Intelligent monitoring method and system for electrical equipment

Abstract

The application provides an electrical equipment intelligent monitoring method and system, and belongs to the technical field of electrical equipment monitoring. The method comprises the following steps: determining an initial collection frequency of a target sensor based on temperature information, humidity information and seasonal information of a current environment, and adjusting the initial collection frequency based on the load of a current electrical equipment to obtain a target collection frequency of the target sensor for collecting different types of operation data of the electrical equipment; the target sensor is a sensor for collecting different types of operation data of the electrical equipment; collecting different types of operation data of the electrical equipment based on the target collection frequency, and pre-processing the operation data to obtain target operation data; performing feature extraction on the target operation data to obtain an operation state index of the electrical equipment, and determining an operation state grade of the electrical equipment based on the operation state index. The electrical equipment intelligent monitoring method and system can improve the accuracy and efficiency of electrical equipment monitoring.

Description

Technical Field

[0001] This application belongs to the field of electrical equipment monitoring technology, and more specifically, relates to an intelligent monitoring method and system for electrical equipment. Background Technology

[0002] In the field of electrical equipment operation monitoring, to ensure the safe and stable operation of equipment, it is necessary to collect operational data and assess its status through sensors. Currently, electrical equipment monitoring largely relies on collecting sensor data at fixed frequencies. While this method can obtain basic operational information, it does not consider the impact of environmental conditions and equipment load on data collection requirements. In harsh environments or when equipment is under high load, fixed low-frequency acquisition is prone to missing key status changes. In mild environments or when equipment is under low load, fixed high-frequency acquisition generates redundant data, increasing hardware energy consumption and data storage pressure. This results in low accuracy and efficiency of traditional monitoring methods, making it difficult to accurately reflect the true operating status of the equipment. Summary of the Invention

[0003] The purpose of this application is to provide an intelligent monitoring method and system for electrical equipment, so as to improve the accuracy and efficiency of electrical equipment monitoring.

[0004] A first aspect of this application provides an intelligent monitoring method for electrical equipment, comprising: The initial acquisition frequency of the target sensor is determined based on the current temperature, humidity, and seasonal information of the environment. The initial acquisition frequency is then adjusted based on the current load of the electrical equipment to obtain the target acquisition frequency for collecting different types of operating data from the electrical equipment. The target sensor is the sensor that collects different types of operating data from the electrical equipment. Based on the target acquisition frequency, different types of operating data of electrical equipment are collected, and the operating data is preprocessed to obtain the target operating data; Feature extraction is performed on the target operating data to obtain the operating status indicators of the electrical equipment, and the operating status level of the electrical equipment is determined based on the operating status indicators.

[0005] A second aspect of this application provides an intelligent monitoring system for electrical equipment, comprising: The sampling frequency determination module is used to determine the initial sampling frequency of the target sensor based on the current environmental temperature, humidity, and seasonal information, and to adjust the initial sampling frequency based on the current load of the electrical equipment to obtain the target sampling frequency for the target sensor to collect different types of operating data from the electrical equipment; the target sensor is the sensor that collects different types of operating data from the electrical equipment. The data processing module is used to collect different types of operating data from electrical equipment based on the target acquisition frequency, and to preprocess the operating data to obtain the target operating data. The status monitoring module is used to extract features from the target operating data, obtain the operating status indicators of the electrical equipment, and determine the operating status level of the electrical equipment based on the operating status indicators.

[0006] A third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the above-described intelligent monitoring method for electrical equipment.

[0007] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described intelligent monitoring method for electrical equipment.

[0008] The beneficial effects of the intelligent monitoring method and system for electrical equipment provided in this application are as follows: This application dynamically determines the target acquisition frequency, adapting the sensor acquisition density to environmental and load changes. The frequency is increased in harsh environments or under high loads to capture critical state changes, and conversely, the frequency is reduced to decrease redundancy, ensuring data validity while reducing energy consumption and storage pressure. Preprocessed target operating data eliminates noise, missing values, and magnitude differences, providing reliable input for feature extraction and avoiding analytical biases caused by defects in the original data. Based on the operating status indicators and level classifications obtained from feature extraction, reliable assessment of equipment status is achieved, replacing traditional fuzzy judgment. This improves monitoring accuracy and efficiency, reduces misjudgments and omissions, and ensures the safe and stable operation of electrical equipment. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of this application, 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 this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 A flowchart illustrating an embodiment of the intelligent monitoring method for electrical equipment provided in this application; Figure 2 This is a structural block diagram of an intelligent monitoring system for electrical equipment provided in an embodiment of this application; Figure 3 This is a schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0011] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0012] To make the objectives, technical solutions, and advantages of this application clearer, the following description will be provided in conjunction with the accompanying drawings and specific embodiments.

[0013] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating an embodiment of an intelligent monitoring method for electrical equipment provided in this application. The method can be executed by an electronic device and may include: S101: Determine the initial acquisition frequency of the target sensor based on the current environmental temperature, humidity and seasonal information, and adjust the initial acquisition frequency based on the current load of the electrical equipment to obtain the target acquisition frequency for the target sensor to collect different types of operating data of the electrical equipment; the target sensor is the sensor that collects different types of operating data of the electrical equipment.

[0014] In this embodiment, electrical equipment refers to all types of equipment used in power systems for power generation, transmission, distribution, and consumption, realizing the production, conversion, transmission, distribution, control, or consumption of electrical energy. It is a core component of the normal operation of the power system. In this embodiment, the electrical equipment can be transmission and transformation equipment, such as transformers.

[0015] A target sensor is a combination of sensors used to collect various operating parameters of electrical equipment. Depending on the type of equipment, it may include current sensors, temperature sensors, voltage sensors, and vibration sensors, and can simultaneously acquire multi-dimensional data reflecting the status of the equipment.

