Power equipment operation state analysis method, system, equipment and medium

Abstract

The invention relates to the technical field of power equipment operation state analysis, and discloses a power equipment operation state analysis method, system, equipment and medium, and the method comprises the steps: obtaining a first parameter of target power equipment, carrying out the first classification of the first parameter, carrying out the first preprocessing of the first parameter after the first classification, and obtaining a second parameter, and establishing a first analysis model, training the first analysis model according to the second parameter to obtain the trained first analysis model, recording the trained first analysis model as a second analysis model, and inputting the second analysis model in response to the real-time data of the target power equipment to obtain a power equipment operation state analysis result. According to the method, the complexity and diversity of the operation state of the power equipment are fully considered, and accurate analysis and real-time monitoring of the operation state of the power equipment are realized through fine parameter classification and preprocessing and an advanced four-layer network architecture analysis model.

Description

Technical Field

[0001] The present invention relates to the technical field of power equipment operation state analysis, and particularly to a method, system, device and medium for analyzing the operation state of power equipment. Background Art

[0002] With the continuous increase in the scale and complexity of modern power systems, the demand for accurate analysis and real-time monitoring of the operation state of power equipment is growing. Traditional power equipment monitoring methods mainly rely on regular maintenance and manual inspections. This method is not only inefficient but also difficult to detect potential problems in a timely manner, which may lead to equipment failures or power outages. In addition, the operation state of power equipment is affected by various factors, including but not limited to internal parameters (such as temperature, voltage, etc.) and external environmental parameters (such as weather conditions, geographical factors, etc.). The complex interaction between these parameters makes it more difficult to accurately evaluate the state of power equipment.

[0003] In recent years, with the development of artificial intelligence and machine learning technologies, using data-driven methods to analyze the operation state of power equipment has gradually become a research hotspot. By collecting and analyzing a large amount of historical data, an effective model can be established to predict possible problems of power equipment, thereby realizing preventive maintenance. However, existing models often fail to fully consider the differences between dynamic parameters and static parameters and their different impacts on the operation state of power equipment, resulting in inaccurate prediction results. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a method, system, device and medium for analyzing the operation state of power equipment, which can solve the problem of inaccurate analysis of the operation state of power equipment in the prior art, especially the problem of not fully considering the differences between dynamic parameters and static parameters.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a method for analyzing the operation state of power equipment, including: Obtaining first parameters of a target power equipment and performing a first classification on the first parameters; The first classification is used to classify the first parameters into first dynamic parameters and first static parameters; Performing a first preprocessing on the first parameters after the first classification to obtain second parameters, where the second parameters include second dynamic parameters and second static parameters; Establishing a first analysis model, where the first analysis model includes two input ends, a four-layer network architecture and an output end; The four-layer network architecture includes a first feature judgment network architecture and a second feature judgment network architecture connected in parallel, a third state analysis network structure connected in series with the first feature judgment network architecture and the second feature judgment network architecture, and a fourth decision-making network structure connected in series with the third state analysis network structure; Train the first analysis model according to the second parameter to obtain the first analysis model after training, denoted as the second analysis model; In response to the real-time data of the target power equipment being input into the second analysis model, obtain the analysis result of the operating state of the power equipment.

[0007] As a preferred solution of the power equipment operating state analysis method described in the present invention, wherein: the first classification of the first parameter includes: The first parameter includes a first internal parameter and a second external parameter; The first classification includes: Preset a first parameter classification library and a second parameter classification judgment logic; Obtain the classified parameters and unclassified parameters in the first internal parameter and the second external parameter according to the first parameter classification library; Perform classification judgment on the unclassified parameters according to the second parameter classification judgment logic to obtain a classification judgment result; Obtain a first dynamic parameter and a first static parameter according to the classified parameters and the classification judgment result.

[0008] As a preferred solution of the power equipment operating state analysis method described in the present invention, wherein: the first classification further includes: The initial first parameter classification library is a blank library; The initial first classification only performs initial classification judgment through the second parameter classification judgment logic. The initial first classification performs classification judgment on the initial first parameter according to the second parameter classification judgment logic; Establish a non-blank first parameter classification library according to the parameters in the initial classification judgment result.

