A state intelligent monitoring and early warning integrated device of power distribution equipment

Abstract

The application discloses a kind of state intelligent monitoring and early warning integration device of power distribution equipment, it is related to equipment state monitoring technical field, the device includes: component information acquisition module, for the component splitting of power distribution equipment three-dimensional model, obtain component model, component service length and component position;Correlation analysis module, for based on component model, abnormal correlation analysis is carried out to equipment state factor, and sensitive state factor is obtained;Sensitive factor statistical module, for retrieving component health service sample set, and the standard interval of sensitive state factor is counted;Regional clustering module, for the regional clustering of power distribution equipment three-dimensional model, obtain first regional state standard interval until Nth regional state standard interval;Abnormal early warning module, for when any regional state monitoring value does not satisfy standard interval, execute power distribution equipment abnormal early warning.The application solves the technical problem of poor monitoring and early warning accuracy of power distribution equipment in the prior art.

Description

Technical Field

[0001] This invention relates to the field of equipment condition monitoring technology, and specifically to an integrated device for intelligent monitoring and early warning of the condition of power distribution equipment. Background Technology

[0002] With the development of intelligent power systems, the condition monitoring technology for power distribution equipment has gradually adopted sensor networks and data analysis methods. Existing technologies primarily monitor the condition factors of equipment components independently, such as temperature, current, and vibration, by setting static thresholds. An early warning is triggered when a single parameter exceeds the preset threshold. However, this approach has significant drawbacks: First, static thresholds cannot adapt to dynamic changes in actual equipment operation (such as load fluctuations, differences in ambient temperature and humidity, and component aging), leading to increased false alarms or missed alarms. For example, the normal state range of the same component under high-temperature, high-load conditions may be significantly wider than under low-temperature, low-load conditions, but traditional methods do not establish a dynamic correlation mechanism between the threshold and the operating environment. Second, existing technologies use independent components as monitoring units, performing state calculations and early warning judgments for each individual component, resulting in a large amount of redundant analysis and significantly increasing the system's computational burden. Especially in large-scale power distribution equipment, the independent monitoring of a massive number of components will lead to a decrease in real-time performance, making it difficult to meet the demand for rapid fault early warning. Summary of the Invention

[0003] This application provides an integrated intelligent monitoring and early warning device for the status of power distribution equipment, which addresses the technical problems of independent and rigid status monitoring and low early warning efficiency of power distribution equipment in the prior art.

[0004] In view of the above problems, this application provides an integrated intelligent monitoring and early warning device for the status of power distribution equipment, the device comprising:

[0005] The component information acquisition module is used to decompose the three-dimensional model of the power distribution equipment in a spatial coordinate system to obtain the component model, component service life and component location;

[0006] The correlation analysis module is used to perform abnormal correlation analysis on the equipment status factors based on the component model to obtain sensitive status factors;

[0007] The sensitivity factor statistics module is used to retrieve the component health service sample set and statistically analyze the standard interval of the sensitivity state factor, based on the power distribution equipment operating environment parameters, power distribution load parameters, the service life of the component, the component model and the sensitivity state factor.

[0008] The region clustering module is used to perform region clustering on the three-dimensional model of the power distribution equipment based on the sensitive state factor standard interval and the component location, to obtain the first region state standard interval up to the Nth region state standard interval.

[0009] The anomaly warning module is used to execute an anomaly warning for power distribution equipment when the status monitoring value of any area does not meet the status standard range of the first area up to the status standard range of the Nth area.

[0010] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0011] This application proposes an integrated intelligent monitoring and early warning device for power distribution equipment. By dynamically generating standard intervals of state factors adapted to operating conditions and optimizing the division of monitoring units, it significantly improves the accuracy of state warnings and the efficiency of system computing power utilization. Compared with traditional methods, the technical solution provided in this application significantly overcomes the shortcomings of rigid static thresholds and redundant analysis of independent components. Specifically, firstly, it dynamically retrieves a health sample set based on component model, service life, environmental parameters, and load parameters. Through box plot analysis, it generates flexible standard intervals of sensitive state factors, enabling state thresholds to adapt to equipment aging, load fluctuations, and environmental changes, effectively avoiding false alarms or missed alarms caused by differences in operating conditions. Secondly, through spatial clustering, it aggregates neighboring components with similar standard intervals into a unified monitoring area, performing state judgment on an area-by-area basis, greatly reducing the number of monitoring points that need to be analyzed independently. This not only reduces the computing power burden caused by the independent calculation of massive components but also enhances the typicality of cluster state characteristics by utilizing the centroid interval of the area.