[0016] The initial acquisition frequency is a basic data acquisition frequency calculated based on the current ambient temperature, humidity, and seasonal information, reflecting the intensity of environmental factors' demand for equipment status monitoring.

[0017] The target acquisition frequency is the final acquisition frequency obtained after correcting the initial acquisition frequency for equipment load. It takes into account the influence of the environment and the equipment's own operating load, and is the actual data acquisition frequency performed by the target sensor.

[0018] Different types of operational data are raw data acquired by the target sensor at the target acquisition frequency, including physical or electrical parameters of electrical equipment such as current, temperature, and voltage.

[0019] In this embodiment, the temperature, humidity and season of the environment directly affect the operational risk of electrical equipment. Excessive temperature will accelerate the aging of the equipment insulation layer (such as transformer winding insulation failure), and excessive humidity will cause corrosion and short circuit of metal parts. Different seasons (such as high temperature in summer and low temperature in winter) will form different environmental risk baselines. Therefore, it is necessary to combine these three types of environmental information and calculate the initial acquisition frequency through preset logic.

[0020] The initial acquisition frequency is the basic data density set for the target sensor to adapt to the current environmental risks. For example, the initial frequency is higher in hot and humid seasons than in normal temperature and humidity seasons to prevent sudden changes in equipment status caused by the environment from being missed due to excessively long acquisition intervals.

[0021] In this embodiment, the load of electrical equipment is a more direct risk-related factor (e.g., the risk of current and temperature fluctuations when electrical equipment is running at full load is much higher than that under light load). Therefore, the initial sampling frequency needs to be adjusted based on the current load. The higher the load, the higher the probability and risk of changes in the state of electrical equipment, and the sampling frequency needs to be increased accordingly; the lower the load, the lower the risk, and the sampling frequency can be appropriately reduced, ultimately obtaining the target sampling frequency.

[0022] This embodiment abandons the static mode of fixed acquisition frequency in traditional monitoring. By associating environmental influencing factors with the load of the electrical equipment itself, it dynamically adjusts the acquisition frequency of the target sensor to ensure that the acquired data can cover the state changes under critical operating conditions and avoid redundant acquisition in low-risk scenarios.

[0023] S102: Collect different types of operating data of electrical equipment based on the target acquisition frequency, and preprocess the operating data to obtain the target operating data.

[0024] In this embodiment, the target sensor can synchronously or asynchronously acquire different types of data such as current, voltage, temperature, or vibration of the electrical equipment based on a determined target acquisition frequency. These data can directly reflect the real-time operating status of the electrical equipment; for example, current data reflects load distribution, and temperature data reflects the equipment's heat dissipation. Because the target acquisition frequency is adapted to the current risks of the electrical equipment, the data acquired by the target sensor will neither lose critical information due to excessively long intervals nor cause storage and computational redundancy due to excessive density.

[0025] Raw operational data collected by target sensors may contain noise, missing values, and magnitude differences. Using this data directly without processing can lead to biased or even erroneous analysis results for electrical equipment, failing to accurately reflect its true operating status. Therefore, preprocessing of the raw operational data is necessary. Preprocessing eliminates these interferences, ensuring data quality; it also standardizes the data.

[0026] In this embodiment, outlier removal algorithms (such as the 3σ principle) are used to eliminate noise interference in the original operating data; linear interpolation is used to fill in missing values; and different data types have different units (such as current "A", temperature "°C") and numerical ranges, which are standardized (such as mapping to the [0, 1] interval) to unify the magnitude. After the above processing, the target operating data is obtained.

[0027] This embodiment transforms the various types of noisy and unstructured raw operational data collected by the target sensor at the target acquisition frequency into interference-free, standardized, and structured target operational data, providing a qualified data carrier for subsequent processing.

[0028] For example, taking a 10kV distribution transformer as an example, the target sensors include a current transformer, a temperature sensor, and an oil level sensor. Data is continuously collected for 5 hours (covering one monitoring cycle) at a frequency of 16 times per hour to obtain the following three types of operational data: Three-phase current data (unit: A): The current values ​​of phases A, B, and C are recorded every 3.75 minutes; Winding temperature data (unit: °C): Top layer oil temperature is recorded every 3.75 minutes; Oil level data (unit: mm): The oil level is recorded every 3.75 minutes.

[0029] The 3σ principle was used to remove outliers caused by electromagnetic interference in the current data (such as a sudden increase in phase A current to 1.5 times the rated value). For the two missing temperature data caused by sensor offline, linear interpolation was used to fill them (based on the temperature trend before and after the time). The current (0-100A), temperature (0-100℃), and oil level (300-350mm) data were normalized to the [0,1] interval through min-max normalization. The data were integrated according to the acquisition window. Each window contains normalized data from 5 consecutive acquisitions (covering about 18.75 minutes) to form 8 sets of target operating data (5 hours ÷ 0.3125 hours / window ≈ 16 windows, and the first 8 sets were used for feature extraction).

[0030] S103: Extract features from the target operating data to obtain the operating status indicators of the electrical equipment, and determine the operating status level of the electrical equipment based on the operating status indicators.

[0031] In this embodiment, the operating status level is a category of equipment status (such as normal or slightly abnormal) based on operating status indicators, and each level corresponds to a specific indicator range.

[0032] In this embodiment, a machine learning model can be used to extract key information characterizing the state of electrical equipment from target operating data. For example, current stability features reflecting whether current fluctuations are within a safe range can be extracted from current data; temperature anomaly risk features reflecting whether the temperature is close to a fault threshold can be extracted from temperature data; and voltage fluctuation amplitude features reflecting power supply stability can be extracted from voltage data. These features are integrated into operating status indicators. For instance, by assigning weights to the impact of features on the safety of electrical equipment, the feature values ​​corresponding to these features are weighted and fused to obtain an operating status indicator that comprehensively reflects the overall state of the equipment. For example, "current stability 0.9 + temperature anomaly risk 0.2 + voltage fluctuation amplitude 0.8," after weighted calculation, forms an operating status indicator representing the overall health of the electrical equipment.