[0009] As a preferred solution of the power equipment operating state analysis method described in the present invention, wherein: the first analysis model further includes: The two input ends are respectively the input ends of the first feature judgment network architecture and the second feature judgment network architecture; One of the first feature judgment network architecture and the second feature judgment network architecture is used to process the second dynamic data in the second parameter, and the other is used to process the second static data in the second parameter; The output of the first feature judgment network architecture is the first state analysis result; The output of the second feature judgment network architecture is the second state analysis result; The first state analysis result and the second state analysis result are the inputs of the third state analysis network structure.

[0010] As a preferred solution of the power equipment operation state analysis method of the present invention, wherein: the first analysis model further includes: The third state analysis network structure is used to perform a third state analysis on the first state analysis result and the second state analysis result; A preset third state analysis logic, and the third state analysis network structure is established according to the third state analysis logic; The result of the third state analysis is the final analysis state of the target power equipment; Taking the final analysis state as an input, input it into the fourth decision-making network structure.

[0011] As a preferred solution of the power equipment operation state analysis method of the present invention, wherein: the first analysis model further includes: The fourth decision-making network structure is used to decide a state processing strategy according to the final analysis state; The state processing strategy includes several processing strategies for different operating states of the target power equipment.

[0012] As a preferred solution of the power equipment operation state analysis method of the present invention, wherein: each of the four-layer network architectures includes different network objective functions and network constraint conditions.

[0013] In a second aspect, the present invention provides a power equipment operation state analysis system, including: A data acquisition and classification module, configured to acquire first parameters of a target power equipment and perform a first classification on the first parameters; The first classification is used to classify the first parameters into first dynamic parameters and first static parameters; A preprocessing module, configured to perform a first preprocessing on the first parameters after the first classification to obtain second parameters, where the second parameters include second dynamic parameters and second static parameters; A model establishment module, configured to establish a first analysis model, where the first analysis model includes two input ends, a four-layer network architecture, and an output end; The four-layer network architecture includes a first feature judgment network architecture and a second feature judgment network architecture in parallel, a third state analysis network structure connected in series with the first feature judgment network architecture and the second feature judgment network architecture, and a fourth decision-making network structure connected in series with the third state analysis network structure; A model training module, configured to train the first analysis model according to the second parameter to obtain the first analysis model after training, denoted as the second analysis model; An analysis module, configured to obtain an analysis result of the operating state of the power equipment in response to real-time data of the target power equipment being input into the second analysis model.

[0014] In a third aspect, the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, the steps of the method described above are implemented.

[0015] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method described above are implemented.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention proposes a method, system, device and medium for analyzing the operating state of power equipment, obtains the first parameter of the target power equipment, and performs a first classification on the first parameter; performs a first preprocessing on the first parameter after the first classification to obtain a second parameter, where the second parameter includes a second dynamic parameter and a second static parameter; establishes a first analysis model, where the first analysis model includes two input ends, a four-layer network architecture and an output end; trains the first analysis model according to the second parameter to obtain the first analysis model after training, denoted as the second analysis model; obtains an analysis result of the operating state of the power equipment in response to real-time data of the target power equipment being input into the second analysis model. This method fully considers the complexity and diversity of the operating state of power equipment, and through fine parameter classification and preprocessing, as well as an advanced four-layer network architecture analysis model, realizes the accurate analysis and real-time monitoring of the operating state of power equipment. Compared with traditional power equipment monitoring methods, the method of the present invention not only improves the analysis efficiency, but also can timely discover potential problems, avoiding economic losses and social impacts caused by equipment failures or power outages. In addition, this method can also dynamically adjust the analysis model according to the historical data and real-time data of power equipment, improving the accuracy and reliability of the analysis. Therefore, the present invention has broad application prospects and important practical significance in the technical field of analyzing the operating state of power equipment.