[0012] This application achieves the technical effect of dynamic, accurate, and lightweight early warning for power distribution equipment under adaptive operating conditions, providing reliable support for the intelligent operation and maintenance of complex power distribution equipment. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a schematic diagram of the structure of an integrated intelligent monitoring and early warning device for the status of power distribution equipment provided in an embodiment of this application;

[0015] Figure 2 A schematic diagram illustrating the execution steps of the sensitive factor statistics module in an integrated intelligent monitoring and early warning device for power distribution equipment provided in this application embodiment;

[0016] The components represented by each number in the attached diagram are explained below:

[0017] The system includes a component information acquisition module (100), a correlation analysis module (200), a sensitivity factor statistics module (300), a region clustering module (400), and an anomaly early warning module (500). Detailed Implementation

[0018] This application provides an integrated intelligent monitoring and early warning device for the status of power distribution equipment, which addresses the technical problems of independent and rigid status monitoring and low early warning efficiency of power distribution equipment in the prior art.

[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0020] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0021] Example 1, as Figure 1 As shown, this application provides an integrated intelligent monitoring and early warning device for the status of power distribution equipment, wherein the device includes:

[0022] The component information acquisition module 100 is used to decompose the three-dimensional model of the power distribution equipment in a spatial coordinate system to obtain the component model, component service life and component location.

[0023] Traditional power distribution equipment condition monitoring relies on the collection of discrete component information and lacks spatial topological correlation. When it is necessary to combine spatial distribution to analyze the condition threshold, an additional complex data mapping operation is required, which not only increases the preprocessing burden, but also makes it difficult to support subsequent regional monitoring due to the lack of coordinate information.

[0024] In this embodiment, three-dimensional mesh model data (such as OBJ format three-dimensional model files) of the power distribution equipment are extracted from the design data of the power distribution equipment. The KDTree algorithm is used for nearest neighbor search, and connected component analysis is combined to identify independent components in the model. For each identified component, the component model and service life are obtained from a preset equipment description, and the arithmetic mean coordinates of all vertices of each component are calculated as the component's position. Finally, a structured list is output, containing the model, service life (in months), and position (three-dimensional coordinate set) of each component.

[0025] By directly disassembling components using the spatial coordinate system of a 3D model and binding them with model number, service duration, and location information, a structured component database with spatial topological attributes is constructed. This module provides underlying spatial association support for dynamic threshold generation and region clustering, eliminates the need for secondary alignment of multi-source data, and significantly improves the integrity of preliminary data for subsequent state analysis.

[0026] The correlation analysis module 200 is used to perform abnormal correlation analysis on the equipment status factors based on the component model to obtain sensitive status factors.

[0027] In existing technologies, monitoring all states equally fails to distinguish the differences in the sensitivity of different component failures. This easily leads to redundant monitoring of non-critical factors, diluting computing resources, and low-correlation factors interfere with the accuracy of early warnings, resulting in low efficiency in fault tracing.

[0028] In this embodiment of the application, the correlation analysis module 200 includes the following execution steps:

[0029] Based on the component model, a component fault dataset is collected, wherein any component fault data in the component fault dataset includes a set of state attributes that do not meet the preset state standard range.

[0030] Based on the component fault dataset, the correlation degree of several state attributes is calculated, wherein the correlation degree of state attribute anomalies is equal to the ratio of the number of fault data entries triggered by state attributes to the total number of data entries in the component fault dataset.

[0031] Based on the abnormal correlation of the aforementioned state attributes, state attributes with an abnormal correlation greater than or equal to the abnormal correlation threshold are extracted and added to the sensitive state factor.

[0032] In this embodiment, a pre-defined power distribution equipment fault database is connected, and historical fault records are read. Each record includes a component model, a status attribute value, and a preset status standard range. Then, based on the input component model, an SQL query is executed to retrieve all fault data matching that model. Each fault data record is defined as a record whose status attribute value does not meet the preset status standard range; for example, the component temperature exceeds the upper or lower limit of the standard. Finally, a component fault dataset is obtained, where each data entry contains a set of status attributes that do not meet the standard range.

[0033] For each state attribute, such as temperature or current, calculate the number of data entries that trigger a fault, i.e., the number of records where the attribute value does not meet the standard range. The state attribute anomaly correlation degree = number of state attribute fault-triggered data entries ÷ total number of data entries in the component fault dataset. For example, if the number of state attribute fault-triggered data entries is 50 and the total number of data entries in the component fault dataset is 100, then the state attribute anomaly correlation degree = 50 ÷ 100 = 0.5. The higher the state attribute anomaly correlation degree, the more likely that state attribute is to trigger a fault.