[0033] In this embodiment, a classification rule matching the type of electrical equipment and application scenario can be preset. By matching the current operating status index with the preset interval, the specific status level of the equipment can be determined. This embodiment extracts key features of the electrical equipment status from the target operating data, integrates these features to form operating status indicators, and then uses these indicators to achieve a graded judgment of the electrical equipment status. This transforms the traditional fuzzy judgment relying on human experience into a reliable data-based assessment, providing a basis for the operation and maintenance decisions of electrical equipment.

[0034] As can be seen from the above, this embodiment dynamically determines the target acquisition frequency, allowing the sensor acquisition density to adapt to environmental and load changes. The frequency is increased in harsh environments or under high loads to capture critical state changes, while the frequency is reduced to decrease redundancy, ensuring data validity while reducing energy consumption and storage pressure. The pre-processed target operating data eliminates noise, missing values, and magnitude differences, providing reliable input for feature extraction and avoiding analytical biases caused by defects in the original data. Based on the operating status indicators and level classifications obtained from feature extraction, reliable assessment of equipment status is achieved, replacing traditional fuzzy judgment. This improves monitoring accuracy and efficiency, reduces misjudgments and omissions, and ensures the safe and stable operation of electrical equipment.

[0035] In one embodiment of this application, determining the initial sampling frequency of the target sensor based on the current environmental temperature, humidity, and seasonal information includes: The initial acquisition frequency of the target sensor is determined based on the first formula, which is:

[0036] in, Indicates the initial sampling frequency. Indicates the reference sampling frequency. Indicates the temperature effect coefficient. This represents the difference between the current temperature information and the standard temperature information. Indicates the humidity influence coefficient. This represents the difference between the current humidity information and the standard humidity information. This represents the coefficient corresponding to the current season; The initial acquisition frequency is adjusted based on the current load of the electrical equipment to obtain the target acquisition frequency of the target sensor, including: The initial acquisition frequency is adjusted based on the second formula to obtain the target acquisition frequency of the target sensor. The second formula is:

[0037] in, Indicates the target acquisition frequency. This represents the load impact factor, where L represents the current load on the electrical equipment.

[0038] In this embodiment, the initial sampling frequency is determined by a reference sampling frequency. Base value, temperature effect item This reflects the impact of the current temperature deviation from the standard temperature on the required sampling density (e.g., the greater the temperature deviation, the larger this value, driving up the initial frequency); humidity effect item. Similarly, the influence of humidity deviation is reflected; the seasonal coefficient S is used to make an overall correction for the basic environmental risks of different seasons (e.g., higher in summer than in winter). The four factors are superimposed to obtain the initial sampling frequency adapted to the current environment. This enables precise control of the sampling frequency by environmental factors.

[0039] For example, taking a 10kV transformer as an example, the reference sampling frequency... =6 times / hour (basic sampling frequency of the transformer under standard conditions); temperature influence coefficient =0.3, the standard temperature information (the reference temperature for normal transformer operation) is 25℃, and the current temperature information is 33℃, then =33-25=8; Humidity Influence Coefficient =0.2, the standard humidity information (the baseline humidity for normal equipment operation) is 55%, and the current humidity information is 70%, then =70-55=15; The current season is summer, and the corresponding coefficient S=3 (summer has a higher risk of high temperature and humidity, so the coefficient is higher than 1 in spring and autumn and 0.5 in winter). Substitute into the first formula to calculate: times / hour.

[0040] In this embodiment, the second formula adjusts the target acquisition frequency using the initial frequency. Based on this, a load impact factor (1+) is introduced. •L), when the load L of the electrical equipment increases, the load influence factor increases, causing the target frequency to... The frequency is increased proportionally from the initial frequency to ensure more intensive data collection under high load; when the load decreases, the frequency is decreased proportionally. Through the direct proportional relationship between load and frequency, a target collection frequency that adapts to both environment and load is obtained, which ensures data density under critical operating conditions and avoids redundant collection in low-risk scenarios.

[0041] For example, the load influence factor =0.6 (weight of load on sampling frequency, which can be preset according to equipment type); Current load of electrical equipment L=0.7 (represents the ratio of actual load to rated load, such as rated capacity 500kVA, current actual load 350kVA), substitute into the second formula for calculation: times / hour.

[0042] As can be seen from the above, this embodiment quantifies the impact of environmental factors on the sampling frequency using the first formula, with reference to the sampling frequency. Provide benchmarks, • and • It accurately reflects the impact of temperature and humidity deviations, with S supplementing the overall seasonal risk, making the initial frequency fᵢ adapt to environmental risks; then, through the second formula, (1+ •L) Transform the load impact into a frequency adjustment coefficient, allowing the target frequency to... The frequency of data acquisition increases proportionally with the load. This allows the acquisition frequency to be increased in harsh environments to capture critical changes, and decreased during low load periods to reduce redundancy. This ensures data validity while reducing energy consumption and storage pressure, solving the problem of balancing monitoring accuracy and cost when using a fixed frequency, and improving the adaptability and economy of data acquisition.

[0043] In one embodiment of this application, feature extraction is performed on target operating data to obtain operating status indicators of electrical equipment, including: The target operating data is input into a pre-trained convolutional neural network model to obtain multiple operating state feature values ​​of the electrical equipment; Determine the weights corresponding to the feature values ​​of each operating state; The operating status index of electrical equipment is obtained by weighted fusion of multiple operating status feature values ​​and their corresponding weights.

[0044] In this embodiment, the pre-trained convolutional neural network model is a convolutional neural network that has been trained in advance using a large amount of historical operating data of electrical equipment (including normal and abnormal state data). It has the ability to automatically extract equipment state features from structured target operating data, and can directly input new data to output feature values, thus adapting to the real-time requirements of monitoring scenarios.