[0017] Specifically, first, obtaining and classifying the first parameters of the target power equipment helps to distinguish parameters of different natures and provides a basis for subsequent processing. Next, preprocessing the classified parameters can further purify and standardize the data, improving the data quality. Then, establishing a first analysis model with two input ends, a four-layer network architecture, and one output end can make full use of the dynamic and static characteristics of the data and improve the accuracy of analysis. By training the first analysis model according to the second parameter, a second analysis model after training is obtained, which can ensure that the model can adapt to changes in the actual operating state of the power equipment. Finally, in response to the real-time data of the target power equipment being input into the second analysis model, an analysis result of the operating state of the power equipment is obtained, enabling real-time monitoring and accurate evaluation of the operating state of the power equipment. Each step provides key support for the final accurate analysis and real-time monitoring, ensuring the effectiveness and reliability of the entire method. Description of the Drawings

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0019] Figure 1 It is a flowchart of a method for analyzing the operating state of a power equipment provided by an embodiment of the present invention.

[0020] Figure 2 It is a first classification logic schematic diagram of a method for analyzing the operating state of a power equipment provided by an embodiment of the present invention.

[0021] Figure 3 It is a fourth decision network structure diagram of a method for analyzing the operating state of a power equipment provided by an embodiment of the present invention.

[0022] Figure 4 It is a first analysis model structure diagram of a method for analyzing the operating state of a power equipment provided by an embodiment of the present invention.

[0023] Figure 5 It is an internal structure diagram of a computer device of a method for analyzing the operating state of a power equipment provided by an embodiment of the present invention. Detailed Embodiments

[0024] To make the above objects, features, and advantages of the present invention more apparent and understandable, the following provides a detailed description of the specific embodiments of the present invention in conjunction with the accompanying drawings of the specification. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present invention.

[0025] Example 1, referring to Figures 1 - 5 , which is the first embodiment of the present invention. This embodiment provides a method for analyzing the operating state of power equipment, including: Before introducing the embodiments of the present application in detail, for clarity, some related concepts are first explained.

[0026] A Convolutional Neural Network (CNN) is a deep learning model, especially suitable for processing data with a grid structure, such as images. It automatically extracts feature representations from the input data through a series of layers: Convolutional layer: Scans the input data using filters to identify patterns or features within local regions.

[0027] Activation function: Usually adopts non-linear functions such as ReLU to increase the expressive power of the model.

[0028] Pooling layer: Reduces the spatial size of the data, reduces computational complexity, and improves invariance to input transformations.

[0029] Fully connected layer: Converts the previously obtained feature maps into specific outputs, such as classification results.

[0030] Through the way of sharing weights and local receptive fields, CNN can effectively capture the spatial hierarchical structure of the input data.

[0031] Reinforcement Learning (RL) is a branch of machine learning that focuses on how to interact with the environment based on environmental behaviors and learn the optimal behavior strategy through trial and error. Its core concepts include: Agent: The learner that executes actions and interacts with the environment.

[0032] Environment: The world where the agent is located, which gives feedback according to the agent's actions.

[0033] Reward: The numerical feedback provided by the environment to the agent, guiding the agent to learn the optimal strategy.

[0034] Policy: The method by which an agent selects actions based on the current state.

[0035] Value Function: Evaluates the goodness of a particular policy in a given state.

[0036] The goal of reinforcement learning is to find a policy that maximizes the cumulative reward obtained by the agent in the long run.

[0037] Network Architecture refers to the specific structural arrangement for constructing an artificial neural network or other types of computational models. This includes, but is not limited to: The number of layers and the type of each layer: such as the input layer, hidden layers (which can be convolutional layers, recurrent layers, etc.), and output layer.

[0038] The number of neurons in each layer: Determines the width of the layer and affects the capacity and complexity of the model.

[0039] The connection methods between layers: For example, fully connected, convolutional connection, recursive connection, etc.

[0040] The choice of activation function: Determines how each neuron transforms its input into an output.

[0041] The design of forward propagation and backpropagation algorithms: Used to calculate the prediction error and adjust the network weights.