[0034] Repeat this calculation for all state attributes and add them to the sensitive state factor.

[0035] This module filters fault datasets based on component models and extracts sensitive state factors by quantifying the correlation between abnormal state attributes. It precisely focuses on factors with high fault contribution rates, eliminates noisy monitoring items, and provides core input for dynamic threshold calculation, significantly improving the effectiveness and relevance of subsequent analyses.

[0036] The sensitivity factor statistics module 300 is used to retrieve a sample set of healthy service conditions of components and to statistically analyze the standard intervals of sensitivity factors, based on constraints such as the operating environment parameters of the power distribution equipment, the power load parameters, the service life of the components, the model of the components, and the sensitivity state factors. For example... Figure 2 As shown, the sensitivity factor statistics module 300 includes the following execution steps:

[0037] A first constraint condition is constructed using the operating environment parameters of the power distribution equipment, the power distribution load parameters, and the service life of the components as constraints.

[0038] The first constraint condition is constructed using the operating environment parameters of the power distribution equipment, the power distribution load parameters, and the service life of the components as constraints, including:

[0039] Using the component model as a constraint, and based on the sensitive state factor, the operating environment parameters of the power distribution equipment, the power distribution load parameters, and the service life of the component are weighted and distributed to obtain the influence weight distribution results;

[0040] Specifically, using the component model as a constraint, and based on the sensitive state factor, a weighted distribution is applied to the operating environment parameters of the power distribution equipment, the power distribution load parameters, and the service life of the component to obtain the influence weight distribution results, including:

[0041] The operating environment parameters of the power distribution equipment, the power distribution load parameters, and the service life of the components are set as a set of limiting condition attributes, and the first limiting condition attribute is extracted.

[0042] Through the user terminal, the tolerance deviation of the limiting condition attribute is configured, wherein the tolerance deviation of the limiting condition attribute includes a first tolerance deviation of the limiting condition attribute and other tolerance deviations of the limiting condition attribute.

[0043] The set of abnormal attribute changes of sensitive state factors is collected, with the first constraint attribute deviation being greater than or equal to the first constraint attribute fault tolerance deviation and the other constraint attribute deviations being less than the other constraint attribute fault tolerance deviations.

[0044] Calculate the mean of the set of abnormal attribute changes of the sensitive state factors, set it as the first constraint condition attribute influence parameter, and add it to the set of constraint condition attribute influence parameters.

[0045] Calculate the ratio of the first constraint condition attribute influence parameter to the sum of the set of constraint condition attribute influence parameters, set it as the first constraint condition attribute weight, and add it to the influence weight distribution result;

[0046] The operating environment parameters of the power distribution equipment, the power distribution load parameters, and the service life of the components are normalized to obtain the baseline parameters for the limiting conditions.

[0047] Construct deviation assessment rules:

[0048] Step 1: Calculate the magnitude of the constraint deviation between the constraint condition sample parameters and the constraint condition reference parameters;

[0049] Step 2: Calculate the weighted average of the deviation modulus of the constraint conditions based on the influence weight distribution results to obtain the deviation evaluation value;

[0050] When the deviation evaluation value is greater than or equal to the deviation evaluation threshold, it is considered that the first constraint condition is not met.

[0051] When the deviation evaluation value is less than the deviation evaluation threshold, it is considered that the first constraint condition is met.

[0052] A second constraint condition is constructed using the component model and the sensitive state factor as constraints;

[0053] Retrieve a first component health service sample set that satisfies the first constraint condition;

[0054] From the first component health service sample set, select the component health service sample set that satisfies the second restriction condition, wherein any component health service sample in the component health service sample set includes a sensitive state factor record value.

[0055] Based on the component health service sample set, the recorded values ​​of the sensitive state factors are analyzed by the same attribute box plot to obtain the recorded values ​​of the sensitive state factors of the first attribute box up to the recorded values ​​of the sensitive state factors of the Mth attribute box.

[0056] Traverse the first attribute box sensitive state factor record value until the Mth attribute box sensitive state factor record value, take the box boundary, construct the first attribute box sensitive state factor standard interval until the Mth attribute box sensitive state factor standard interval, and add it into the sensitive state factor standard interval.

[0057] Static thresholds cannot adapt to the dynamic operating conditions of power distribution equipment, such as load fluctuations, ambient temperature changes, and component aging. Manually adjusting thresholds lacks real-time accuracy and objectivity. Existing health sample retrieval ignores the coupling effects of multiple parameters, leading to a mismatch between thresholds and actual operating conditions.