[0045] The operating status feature value is a quantified value (usually in the range of 0-1) extracted from the target operating data by a pre-trained convolutional neural network model. Each feature value corresponds to a key operating dimension of the equipment (such as current stability and temperature risk). The magnitude of the value directly reflects the quality of the state of that dimension (e.g., the closer the temperature anomaly risk feature value is to 1, the lower the temperature risk).

[0046] Operational status indicators represent the overall operating status of electrical equipment. The value of the corresponding operational status indicator is positively correlated with the health of the equipment (the higher the value, the better the status).

[0047] In this embodiment, the convolutional neural network, through its hierarchical convolution and pooling structure, can automatically extract deep features from target operational data. By using sliding calculations across multiple convolutional kernels, it captures key local patterns in the data (such as frequency characteristics of current fluctuations and trend characteristics of temperature changes). Pooling layers then retain salient features and filter redundant information, ultimately outputting multiple operational state feature values. These feature values ​​describe different dimensions of the electrical equipment's state (such as current stability and temperature anomaly risk), overcoming the limitations of manual feature extraction.

[0048] In this embodiment, based on the degree of impact of different features on the safe operation of the equipment (e.g., the impact of abnormal temperature on transformers is usually higher than that of oil level fluctuations), a corresponding weight can be assigned to each feature value using the entropy weighting method, ensuring that important features occupy a higher proportion in the final evaluation. Finally, weighted fusion integrates multiple single-dimensional feature values ​​into a comprehensive operating status index by multiplying each feature value by its weight and then summing the results. This index retains the key information of each feature and reflects its relative importance through weights, ultimately achieving a quantitative representation of the overall operating status of the equipment and providing accurate and comparable numerical basis for subsequent status level determination.

[0049] As can be seen from the above, this embodiment extracts multi-dimensional operational status feature values ​​through a pre-trained convolutional neural network model, overcoming the limitations of manual feature selection and accurately extracting equipment status details. Furthermore, by weighting and allocating key features to highlight their impact, multi-dimensional information is weighted, fused, and integrated to form a comprehensive operational status index. This achieves accurate judgment from scattered data to the overall status, improving the comprehensiveness and accuracy of the assessment.

[0050] In one embodiment of this application, determining the weights corresponding to each running state feature value includes: Construct a target feature matrix based on multiple operational state feature values; Calculate the information entropy corresponding to each running state feature value in the feature matrix; The weight corresponding to each running state feature value is calculated based on the information entropy of each running state feature value.

[0051] In this embodiment, firstly, a target feature matrix is ​​constructed based on multiple operating status feature values. The feature values ​​of different acquisition windows are then structurally integrated according to sample (row) × feature (column) to provide a standardized data carrier for subsequent calculations.

[0052] In one embodiment of this application, a target feature matrix is ​​constructed based on multiple operating state feature values, including: Within a preset monitoring period, multiple sets of target operating data are acquired based on a preset acquisition window. The acquisition window is for continuously acquiring target operating data a preset number of times. Each set of target operating data corresponds to multiple operating status characteristic values ​​of electrical equipment during a preset acquisition window period. An initial feature matrix is ​​constructed based on multiple sets of target operation data and multiple operation state feature values; the rows of the initial feature matrix represent each set of target operation data, and the columns of the initial feature matrix represent multiple operation state feature values ​​corresponding to each set of target operation data; The initial feature matrix is ​​standardized to obtain the target feature matrix.

[0053] In this embodiment, the preset monitoring period is a pre-defined time interval for centralized collection and analysis of electrical equipment operation data. It is the boundary of the data collection time range, ensuring that the data collection is regular and targeted, and avoiding data redundancy or missing data caused by unlimited collection.

[0054] The preset acquisition window is a pre-set threshold for the number of times the target operation data is continuously acquired. Each window corresponds to a complete set of multi-feature operation data, which can reflect the status snapshot of the device in a short period of time. By superimposing data from multiple windows, the error caused by single data fluctuations can be reduced and the representativeness of the data can be improved.

[0055] In this embodiment, a monitoring period (e.g., 1 hour) and a data acquisition window (e.g., acquiring 5 consecutive sets of target operation data as one window) are first set. Multiple sets of data are acquired within the monitoring period (e.g., acquiring 6 sets of data from 6 windows within 1 hour). Each set of data corresponds to multiple operational status feature values ​​(e.g., current fluctuation, temperature peak, voltage stability) within a single window period, providing multi-dimensional and multi-time-period raw data support for matrix construction and avoiding the random bias of single-window data. An initial feature matrix is ​​constructed by using rows to represent each set of target operation data and columns to represent the multiple operational status feature values ​​corresponding to each set of target operation data.

[0056] For example, taking a 10kV transformer as an example, the preset monitoring cycle is 2 hours (the time range for centralized data collection and analysis). Preset acquisition window: Four consecutive acquisitions of target running data constitute one window (each acquisition is 5 minutes apart, and one window covers 20 minutes). Target operating data types: three operating status feature values ​​(range 0-1) of preprocessed current stability (feature 1), temperature anomaly risk (feature 2), and oil level fluctuation (feature 3).

[0057] Within a 2-hour monitoring period, data was collected 24 times at a target collection frequency of 5 minutes per time. This data was divided into 6 collection windows (24 times ÷ 4 times / window = 6 windows), resulting in 6 sets of target operation data. Each set contains 3 feature values, as shown in Table 1, the feature value comparison table.

[0058] As shown in Table 1, the feature value comparison table

[0059] Construct an initial feature matrix M, where rows represent each set of target running data and columns represent multiple running state feature values ​​corresponding to each set of target running data.

[0060] The matrix has a dimension of 6×3 (6 rows correspond to 6 windows, and 3 columns correspond to 3 feature values), directly carrying the original feature information.

[0061] Because the dimensions and numerical ranges of feature values ​​differ significantly across different operating states, direct calculation can lead to weight bias towards features with larger numerical values. Standardization eliminates interference from dimensions and ranges, ensuring that all feature values ​​are on the same dimension. After standardization, the target feature matrix is ​​obtained.

