved using matrix operations or parameter-by-parameter iterative calculations. For example, for the temperature parameter, the temperature values ​​of each device within the area can be multiplied by their respective temperature weights and then summed to form the temperature characteristic value for that area.

[0062] Specifically, after acquiring the operating parameter data of all cable branch boxes within the environmental area during a specified time period, a weight is first assigned to each operating parameter. The weight value is determined based on the correlation strength between the parameter and the target latent fault. For example, the temperature parameter, directly related to insulation aging, is assigned a weight of 0.6, and the current parameter, related to overload faults, is assigned a weight of 0.3. Subsequently, the temperature data of all equipment in the area are multiplied by 0.6 and summed to obtain the temperature characteristic component of the area; the current data are multiplied by 0.3 and summed to obtain the current characteristic component. Finally, the characteristic components of each parameter are combined to form the operating parameter characteristics of the environmental area. This process avoids the problem of key parameters being diluted by secondary parameters in traditional averaging calculations, making temperature anomalies or current fluctuations more prominent in the regional characteristics.

[0063] Existing methods for calculating environmental operating parameter characteristics typically employ simple arithmetic averages or extract only a single parameter, failing to consider the varying contributions of different operating parameters to fault detection. For example, traditional averaging methods treat temperature and humidity parameters equally, leading to the risk of high temperatures being neutralized by normal humidity values. This proposed solution, however, uses parameter weighting to ensure that temperature anomalies dominate the calculation of regional characteristics, directly reflecting potential insulation aging hazards. Furthermore, existing technologies using fixed weights or equipment-dimensional weights cannot achieve precise matching between parameter dimensions and fault types. This solution dynamically sets weights based on parameter-fault correlation, better meeting the needs of multi-parameter monitoring scenarios for cable branch boxes.

[0064] This application addresses the problem of inaccurate regional features caused by the failure to differentiate the importance of operating parameters, enabling environmental regional operating parameter features to accurately reflect key fault causes. For example, in the detection of latent faults such as joint overheating, because temperature parameters are given high weight, even if individual devices in the area have abnormal humidity, the regional features are still dominated by temperature data, avoiding interference from secondary parameters in fault identification. Simultaneously, this scheme strengthens the influence of parameters highly correlated with the target fault through weighted calculations, providing more discriminative input data for the fault detection model and improving the detection sensitivity for latent faults such as insulation aging and increased contact resistance.

[0065] Example 2

[0066] This application further proposes a computing device including at least one processor and a memory, the memory storing instructions that, when executed by at least one processor, cause at least one processor to execute a fault detection method based on multidimensional data fusion.

[0067] The processor refers to the hardware unit that executes computational logic. It can be implemented using a multi-core central processing unit or an embedded microcontroller. It is used to process operating parameters and environmental characteristic data in real time, perform weighted calculations and model inference, and ensure that computing power is maintained in low-power mode. The memory refers to the physical medium that stores program instructions and data. It can be implemented using flash memory chips or solid-state drives. It is used to solidify the fault detection algorithm process, save historical operating parameters and environmental characteristic data, and support the retrieval of preset detection logic in offline mode.

[0068] Specifically, the processor continuously acquires the current, voltage, and temperature parameters of the cable branch boxes by reading instructions stored in memory, and determines their location by combining this with environmental change trajectories. For each environmental region, the processor calculates the average operating parameters of all cable branch boxes within that region, forming environmental characteristic data. Furthermore, the processor dynamically weights the operating parameter characteristics based on the frequency of occurrence of the environmental region and the distribution density of the cable branch boxes, inputting the weighted result into the fault detection model to identify the initial characteristics of valve core jamming or spring fatigue. When power supply is limited, the processor switches to a low-power mode, prioritizing the execution of the core detection algorithm, while memory maintains a cache of critical data to prevent the detection process from terminating due to power interruption.

[0069] Existing solutions rely on variable reluctance generators to provide power during discharge conditions, which cannot maintain energy storage during low-load or no-discharge phases, leading to monitoring interruptions. This solution utilizes a collaborative architecture of processor and memory to maintain basic detection functions using pre-stored instructions when there is no external power supply. The processor dynamically adjusts its computational load based on the power supply status, ensuring continuous acquisition and analysis of key parameters.

[0070] This application enables continuous monitoring of hidden faults in safety valves under power fluctuation or low load scenarios. By dynamically weighting and fusing environmental characteristics and operating parameters, it effectively identifies the sealing surface adhesion characteristics caused by valve core jamming and the pressure drift trend caused by spring fatigue, thus avoiding monitoring blind spots caused by power outages.

[0071] Example 3

[0072] This application further proposes a non-transitory machine-readable storage medium storing executable instructions that, when executed, cause the machine to perform the method.

[0073] Non-transitory machine-readable storage media refers to physical storage media capable of long-term data retention without relying on continuous power supply. Specifically, it can be implemented using solid-state drives (SSDs), flash memory chips, or read-only memory (ROMs). Its key technical feature is that it retains its stored content even after power failure, preventing program instructions from becoming unrecoverable due to power interruption. The "method for machine execution" refers to the binding relationship between the algorithm logic embedded in the storage medium and the hardware execution unit. This can be achieved through the boot loading mechanism of an embedded system or firmware pre-programming. Its function is to ensure that the system can automatically reload the detection process after power is restored, maintaining the continuity of fault identification and early warning functions.