[0058] In this embodiment, a first limiting condition is constructed. First, based on historical data, operating environment parameters of the power distribution equipment (e.g., temperature, in °C), power distribution load parameters (e.g., current percentage, in %), and component service life (in months) are collected to set a limiting condition attribute set. For example, a temperature range of -25°C to +50°C is set as a limiting condition, a current range of 20% to 100% of the component's rated value is set as a limiting condition, and a component service life range of 0 to 120 months is set as a limiting condition. The limiting condition attribute set is then integrated to obtain the first limiting condition, which includes all limiting conditions in the limiting condition set.

[0059] Through the user interface, the tolerance deviation for the limiting condition attribute is configured. This tolerance deviation includes a first tolerance deviation for the limiting condition attribute and other tolerance deviations for other limiting condition attributes. The tolerance deviation for the limiting condition attribute refers to the deviation value of each condition's tolerance. Deviations within the tolerance range can be considered minor, occasional deviations. For example, setting the tolerance deviation for the limiting condition attribute to 5% means that deviations within 5% are within the tolerance range for the limiting condition attribute.

[0060] Traverse the historical dataset and collect the set of abnormal attribute changes of sensitive state factors, with the constraint that the deviation of the first constraint attribute is greater than or equal to the fault tolerance deviation of the first constraint attribute, and the deviation of other constraint attributes is less than the fault tolerance deviation of other constraint attributes.

[0061] Calculate the mean of the set of abnormal attribute changes of sensitive state factors, set the mean as the first limiting condition attribute influence parameter, and add it to the set of limiting condition attribute influence parameters.

[0062] Calculate the ratio of the influence parameter of the first limiting condition attribute to the sum of the influence parameters of the set of limiting condition attributes, and set this as the weight of the first limiting condition attribute. Add this weight to the influence weight distribution result. The weight of the first limiting condition attribute = influence parameter of the first limiting condition attribute ÷ sum of influence parameters of the limiting condition attributes. Based on the sensitive state factor, distribute the weights of the operating environment parameters of the power distribution equipment, the power distribution load parameters, and the service life of the components to obtain the influence weight distribution result. When the deviation of the limiting attribute is large, observe the change in the number of abnormal attributes of the sensitive state factor. The greater the change, the greater the influence of the first limiting condition attribute, and thus the greater the weight of the first limiting condition attribute. Specifically, for example, if the number of abnormal attributes of sensitive state factors related to the operating environment of power distribution equipment, such as temperature change, changes by 20; the number of abnormal attributes of sensitive state factors related to the load parameters, such as current change, changes by 15; and the number of abnormal attributes of sensitive state factors related to the service life of components, such as the number of months of service, changes by 15, then the influence weight of the operating environment of power distribution equipment = 20 ÷ (20 + 15 + 15) = 0.4, the influence weight of the load parameters = 15 ÷ (20 + 15 + 15) = 0.3, and the influence weight of the service life of components = 15 ÷ (20 + 15 + 15) = 0.3. The distribution of influence weights is then obtained by integrating these factors.

[0063] The operating environment parameters, power load parameters, and component service duration of power distribution equipment are normalized to obtain the baseline parameters for the limiting conditions. Normalization is a method to simplify calculations, transforming dimensional expressions into dimensionless expressions, becoming pure scalars. For example, the operating environment baseline parameter = (operating environment parameter - minimum operating environment parameter) ÷ (maximum operating environment parameter - minimum operating environment parameter), the operating environment baseline parameter = (power load parameter - minimum power load parameter) ÷ (maximum power load parameter - minimum power load parameter), and the operating environment baseline parameter = (component service duration - minimum component service duration) ÷ (maximum component service duration - minimum component service duration).

[0064] Construct deviation assessment rules:

[0065] Step 1: Calculate the magnitude of the constraint deviation between the constraint condition sample parameters and the constraint condition reference parameters. The magnitude of the constraint deviation is: constraint condition sample parameters - constraint condition reference parameters.

[0066] Step 2: Calculate the weighted average of the deviation modulus of the limiting conditions based on the distribution of influence weights to obtain the deviation evaluation value. Deviation evaluation value = ∑ (influence weight × limiting condition deviation modulus).

[0067] When the deviation evaluation value is greater than or equal to the deviation evaluation threshold, it is considered that the first constraint condition is not met. The deviation evaluation threshold is a pre-set threshold for deviation evaluation, that is, the deviation evaluation value of the attribute tolerance deviation exceeding the constraint condition is considered to exceed the deviation evaluation threshold.