[0062] Based on the standardized target feature matrix, the information entropy of each feature is calculated. The greater the fluctuation of the feature value across different windows, the smaller the information entropy, indicating that the feature has a stronger ability to distinguish the device state; conversely, the larger the entropy value, the weaker the distinguishing ability. The method for calculating information entropy is a conventional technique for those skilled in the art and will not be elaborated here.

[0063] Information entropy is used to convert entropy values ​​into the proportion of effective information in features. After normalization, weights are obtained, ultimately assigning higher weights to features with strong discriminative power. This provides an objective and accurate basis for the weighted fusion of operational status indicators. The weight calculation formula for each operational status feature value is as follows:

[0064] in, The weight of the j-th operating state feature value reflects the importance of this feature to the overall state assessment of the electrical equipment. The sum of all feature weights is 1. The information entropy represents the j-th running state feature value; m represents the total number of running state feature values.

[0065] As can be seen from the above, this embodiment integrates multi-dimensional feature data by constructing a target feature matrix, then calculates the entropy of each feature, giving higher weights to features with strong ability to distinguish device states, and finally deriving the weights based on the entropy values. The weight determination method in this embodiment avoids subjective experience bias, ensures a reasonable proportion of key features in the operating status indicators, and improves the objectivity and accuracy of weighted fusion.

[0066] In one embodiment of this application, determining the operating status level of electrical equipment based on operating status indicators includes: A fault sample library is constructed based on the historical fault types of electrical equipment and their corresponding operating status indicators; The K-means clustering algorithm is used to cluster the operating status indicators in the fault sample library to obtain the cluster centers of the indicators corresponding to the operating status levels of electrical equipment. Based on the index cluster center and the preset safety margin, the threshold range corresponding to the operating status level of electrical equipment is obtained. By matching operational status indicators with threshold ranges, the corresponding operational status level of electrical equipment can be determined.

[0067] In this embodiment, the fault sample library is a database that stores various fault types (such as short circuits, insulation failures, etc.) and their corresponding operating status indicators in the historical operation of electrical equipment. It contains information such as the fault type, occurrence time, and operating status indicator values ​​of electrical equipment. It is the basic data source for level classification and reflects the correlation between equipment faults and indicators.

[0068] The preset safety margin is to avoid misjudgment of the level due to small fluctuations in the indicators. The preset allowable fluctuation range (such as ±5% or ±0.05) on both sides of the cluster center can be set according to the equipment operation stability requirements to make the level threshold more in line with the actual working conditions.

[0069] The threshold range is the index interval corresponding to each operating status level determined based on the index cluster center and the preset safety margin (e.g., [0.8, 1.0] corresponds to the normal level). It is the standard boundary for determining the current status level of the equipment, and the index values ​​within the interval are all classified into the same level.

[0070] In this embodiment, different fault types (such as insulation aging, overload overheating, etc.) and their corresponding operating status indicators are collected during the historical operation of electrical equipment to form a sample set containing fault types and operating status indicator values. This provides data basis for the classification of levels and ensures that the level standards are consistent with the actual fault patterns of the equipment.

[0071] In this embodiment, the K-means clustering algorithm is used to automatically classify the status. By clustering, discrete operational status indicators in the sample database are grouped into several classes based on similarity (e.g., normal, slightly abnormal, moderately abnormal, and severely abnormal). The cluster center of each class represents the typical value of the indicator for that status. For example, indicators for the normal state are generally higher, and the cluster centers will be in the high-value region; indicators for the fault state are lower, and the cluster centers will be in the low-value region. This step overcomes the subjectivity of manually setting the level boundaries, making the level classification more closely reflect the data distribution patterns.

[0072] By combining cluster centers and preset safety margins to determine threshold ranges, and setting safety margins around each cluster center, indicator ranges for each level are formed (e.g., normal level corresponds to [0.8, 1.0], and slightly abnormal level corresponds to [0.6, 0.8]). This maintains the objectivity of the clustering results while adapting to indicator fluctuations in actual operation through margin design, avoiding misjudgments of level due to minor fluctuations. Finally, the current operating status indicators are matched with threshold ranges. If an indicator falls into a certain range, the equipment is determined to be in the corresponding level, realizing the transformation from quantitative indicators to qualitative levels, providing a clear decision-making basis for operation and maintenance (e.g., falling into the fault range triggers a maintenance warning).

[0073] For example, taking a transformer as an example, three years of historical operating data of a certain 110kV transformer were collected, covering four types of states: normal operation (500 records), core overheating (120 records), oil quality deterioration (90 records), and insulation aging (80 records). Each data record contains the corresponding operating status index (range of 0-1, the higher the value, the better the state), forming a fault sample library with 790 records.

[0074] The number of clusters, K=4 (corresponding to 4 state levels), was set to cluster the operating status indicators in the sample library. After 5 iterations, the cluster centers were obtained as follows: normal state 0.92, mild anomaly (early stage of core overheating) 0.71, moderate anomaly (oil quality deterioration) 0.48, and severe anomaly (insulation aging) 0.23.

[0075] The preset safety margin is ±0.1, and thresholds are defined based on cluster centers: normal [0.82, 1.0], slightly abnormal [0.61, 0.82], moderately abnormal [0.38, 0.61], and severely abnormal [0, 0.38]. If the current transformer operating status index is 0.56, falling within the range of [0.38, 0.61), it is determined to be at the moderately abnormal level, corresponding to the risk of oil quality deterioration.

[0076] As can be seen from the above, this embodiment constructs a sample library using historical fault data, obtains objective index cluster centers using K-means clustering, and forms a reasonable threshold range by combining safety margins, ultimately achieving state level matching. This embodiment avoids the subjectivity of manually setting level boundaries, making the level division conform to the actual operating rules of the equipment. At the same time, by using safety margins to accommodate index fluctuations, it improves the accuracy and reliability of state level determination.