[0074] Specifically, the non-transitory storage medium persistently stores the instruction set containing the environmental feature data fusion algorithm, thus fully preserving the data processing logic during power outages. When power is restored, the processor directly reads the instructions from the storage medium, reconstructs the correlation analysis between the cable branch box operating parameters and environmental features, and continues to execute the fault detection closed loop based on the preset model. This mechanism does not rely on real-time power generation to maintain the operation of the storage unit, making it particularly suitable for long-term low-load or non-discharge scenarios, and solving the problem of algorithm execution interruption caused by energy depletion in traditional solutions.

[0075] Existing technologies use variable reluctance generators to power the monitoring system. However, without discharge, the system cannot continuously replenish power, causing the monitoring function to malfunction after the energy storage module is depleted. This solution, on the other hand, uses non-volatile storage media to solidify the detection algorithm, maintaining program integrity during power outages and immediately restarting the detection process upon power restoration. This eliminates the need for continuous power supply from external generators, achieving full-cycle coverage of the monitoring function.

[0076] In scenarios with unstable or intermittent power supply, this application can ensure the integrity and recoverability of the cable branch box fault detection algorithm, avoid interruption of the environmental feature data fusion process due to insufficient power supply, thereby maintaining the continuous operation capability of the fault early warning system and effectively reducing the safety risk of latent faults developing into manifest faults.

[0077] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A fault detection method for a cable branch box, characterized in that... include: Obtain the operating parameter data of the cable branch box, wherein the operating parameter data includes at least one of current, voltage, temperature, and humidity; The environmental characteristic data of the cable branch box is obtained. The environmental characteristic data includes the environmental characteristic data of at least one environmental area in which the cable branch box is located during a specified time period. The environmental characteristic data of each environmental area is characterized by the operating parameter data of all cable branch boxes in the environmental area during the specified time period. as well as The acquired operating parameter data and environmental characteristic data of the cable branch box are provided to the cable branch box fault detection model for fault identification and early warning.

2. The fault detection method for cable branch boxes according to claim 1, characterized in that, The acquisition of environmental characteristic data of the cable branch box includes: Obtain the environmental change trajectory of the cable branch box within the specified time period; Based on the environmental change trajectory, at least one environmental area where the cable branch box is located is determined, and the environmental area is a preset environmental zone. For each identified environmental area, obtain the operating parameter data of all cable branch boxes in the environmental area during the specified time period to determine the operating parameter characteristics of the environmental area; Based on the operating parameter characteristics corresponding to each environmental region involved in the environmental change trajectory, the environmental characteristic data of the cable branch box are determined.

3. The fault detection method for cable branch boxes according to claim 2, characterized in that, Based on the aforementioned environmental change trajectory, determining at least one environmental area where the cable branch box is located includes: A geographic grid partitioning algorithm is used to determine at least one environmental area where the cable branch box is located based on the environmental change trajectory.

4. The fault detection method for cable branch boxes according to claim 2 or 3, characterized in that, The process of obtaining the environmental characteristic data of the cable branch box also includes: Determine the weights of the operational parameter characteristics corresponding to each environmental region. The environmental characteristic data of the cable branch box, determined based on the operational parameter characteristics corresponding to each acquired environmental area, includes: The environmental characteristic data of the cable branch box are obtained by weighted summation of the operating parameter characteristics of each environmental area.

5. The fault detection method for cable branch boxes according to claim 4, characterized in that, The weights for determining the operational parameter characteristics corresponding to each environmental region include: For each environmental region, the weight of the operating parameter characteristics corresponding to the environmental region is determined based on the frequency of occurrence of the environmental region in the environmental change trajectory, the total number of environmental regions involved in the environmental change trajectory, and the proportion of the number of cable branch boxes in the environmental region to the total number of monitored areas during the specified time period.

6. The fault detection method for cable branch boxes according to claim 2, characterized in that, For each identified environmental zone, the operating parameter data of all cable branch boxes within that environmental zone during the specified time period are obtained to determine the operating parameter characteristics of the environmental zone, including: Calculate the average value of the operating parameter data of each cable branch box in the environmental area, and use it as the operating parameter characteristic of the environmental area.

7. The fault detection method for cable branch boxes according to claim 2, characterized in that, Each of the aforementioned operating parameter data has a corresponding weight. For each determined environmental area, the operating parameter data of all cable branch boxes located within the environmental area during the specified time period are obtained to determine the operating parameter characteristics of the environmental area, including: The operating parameter data of each cable branch box in the environmental area are weighted and summed to obtain the operating parameter characteristics of the environmental area.

8. A computing device, characterized in that, include: At least one processor; as well as A memory storing instructions that, when executed by the at least one processor, cause the at least one processor to perform the fault detection method for a cable branch box as described in any one of claims 1-7.

9. A non-transitory machine-readable storage medium, characterized in that, The non-transitory machine-readable storage medium stores executable instructions that, when executed, cause the machine to perform the fault detection method for the cable branch box as described in any one of claims 1-7.

Images