[0068] When the deviation assessment value is less than the deviation assessment threshold, it is considered that the first constraint condition is met.

[0069] A second constraint is constructed using component model and sensitive state factor as constraints. A component is considered to satisfy the second constraint if it simultaneously satisfies the following conditions: the component model is consistent with the sample component model, and the collected attributes of the state factor are consistent with the attributes of the sensitive state factor.

[0070] All components are searched, and those that meet the first constraint condition are integrated into the first component healthy service sample set.

[0071] From the first set of healthy service samples, select the set of healthy service samples that meet the second constraint condition, wherein any healthy service sample in the set of healthy service samples includes a sensitive state factor record value.

[0072] Based on the component health service sample set, box plot analysis of the sensitive state factor records was performed to obtain the sensitive state factor records of the first attribute box up to the Mth attribute box. A box plot is a statistical chart used to display the central tendency and dispersion of a dataset, allowing for intuitive observation of the data's distribution characteristics, symmetry, and outliers.

[0073] The system iterates through the sensitive state factor records of the first attribute bin up to the Mth attribute bin, taking the bin boundaries and constructing standard intervals for the sensitive state factors of the first attribute bin up to the Mth attribute bin. These standard intervals are then added to the defined sensitive state factor standard intervals. The bin boundaries are the interval between the lower quartile and the upper quartile, which is the middle 50% of the data from the first attribute bin sensitive state factor records to the Mth attribute bin sensitive state factor records. Selecting this interval as the standard interval ensures that the selection of health standards is both broad and more accurate.

[0074] This application embodiment uses operating environment, load parameters, service life and component model as composite constraints to dynamically retrieve a healthy sample set matching the operating conditions. Through box plot analysis, it generates a standard interval for sensitive state factors, realizes the adaptive fitting of threshold to multi-dimensional operating condition parameters, and fundamentally solves the problem of threshold rigidity.

[0075] The region clustering module 400 is used to perform region clustering on the three-dimensional model of the power distribution equipment based on the sensitive state factor standard interval and the component location, to obtain the first region state standard interval up to the Nth region state standard interval.

[0076] The independent component monitoring mode requires performing status judgment on each component separately. When the status characteristics of adjacent components are similar, a large number of duplicate calculations are likely to occur, resulting in a waste of computing power, slowing down the warning time, and the fragmented warning information increases the difficulty of fault location.

[0077] In this embodiment of the application, the region clustering module 400 includes the following execution steps:

[0078] The three-dimensional model of the power distribution equipment is divided into meshes according to a preset side length to construct a three-dimensional mesh model of the power distribution equipment.

[0079] Based on the standard interval of the sensitive state factor, a region clustering analysis is performed on the mesh of the three-dimensional mesh model of the power distribution equipment to obtain the first region up to the Nth region;

[0080] Specifically, based on the standard interval of the sensitive state factor, a region clustering analysis is performed on the mesh of the three-dimensional mesh model of the power distribution equipment to obtain the first region up to the Nth region, including:

[0081] Extract the first and second meshes from the three-dimensional mesh model of the power distribution equipment;

[0082] Extract the first grid sensitive state factor standard interval and the second grid sensitive state factor standard interval from the aforementioned sensitive state factor standard interval;

[0083] Calculate the intersection-union ratio of the first attribute intervals up to the Q-th attribute intervals of the first grid sensitive state factor standard interval and the second grid sensitive state factor standard interval;

[0084] When the cross-union ratio of each attribute interval up to the Qth attribute interval is greater than or equal to the cross-union ratio threshold, the first grid and the second grid are added to the same cluster region.

[0085] If the cross-union ratio of any attribute interval up to the Qth attribute interval is less than the cross-union ratio threshold, then the first grid and the second grid are added to the heterogeneous cluster region.

[0086] Perform centroid interval analysis on the set of standard intervals of sensitive state factors in the first region to obtain the standard intervals of the state in the first region;

[0087] Specifically, centroid interval analysis is performed on the set of standard intervals for sensitive state factors in the first region to obtain the standard intervals for the states of the first region, including:

[0088] Obtain the mean of the first intersection-union ratios (IUCN) of the first type of grid with other grids in the first region;

[0089] Until the mean of the Y-th intersection-union ratios of the Y-th grids within the first region and other grids is obtained;

[0090] Take the interval corresponding to the minimum value of the grid from the first intersection-union ratio mean to the Yth intersection-union ratio mean, set it as the centroid interval, and obtain the first region state standard interval;

[0091] Until the set of standard intervals of sensitive state factors in the Nth region is analyzed by centroid interval analysis, the standard intervals of the state in the Nth region are obtained.