[0077] In one embodiment of this application, the operating status levels of electrical equipment include normal, slightly abnormal, moderately abnormal, and severely abnormal; Determining the operating status level of electrical equipment based on operating status indicators also includes: In response to the electrical equipment's operating status level being a slightly abnormal state, the target acquisition frequency of the target sensor is updated based on a preset adjustment step size to obtain a new target acquisition frequency; In response to the electrical equipment's operating status level being a moderate abnormal state, a corresponding fault warning signal for the electrical equipment is generated; In response to the electrical equipment's operating status level being a severe abnormal state, a corresponding protection control signal for the electrical equipment is generated.

[0078] In this embodiment, the operating status of electrical equipment is divided into normal, slightly abnormal, moderately abnormal, and severely abnormal, with each level corresponding to a different level of risk. Normal indicates no risk, and severely abnormal indicates high risk, providing a basic framework for graded response.

[0079] When a minor anomaly is detected, the target sensor's acquisition frequency is updated by pre-setting the adjustment step size (e.g., increasing by 20% each time). For example, if the original frequency is 10 times / hour, it is updated to 12 times / hour to capture the trend of status changes with more intensive data acquisition, thereby achieving dynamic risk tracking. When a moderate anomaly is detected, a fault warning signal is automatically generated (e.g., pushed to the operation and maintenance platform) to prompt staff to intervene and conduct inspections before the fault escalates. When a serious anomaly is detected, a protection control signal is immediately generated (e.g., triggering trip or shutdown commands) to prevent equipment damage or safety accidents through emergency measures.

[0080] As can be seen from the above, this embodiment divides the operating status into four levels and matches them with differentiated responses. For minor anomalies, the data collection frequency is increased to accurately track trends; for moderate anomalies, an early warning is issued to prompt manual intervention; and for severe anomalies, protective controls are triggered to prevent the accident from escalating. This hierarchical handling mechanism avoids the waste of resources from excessive monitoring and ensures timely response when risks escalate, forming a closed loop from status assessment to proactive handling, thereby improving equipment operating safety and maintenance efficiency.

[0081] In one embodiment of this application, when constructing a fault sample library based on historical fault types of electrical equipment and corresponding operating status indicators, the method further includes: Historical fault data in the fault sample database are classified and labeled according to the fault causes, and each type of fault is associated with corresponding environmental parameters, load parameters and fault duration data. Set dynamic update trigger conditions for the fault sample library. When the feature similarity between a newly added fault type of electrical equipment and an existing fault type in the library is lower than the similarity threshold, the fault sample library update process is automatically triggered. The newly added fault type and its corresponding complete parameter information are entered into the fault sample library to obtain the updated fault sample library. The K-means clustering algorithm is used to cluster the operating status indicators in the updated fault sample library to obtain the cluster centers of the indicators corresponding to the operating status levels of electrical equipment.

[0082] In this embodiment, the original historical fault data only contains basic correlations of fault type and operating status indicators, failing to reflect the context of the fault occurrence (such as the difference in indicators for the same fault under different environments and loads). By classifying and labeling fault causes (such as environmental factors, equipment aging, and overload), a multi-dimensional correlation between fault type, cause, and scenario parameters can be established. Environmental parameters reflect the external environmental impact at the time of the fault, load parameters reflect the equipment's own operating pressure, and the duration of the fault is related to the degree of fault development. This labeling method enriches the sample dimensions and avoids the problem of mixing data from different scenarios of the same fault into one category during clustering due to missing scenario information, providing a more accurate sample basis for subsequent clustering.

[0083] During equipment operation, new types of faults not covered in the database may occur (such as aging faults of new components or special faults under extreme environments). If the sample database remains unchanged, the cluster centers of K-means clustering will gradually deviate from the actual fault patterns. By setting dynamic update trigger conditions, and using the feature similarity between new fault types and existing fault types as the judgment criterion, the operating status indicators, environment, and load parameters of new faults are integrated into feature vectors. Similarity (e.g., cosine similarity) is calculated between these vectors and the feature vectors of each fault type in the database. When the similarity is below a threshold (e.g., 60%), it is determined to be a new fault, and the update process is automatically triggered. This ensures that the sample database continuously covers fault types throughout the entire equipment lifecycle, avoiding clustering bias due to missing samples.

[0084] Based on the updated fault sample library, the algorithm performs clustering to identify the distribution patterns of operational status indicators under different fault causes and scenarios. For example, the cluster centers for operational status indicators of overload faults will differ from those of high humidity faults, and the algorithm can also incorporate the indicator features of novel faults. The resulting indicator cluster centers can more accurately correspond to different fault risk levels in actual equipment operation, providing a more reliable clustering basis for subsequent threshold range matching and equipment status determination based on operational status indicators.

[0085] Corresponding to the intelligent monitoring method for electrical equipment in the above embodiments, Figure 2 This is a structural block diagram of an intelligent monitoring system for electrical equipment provided in one embodiment of this application. For ease of explanation, only the parts relevant to the embodiment of this application are shown. References Figure 2 The intelligent monitoring system 20 for electrical equipment includes: a frequency determination module 21, a data processing module 22, and a status monitoring module 23.

[0086] The acquisition frequency determination module 21 is used to determine the initial acquisition frequency of the target sensor based on the current environmental temperature, humidity and seasonal information, and to adjust the initial acquisition frequency based on the current load of the electrical equipment to obtain the target acquisition frequency for the target sensor to acquire different types of operating data of the electrical equipment; the target sensor is a sensor that acquires different types of operating data of the electrical equipment. The data processing module 22 is used to collect different types of operating data of electrical equipment based on the target acquisition frequency, and to preprocess the operating data to obtain the target operating data; The status monitoring module 23 is used to extract features from the target operating data, obtain the operating status indicators of the electrical equipment, and determine the operating status level of the electrical equipment based on the operating status indicators.