[0042] In the existing related technologies, there are some problems. For example, traditional power equipment monitoring methods often rely on manual inspections and empirical judgments, which are not only inefficient but also difficult to accurately capture potential faults of the equipment. In addition, although some existing automated monitoring methods can collect equipment data in real time, they have deficiencies in data processing and analysis, resulting in inaccurate analysis results and being unable to provide strong support for equipment maintenance and management.

[0043] This application provides a method that can effectively solve the above-mentioned problems. Next, multiple embodiments will be combined to elaborate in detail on how to implement this power equipment operation status analysis method; Figure 1 The method flow chart of a power equipment operation status analysis method is shown, including: S101, Obtain the first parameters of the target power equipment and perform a first classification on the first parameters; It should be noted that the target power equipment in this application can include transformers, generators, circuit breakers, etc. These power equipment generate a large amount of data during operation, including parameters such as current, voltage, and temperature. These data are important bases for analyzing the operation status of power equipment. The purpose of the first classification is to classify these parameters according to their properties and characteristics, providing a basis for subsequent processing and analysis.

[0044] In the embodiment of the present application, the first classification is used to divide the first parameter into a first dynamic parameter and a first static parameter.

[0045] It should be noted that the complexity of the current power system is often difficult to comprehensively reflect through individual dynamic or static parameters. Therefore, in the present application, through the first classification, the first parameter is subdivided into a first dynamic parameter and a first static parameter to more accurately capture the operating state of power equipment.

[0046] It should also be noted that traditionally, dynamic parameters usually refer to parameters that change over time, such as instantaneous values of current and voltage, which can reflect the real-time operating conditions of power equipment. Static parameters, on the other hand, refer to relatively stable parameters, such as the temperature and resistance value of equipment, which more reflect the long-term operating conditions and performance of power equipment. However, this traditional classification does not fit most power equipment, so the present application proposes a method that can perform detailed and accurate classification.

[0047] In the embodiment of the present application, the first classification of the first parameter includes: The first parameter includes a first internal parameter and a second external parameter; Specifically, when the power equipment is a transformer, the first internal parameter includes at least one of the following: oil temperature, winding temperature, oil level, winding current, winding voltage, power, frequency deviation, rated voltage, rated power, and rated capacity, etc.; The second external parameter includes at least one of the following: ambient temperature, humidity, load change, operating environment, altitude, and noise level, etc.

[0048] When the power equipment is a generator, the first internal parameter includes at least one of the following: stator current, stator voltage, rotor current, rotor voltage, power factor, frequency, active power, reactive power, temperature, and vibration, etc.; The second external parameter includes at least one of the following: ambient temperature, humidity, cooling medium flow rate, cooling medium temperature, altitude, and load change, etc.

[0049] When the power equipment is a circuit breaker, the first internal parameter includes at least one of the following: number of operations, opening and closing time, contact wear condition, insulation resistance, leakage current, and gas pressure, etc.; The second external parameter includes at least one of the following: ambient temperature, humidity, installation location, operating frequency, and external electric or magnetic field interference, etc.

[0050] It should be noted that different first internal parameters and second external parameters can be set for other different power equipment, and the selection and setting of specific parameters depend on the type, structure, and operating characteristics of the power equipment. By classifying the first parameters in detail, the operating state of the power equipment can be understood more deeply, providing more accurate and comprehensive data support for subsequent analysis and processing. This application does not make any limitations.

[0051] It should be emphasized that some static parameters in actual operation need to be calculated based on real-time data or directly obtained through sensors. These belong to the preliminary operation steps in this application. For example, the noise level can be collected and calculated in real time through sensors such as microphones, the ambient temperature and humidity can be obtained through temperature and humidity sensors, and the load change can be monitored and calculated through current and voltage sensors. The real-time acquisition and monitoring of these static parameters help to more comprehensively reflect the operating state of the power equipment and improve the accuracy of analysis.