[0092] In this embodiment of the application, the three-dimensional model of the power distribution equipment is divided into grids according to a preset side length to construct a three-dimensional grid model of the power distribution equipment. The preset side length is a moderate length that is set in advance to divide the three-dimensional model of the power distribution equipment into grids of appropriate size. For example, the preset side length is set to 10 centimeters.

[0093] From the three-dimensional mesh model of the power distribution equipment, randomly extract the divided meshes as the first and second meshes;

[0094] Extract the standard intervals of the first grid sensitive state factors and the second grid sensitive state factors from the standard intervals of the sensitive state factors.

[0095] Calculate the intersection-union ratio (IUR) of the first attribute intervals up to the Qth attribute intervals, using the standard intervals of the first and second grid sensitive state factor intervals. IUR = (Standard interval of first grid sensitive state factor ∩ Standard interval of second grid sensitive state factor) ÷ (Standard interval of first grid sensitive state factor ∪ Standard interval of second grid sensitive state factor). The IUR reflects the similarity between the standard intervals of the first and second grid sensitive state factor intervals for the first attribute. A higher IUR indicates greater similarity between the first and second grids for that attribute, and a higher likelihood that their states are identical.

[0096] When the cross-union ratio of each attribute interval up to the Qth attribute interval is greater than or equal to the cross-union ratio threshold, the first and second grids are added to the same cluster region. The resulting same cluster region is set as the first region. The cross-union ratio threshold is a pre-defined threshold that reflects the similarity of grids on the same attribute; for example, the cross-union ratio threshold is set to 0.8.

[0097] When the cross-union ratio of any attribute interval up to the Qth attribute interval is less than the cross-union ratio threshold, the first grid and the second grid are added to the heterogeneous cluster region. Specifically, the first grid is added to the first region, and the second grid is added to the second region.

[0098] The arithmetic mean of the first intersection-union ratios of the first type of grid in the first region with other grids is obtained through calculation.

[0099] Starting with the first type of grid in the first region, and continuing until the Y-th type of grid within the first region is obtained, calculate the Y-th intersection-union ratio (IU / U) of the first type of grid in the first region with other grids, and then calculate the arithmetic mean of all IU / U. The larger the IU / U, the more similar its distribution is to other grids.

[0100] The minimum value among the first intersection-union ratio (IU) mean up to the Yth IU mean is obtained. The interval corresponding to the minimum value grid is retrieved and set as the centroid interval, thus obtaining the standard interval for the first region state. By selecting the interval corresponding to the minimum value grid, the most typical grid is obtained.

[0101] Until the set of standard intervals for sensitive state factors in region N is analyzed, centroid interval analysis is performed to obtain the standard intervals for the state of region N. The standard intervals for the state of regions from the first to the Nth region are obtained, reflecting the overall typical characteristics of similar grids.

[0102] Based on the standard intervals of sensitive state factors and the spatial location of components, this application clusters neighboring components with similar thresholds into a unified monitoring area and generates regional-level state standard intervals. By replacing independent components with regions as monitoring units, the number of monitoring points is reduced, while the typicality of cluster state characteristics is enhanced, laying a data foundation for rapid and accurate early warning.

[0103] The anomaly warning module is used to execute an anomaly warning for power distribution equipment when the status monitoring value of any area does not meet the status standard range of the first area up to the status standard range of the Nth area.

[0104] Traditional early warning mechanisms trigger alarms immediately when component-level anomalies occur, but they do not distinguish between local anomalies and system risks, which may lead to frequent false alarms and increase the burden on operations and maintenance. At the same time, decentralized component-level early warnings make it difficult to quickly locate the scope of fault impact, affecting the effectiveness of early warnings.

[0105] In this embodiment of the application, when the status monitoring value of any area does not meet the status standard range of the first area up to the status standard range of the Nth area, an abnormal warning for the power distribution equipment is executed. For example, an alarm is triggered by using a buzzer, or a warning text message such as "There is an abnormality in the power distribution cabinet, please check and handle" is sent to the user terminal to issue a warning.

[0106] This module performs batch status determination by using regional-level status standard intervals, triggering alerts only when the entire region is abnormal. Using the region as the smallest alert unit, it filters out false alarms from isolated components and quickly pinpoints the scope of fault impact by spatially aggregating abnormal regions, improving alert reliability and operational response efficiency.