[0087] In one embodiment of this application, the sampling frequency determination module 21 is specifically used for: The initial acquisition frequency of the target sensor is determined based on the first formula, which is:

[0088] in, Indicates the initial sampling frequency. Indicates the reference sampling frequency. Indicates the temperature effect coefficient. This represents the difference between the current temperature information and the standard temperature information. Indicates the humidity influence coefficient. This represents the difference between the current humidity information and the standard humidity information. This represents the coefficient corresponding to the current season; The initial acquisition frequency is adjusted based on the second formula to obtain the target acquisition frequency of the target sensor. The second formula is:

[0089] in, Indicates the target acquisition frequency. This represents the load impact factor, where L represents the current load on the electrical equipment.

[0090] In one embodiment of this application, the status monitoring module 23 is specifically used for: The target operating data is input into a pre-trained convolutional neural network model to obtain multiple operating state feature values ​​of the electrical equipment; Determine the weights corresponding to the feature values ​​of each operating state; The operating status index of electrical equipment is obtained by weighted fusion of multiple operating status feature values ​​and their corresponding weights.

[0091] In one embodiment of this application, the status monitoring module 23 is further configured to: Construct a target feature matrix based on multiple operational state feature values; Calculate the information entropy corresponding to each running state feature value in the feature matrix; The weight corresponding to each running state feature value is calculated based on the information entropy of each running state feature value.

[0092] In one embodiment of this application, the status monitoring module 23 is further configured to: Within a preset monitoring period, multiple sets of target operating data are acquired based on a preset acquisition window. The acquisition window is for continuously acquiring target operating data a preset number of times. Each set of target operating data corresponds to multiple operating status characteristic values ​​of electrical equipment during a preset acquisition window period. An initial feature matrix is ​​constructed based on multiple sets of target operation data and multiple operation state feature values; the rows of the initial feature matrix represent each set of target operation data, and the columns of the initial feature matrix represent multiple operation state feature values ​​corresponding to each set of target operation data; The initial feature matrix is ​​standardized to obtain the target feature matrix.

[0093] In one embodiment of this application, the status monitoring module 23 is further configured to: A fault sample library is constructed based on the historical fault types of electrical equipment and their corresponding operating status indicators; The K-means clustering algorithm is used to cluster the operating status indicators in the fault sample library to obtain the cluster centers of the indicators corresponding to the operating status levels of electrical equipment. Based on the index cluster center and the preset safety margin, the threshold range corresponding to the operating status level of electrical equipment is obtained. By matching operational status indicators with threshold ranges, the corresponding operational status level of electrical equipment can be determined.

[0094] In one embodiment of this application, the status monitoring module 23 is further configured to: Determining the operating status level of electrical equipment based on operating status indicators also includes: In response to the electrical equipment's operating status level being a slightly abnormal state, the target acquisition frequency of the target sensor is updated based on a preset adjustment step size to obtain a new target acquisition frequency; In response to the electrical equipment's operating status level being a moderate abnormal state, a corresponding fault warning signal for the electrical equipment is generated; In response to the electrical equipment's operating status level being a severe abnormal state, a corresponding protection control signal for the electrical equipment is generated.

[0095] See Figure 3 , Figure 3 This is a schematic block diagram of an electronic device provided according to an embodiment of this application. Figure 3 The electronic device 300 in this embodiment may include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memories 304 store computer programs, including program instructions. The processors 301 execute the program instructions stored in the memories 304. Specifically, the processors 301 are configured to invoke the program instructions to perform the functions of the modules in the aforementioned system embodiments, for example... Figure 2 The functions of the sampling frequency determination module 21, data processing module 22, and status monitoring module 23 are shown.

[0096] It should be understood that, in the embodiments of this application, the processor 301 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0097] Input device 302 may include a touchpad, a fingerprint sensor (for collecting the user's fingerprint information and fingerprint orientation information), a microphone, etc., and output device 303 may include a display (LCD, etc.), a speaker, etc.

[0098] The memory 304 may include read-only memory and random access memory, and provides instructions and data to the processor 301. A portion of the memory 304 may also include non-volatile random access memory. For example, the memory 304 may also store information such as a fault sample library.

[0099] In specific implementations, the processor 301, input device 302, and output device 303 described in the embodiments of this application can execute the implementation method described in the intelligent monitoring method for electrical equipment provided in the embodiments of this application, or they can execute the implementation method of the electronic equipment described in the embodiments of this application, which will not be repeated here.

[0100] In another embodiment of this application, a computer-readable storage medium is provided. This computer-readable storage medium stores a computer program, which includes program instructions. When executed by a processor, the program instructions implement all or part of the processes in the methods described above. Alternatively, the computer program can instruct related hardware to implement these processes. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include any entity or system capable of carrying computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0101] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the foregoing embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD) card, flash card, etc., equipped on the electronic device. Furthermore, the computer-readable storage medium can include both internal and external storage units of the electronic device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0102] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0103] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the electronic devices and units described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