[0052] It should be noted that the advantage of dividing into the first internal parameters and the second external parameters is that it can more precisely reflect the operating state of the power equipment, taking into account various influencing factors inside and outside the equipment. The first internal parameters mainly reflect the operating conditions of the power equipment itself, such as oil temperature, winding temperature, etc. These parameters are crucial for judging the health state of the equipment. The second external parameters consider the environmental conditions where the equipment is located, such as ambient temperature, humidity, etc. Although these parameters do not directly reflect the internal state of the equipment, they also have an impact on the operation of the equipment. By comprehensively considering these two types of parameters, the operating state of the power equipment can be more comprehensively evaluated, improving the accuracy of analysis.

[0053] It should also be noted that to avoid excessive length, the subsequent content of this application only takes the "transformer" as an exemplary electronic device for description.

[0054] As Figure 2 shown, in the embodiment of this application, the first classification includes: A preset first parameter classification library and a second parameter classification judgment logic; Obtain the classified parameters and unclassified parameters in the first internal parameters and the second external parameters according to the first parameter classification library; Perform a classification judgment on the unclassified parameters according to the second parameter classification judgment logic to obtain a classification judgment result; Obtain a first dynamic parameter and a first static parameter according to the classified parameters and the classification judgment result.

[0055] Exemplarily, for example, when the power device is a transformer, and the dynamic parameters in the first classification library are: "power, frequency offset", and the static parameters are: "rated voltage, rated power, and rated capacity", then the classified parameters in the first internal parameters and the second external parameters are several of "power, frequency offset, rated voltage, rated power, and rated capacity"; the unclassified parameters are the parameters in the first internal parameters and the second external parameters excluding "power, frequency offset, rated voltage, rated power, and rated capacity". Furthermore, perform a second parameter classification judgment logic judgment on the unclassified parameters (that is, whether there is a judgment logic for the unclassified parameters), classify and judge those that meet the judgment logic, and judge the importance of those that do not meet the parameters; If the importance is insufficient, do not consider the parameter that does not meet the requirements; If the importance is sufficient, regenerate the second parameter classification judgment logic in combination with the characteristics of the parameter that does not meet the requirements; For example, if the parameter that does not meet the requirements is the oil temperature change trend, and its importance is sufficient because it can reflect the heat dissipation situation and internal temperature change of the transformer, which is of great significance for judging whether the transformer is overheated. Therefore, in combination with the characteristics of the oil temperature change trend, the second parameter classification judgment logic can be regenerated, classify it as a dynamic parameter, and separately generate the judgment criterion for the oil temperature change trend, and incorporate the judgment criterion into the second parameter classification judgment logic to complete the update.

[0056] It should be noted that the importance judgment standard can obtain the threshold through experiments with historical data, judge through the threshold, or establish an importance judgment library through expert experience for judgment. There are many other ways to judge, and this application does not limit it.

[0057] In the embodiment of the present application, the first classification further includes: The initial first parameter classification library is an empty library; The initial first classification only performs an initial classification judgment through the second parameter classification judgment logic. The initial first classification classifies and judges the initial first parameters according to the second parameter classification judgment logic; Establish a non-empty first parameter classification library according to the parameters in the initial classification judgment result.

[0058] Exemplarily, assume that in the initial state, there is a target power device (such as a transformer) in the example, and its first parameter set P = {p1, p2,..., pn} contains n different parameters (such as voltage, current, temperature, etc.).

[0059] Initially, the dynamic parameter set D = {} in the first parameter classification library is an empty set, and the static parameter set S = {} is an empty set.

[0060] The second parameter classification judgment logic evaluates and classifies each parameter pi in P according to a preset time change pattern or other characteristics.

[0061] Suppose that after the initial classification, the results are as follows: The dynamic parameter set D’ = {p1, p3, p5}, which satisfies the dynamic parameter judgment in the second parameter classification judgment logic.

[0062] The static parameter set S’ = {p2, p4, p6}, which satisfies the static parameter judgment in the second parameter classification judgment logic.

[0063] Based on the results of the above initial classification, a non-empty first parameter classification library can be established in the example: The updated dynamic parameter set D = D’.

[0064] The updated static parameter set S = S’.

[0065] Therefore, after the initial classification is completed, the first parameter classification library is no longer blank, but contains the specific dynamic parameter set D and static parameter set S. This classification library can be further updated and improved according to subsequent data and analysis to adapt to different types or models of power equipment. For the same type of power equipment, the generated classification library can be directly applied; while for different models of equipment, the corresponding classification library needs to be regenerated.