[0107] In summary, the embodiments of this application have at least the following technical effects:

[0108] This application proposes an integrated intelligent monitoring and early warning device for power distribution equipment. By dynamically generating standard intervals of state factors adapted to operating conditions and optimizing the division of monitoring units, it significantly improves the accuracy of state warnings and the efficiency of system computing power utilization. Compared with traditional methods, the technical solution provided in this application significantly overcomes the shortcomings of rigid static thresholds and redundant analysis of independent components. Specifically, firstly, it dynamically retrieves a health sample set based on component model, service life, environmental parameters, and load parameters. Through box plot analysis, it generates flexible standard intervals of sensitive state factors, enabling state thresholds to adapt to equipment aging, load fluctuations, and environmental changes, effectively avoiding false alarms or missed alarms caused by differences in operating conditions. Secondly, through spatial clustering, it aggregates neighboring components with similar standard intervals into a unified monitoring area, performing state judgment on an area-by-area basis, greatly reducing the number of monitoring points that need to be analyzed independently. This not only reduces the computing power burden caused by the independent calculation of massive components but also enhances the typicality of cluster state characteristics by utilizing the centroid interval of the area.

[0109] This application achieves the technical effect of dynamic, accurate, and lightweight early warning for power distribution equipment under adaptive operating conditions, providing reliable support for the intelligent operation and maintenance of complex power distribution equipment.

[0110] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0111] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0112] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. An integrated intelligent monitoring and early warning device for the status of power distribution equipment, characterized in that, include: The component information acquisition module is used to decompose the three-dimensional model of the power distribution equipment in a spatial coordinate system to obtain the component model, component service life and component location; The correlation analysis module is used to perform abnormal correlation analysis on the equipment status factors based on the component model to obtain sensitive status factors; The sensitivity factor statistics module is used to retrieve the component health service sample set and statistically analyze the standard interval of the sensitivity state factor, based on the power distribution equipment operating environment parameters, power distribution load parameters, the service life of the component, the component model and the sensitivity state factor. The region clustering module is used to perform region clustering on the three-dimensional model of the power distribution equipment based on the sensitive state factor standard interval and the component location, to obtain the first region state standard interval up to the Nth region state standard interval. The abnormal warning module is used to execute an abnormal warning for power distribution equipment when the status monitoring value of any area does not meet the corresponding area status standard interval from the first area status standard interval to the Nth area status standard interval. Among them, constrained by the operating environment parameters of the power distribution equipment, the power load parameters, the service life of the component, the component model, and the sensitive state factor, a sample set of healthy service of the component is retrieved, and the standard interval of the sensitive state factor is statistically analyzed, including: A first constraint condition is constructed using the operating environment parameters of the power distribution equipment, the power distribution load parameters, and the service life of the components as constraints. A second constraint condition is constructed using the component model and the sensitive state factor as constraints; Retrieve a first component health service sample set that satisfies the first constraint condition; From the first component health service sample set, select the component health service sample set that satisfies the second restriction condition, wherein any component health service sample in the component health service sample set includes a sensitive state factor record value. Based on the component health service sample set, the recorded values ​​of the sensitive state factors are analyzed by the same attribute box plot to obtain the recorded values ​​of the sensitive state factors of the first attribute box up to the recorded values ​​of the sensitive state factors of the Mth attribute box. Traverse the first attribute box sensitive state factor record value until the Mth attribute box sensitive state factor record value, take the box boundary, construct the first attribute box sensitive state factor standard interval until the Mth attribute box sensitive state factor standard interval, and add it into the sensitive state factor standard interval.

2. The apparatus as claimed in claim 1, characterized in that, Based on the component model, anomaly correlation analysis is performed on the equipment state factors to obtain sensitive state factors, including: Based on the component model, a component fault dataset is collected, wherein any component fault data in the component fault dataset includes a set of state attributes that do not meet the preset state standard range. Based on the component fault dataset, the correlation degree of several state attributes is calculated, wherein the correlation degree of state attribute anomalies is equal to the ratio of the number of fault data entries triggered by state attributes to the total number of data entries in the component fault dataset. Based on the abnormal correlation of the aforementioned state attributes, state attributes with an abnormal correlation greater than or equal to the abnormal correlation threshold are extracted and added to the sensitive state factor.