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

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

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

[0107] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for intelligent monitoring of electrical equipment, characterized in that, include: The initial acquisition frequency of the target sensor is determined based on the current environmental temperature, humidity, and seasonal information. The initial acquisition frequency is then adjusted based on the current load of the electrical equipment to obtain the target acquisition frequency for collecting different types of operating data from the electrical equipment. The target sensor is a sensor that collects different types of operating data from the electrical equipment. Based on the target acquisition frequency, different types of operating data of electrical equipment are collected, and the operating data is preprocessed to obtain the target operating data; Feature extraction is performed on the target operating data to obtain the operating status index of the electrical equipment, and the operating status level of the electrical equipment is determined based on the operating status index; The step of extracting features from the target operating data to obtain the operating status indicators of the electrical equipment includes: The target operating data is input into a pre-trained convolutional neural network model to obtain multiple operating state feature values ​​of the electrical equipment; Within a preset monitoring period, multiple sets of target operating data are acquired based on a preset acquisition window. The acquisition window is for continuously acquiring the target operating data a preset number of times. Each set of target operating data corresponds to multiple operating status feature values ​​of electrical equipment during a preset acquisition window operation period. An initial feature matrix is ​​constructed based on the multiple sets of target operation data and the multiple operation state feature values; the rows of the initial feature matrix represent each set of target operation data, and the columns of the initial feature matrix represent the multiple operation state feature values ​​corresponding to each set of target operation data; The initial feature matrix is ​​standardized to obtain the target feature matrix; Calculate the information entropy corresponding to each running state feature value in the target feature matrix; The weight corresponding to each running state feature value is calculated based on the information entropy of each running state feature value; The weight calculation formula for each running state feature value is as follows: in, The weight of the j-th operating state feature value reflects the importance of this feature to the overall state assessment of the electrical equipment. The sum of all feature weights is 1. The information entropy represents the j-th running state feature value; m represents the total number of running state feature values; Based on the multiple operating status feature values ​​and their corresponding weights, a weighted fusion is performed to obtain the operating status index of the electrical equipment. The step of determining the operating status level of the electrical equipment based on the operating status indicators includes: A fault sample library is constructed based on the historical fault types of electrical equipment and their corresponding operating status indicators; The operating status indicators in the fault sample library are clustered based on the K-means clustering algorithm to obtain the cluster centers of the indicators corresponding to the operating status levels of electrical equipment. Based on the cluster centers of the indicators and the preset safety margin, the threshold range corresponding to the operating status level of the electrical equipment is obtained; The operating status indicators are matched with the threshold range to determine the operating status level of the electrical equipment.

2. The intelligent monitoring method for electrical equipment as described in claim 1, characterized in that, The determination of the initial acquisition frequency of the target sensor based on the current environmental temperature, humidity, and seasonal information includes: The initial acquisition frequency of the target sensor is determined based on the first formula, which is: in, Indicates the initial sampling frequency. Indicates the reference sampling frequency. Indicates the temperature effect coefficient. This represents the difference between the current temperature information and the standard temperature information. Indicates the humidity influence coefficient. This represents the difference between the current humidity information and the standard humidity information. This represents the coefficient corresponding to the current season; The step of adjusting the initial acquisition frequency based on the current load of the electrical equipment to obtain the target acquisition frequency of the target sensor includes: The initial acquisition frequency is adjusted based on the second formula to obtain the target acquisition frequency of the target sensor. The second formula is: in, Indicates the target acquisition frequency. This represents the load impact factor, where L represents the current load on the electrical equipment.

3. The intelligent monitoring method for electrical equipment as described in claim 1, characterized in that, The operating status levels of the electrical equipment include normal, slightly abnormal, moderately abnormal, and severely abnormal; The step of determining the operating status level of the electrical equipment based on the operating status indicators further includes: In response to the electrical equipment's operating status level being a slightly abnormal state, the target acquisition frequency of the target sensor is updated based on a preset adjustment step size to obtain a new target acquisition frequency; In response to the electrical equipment's operating status level being a moderate abnormal state, a corresponding fault warning signal for the electrical equipment is generated; In response to the electrical equipment's operating status level being a severe abnormal state, a corresponding protection control signal for the electrical equipment is generated.

4. An intelligent monitoring system for electrical equipment, characterized in that, include: The sampling frequency determination module is used to determine the initial sampling frequency of the target sensor based on the current environmental temperature, humidity, and seasonal information, and to adjust the initial sampling frequency based on the current load of the electrical equipment to obtain the target sampling frequency for the target sensor to collect different types of operating data of the electrical equipment; the target sensor is a sensor that collects different types of operating data of the electrical equipment. The data processing module is used to collect different types of operating data of electrical equipment based on the target acquisition frequency, and to preprocess the operating data to obtain the target operating data; The status monitoring module is used to extract features from the target operating data to obtain the operating status indicators of the electrical equipment, and to determine the operating status level of the electrical equipment based on the operating status indicators. The status monitoring module is specifically used for: The target operating data is input into a pre-trained convolutional neural network model to obtain multiple operating state feature values ​​of the electrical equipment; Within a preset monitoring period, multiple sets of target operating data are acquired based on a preset acquisition window. The acquisition window is for continuously acquiring the target operating data a preset number of times. Each set of target operating data corresponds to multiple operating status feature values ​​of electrical equipment during a preset acquisition window operation period. An initial feature matrix is ​​constructed based on the multiple sets of target operation data and the multiple operation state feature values; The rows of the initial feature matrix represent each set of target running data, and the columns of the initial feature matrix represent multiple running state feature values ​​corresponding to each set of target running data; The initial feature matrix is ​​standardized to obtain the target feature matrix; Calculate the information entropy corresponding to each running state feature value in the target feature matrix; The weight corresponding to each running state feature value is calculated based on the information entropy of each running state feature value; The weight calculation formula for each running state feature value is as follows: in, The weight of the j-th operating state feature value reflects the importance of this feature to the overall state assessment of the electrical equipment. The sum of all feature weights is 1. The information entropy represents the j-th running state feature value; m represents the total number of running state feature values; Based on the multiple operating status feature values ​​and their corresponding weights, a weighted fusion is performed to obtain the operating status index of the electrical equipment. The status monitoring module is also specifically used for: A fault sample library is constructed based on the historical fault types of electrical equipment and their corresponding operating status indicators; The operating status indicators in the fault sample library are clustered based on the K-means clustering algorithm to obtain the cluster centers of the indicators corresponding to the operating status levels of electrical equipment. Based on the cluster centers of the indicators and the preset safety margin, the threshold range corresponding to the operating status level of the electrical equipment is obtained; The operating status indicators are matched with the threshold range to determine the operating status level of the electrical equipment.

5. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 3.

6. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 3.

Images