3. The apparatus as described in claim 1, characterized in that, Using the operating environment parameters of the power distribution equipment, the power distribution load parameters, and the service life of the components as constraints, a first limiting condition is constructed, including: Using the component model as a constraint, and based on the sensitive state factor, the operating environment parameters of the power distribution equipment, the power distribution load parameters, and the service life of the component are weighted and distributed to obtain the influence weight distribution results; The operating environment parameters of the power distribution equipment, the power distribution load parameters, and the service life of the components are normalized to obtain the baseline parameters for the limiting conditions. Construct deviation assessment rules: Step 1: Calculate the magnitude of the constraint deviation between the constraint condition sample parameters and the constraint condition reference parameters; Step 2: Calculate the weighted average of the deviation modulus of the constraint conditions based on the influence weight distribution results to obtain the deviation evaluation value; When the deviation evaluation value is greater than or equal to the deviation evaluation threshold, it is considered that the first constraint condition is not met. When the deviation evaluation value is less than the deviation evaluation threshold, it is considered that the first constraint condition is met.

4. The apparatus as described in claim 3, characterized in that, Using the component model as a constraint, and based on the sensitive state factor, a weighted distribution is applied to the operating environment parameters of the power distribution equipment, the power distribution load parameters, and the service life of the component to obtain the influence weight distribution results, including: The operating environment parameters of the power distribution equipment, the power distribution load parameters, and the service life of the components are set as a set of limiting condition attributes, and the first limiting condition attribute is extracted. Through the user terminal, the tolerance deviation of the limiting condition attribute is configured, wherein the tolerance deviation of the limiting condition attribute includes a first tolerance deviation of the limiting condition attribute and other tolerance deviations of the limiting condition attribute. The set of abnormal attribute changes of sensitive state factors is collected, with the first constraint attribute deviation being greater than or equal to the first constraint attribute fault tolerance deviation and the other constraint attribute deviations being less than the other constraint attribute fault tolerance deviations. Calculate the mean of the set of abnormal attribute changes of the sensitive state factors, set it as the first constraint condition attribute influence parameter, and add it to the set of constraint condition attribute influence parameters. Calculate the ratio of the first constraint condition attribute influence parameter to the sum of the set of constraint condition attribute influence parameters, set it as the first constraint condition attribute weight, and add it to the influence weight distribution result.

5. The apparatus as claimed in claim 1, characterized in that, Based on the aforementioned sensitive state factor standard intervals and in conjunction with the component locations, the three-dimensional model of the power distribution equipment is subjected to region clustering to obtain the first region state standard interval up to the Nth region state standard interval, including: The three-dimensional model of the power distribution equipment is divided into meshes according to a preset side length to construct a three-dimensional mesh model of the power distribution equipment. Based on the standard interval of the sensitive state factor, a region clustering analysis is performed on the mesh of the three-dimensional mesh model of the power distribution equipment to obtain the first region up to the Nth region; Perform centroid interval analysis on the set of standard intervals of sensitive state factors in the first region to obtain the standard intervals of the state in the first region; Until the set of standard intervals of sensitive state factors in the Nth region is analyzed by centroid interval analysis, the standard intervals of the state in the Nth region are obtained.

6. The apparatus as claimed in claim 5, characterized in that, Based on the aforementioned sensitive state factor standard interval, a region clustering analysis is performed on the mesh of the three-dimensional mesh model of the power distribution equipment to obtain the first region up to the Nth region, including: Extract the first mesh and the second mesh from the three-dimensional mesh model of the power distribution equipment; Extract the first grid sensitive state factor standard interval and the second grid sensitive state factor standard interval from the aforementioned sensitive state factor standard interval; Calculate the intersection-union ratio of the first attribute intervals up to the Q-th attribute intervals of the first grid sensitive state factor standard interval and the second grid sensitive state factor standard interval; When the cross-union ratio of each attribute interval up to the Qth attribute interval is greater than or equal to the cross-union ratio threshold, the first grid and the second grid are added to the same cluster region. If the cross-union ratio of any attribute interval up to the Qth attribute interval is less than the cross-union ratio threshold, then the first grid and the second grid are added to the heterogeneous cluster region.

7. The apparatus as claimed in claim 5, characterized in that, Centroid interval analysis is performed on the set of standard intervals of sensitive state factors in the first region to obtain the standard intervals of the state in the first region, including: Obtain the mean of the first intersection-union ratios (IUCN) of the first type of grid with other grids in the first region; Until the mean of the Y-th intersection-union ratios of the Y-th grids within the first region and other grids is obtained; Take the interval corresponding to the minimum value of the first intersection-union ratio (IU) up to the minimum value of the Yth IU, and set it as the centroid interval to obtain the standard interval of the first region state.

Images