[0066] In an optional embodiment, the second parameter classification judgment logic includes: Presetting a first cycle threshold and a second fluctuation threshold; If there is a parameter in the unclassified parameters whose fluctuation (i.e., data change) generated within the first cycle threshold is greater than the second fluctuation threshold, then this parameter is recorded as a dynamic acquisition parameter, otherwise it is a static parameter; Exemplarily, suppose the oil temperature of a certain transformer is being evaluated as an unclassified parameter, and the first cycle threshold is set to one day (24 hours), and the second fluctuation threshold is set to 5 degrees Celsius. If within 24 consecutive hours, the oil temperature change exceeds 5 degrees Celsius, then according to the above logic, the oil temperature will be classified as a dynamic parameter; if it does not exceed this fluctuation threshold, then the oil temperature will be regarded as a static parameter.

[0067] It should be noted that the first cycle threshold is a value that can be adjusted according to the needs of relevant technical personnel. The second fluctuation threshold is initially an empirical value, which is obtained based on previous historical data. After the initial judgment, the second fluctuation threshold is updated according to the cycle in the set first cycle threshold, that is, when entering the next cycle, the value of the second fluctuation threshold is re-obtained according to the data of this cycle and the previous historical data.

[0068] It should also be noted that the second fluctuation threshold includes several different branch thresholds. The number of branch thresholds is determined according to the number of dynamic parameters or static parameters obtained from historical data. For example, if there are 5 dynamic parameters and 3 static parameters obtained from historical data, then the number of branch thresholds is 5 + 3 = 8. These branch thresholds correspond to different types of parameters respectively to more finely control the classification process. In practical applications, when the fluctuation of the unclassified parameter exceeds its corresponding branch threshold, the parameter is classified as a dynamic parameter; otherwise, it is regarded as a static parameter.

[0069] In this way, the accuracy and flexibility of parameter classification can be further improved, so as to better support the analysis and monitoring of the operating state of power equipment, and avoid the one-sidedness caused by the traditional selection of dynamic data or static data.

[0070] The combination of the dynamic parameters in the above classified parameters and the dynamic parameters in the classification judgment result is denoted as the first dynamic parameter, and the combination of the static parameters in the above classified parameters and the static parameters in the classification judgment result is denoted as the second static parameter.

[0071] It should be noted that as power equipment is continuously updated and upgraded, the data types and characteristics generated by it will also change, and traditional classification methods may not be able to adapt to these changes. The first parameter classification library and the second parameter classification judgment logic of this application can both be updated. The advantage of this is that it can adapt to different types of power equipment and different operating environments, and improve the accuracy and flexibility of classification.

[0072] It should also be noted that obtaining the first parameters of the target power equipment and performing the first classification on the first parameters can provide more accurate and comprehensive data support for subsequent analysis and processing. By classifying the first parameters in detail and accurately, this application can more deeply understand the operating state of the power equipment, capture the dynamic and static characteristics of the equipment, so as to provide a scientific basis for equipment health monitoring, fault prediction, and maintenance decision-making. In addition, this classification method helps to improve the efficiency of data analysis, reduce unnecessary data redundancy, and make the analysis results more intuitive and easy to understand. In the next steps, this application will further construct an analysis model for the operating state of power equipment based on these classified parameters to achieve real-time monitoring and accurate analysis of the operating state of power equipment.

[0073] S102. Perform first preprocessing on the first parameters after the first classification to obtain second parameters, where the second parameters include second dynamic parameters and second static parameters; In an optional embodiment, the first preprocessing may include, but is not limited to, operations such as data cleaning, data normalization, and data smoothing. Data cleaning aims to remove outliers, missing values, or duplicate data to ensure the accuracy and reliability of the data used for analysis. Data normalization scales the data proportionally to a small specific interval, such as between 0 and 1, so that parameters with different dimensions or magnitudes can be compared and analyzed under the same standard. Data smoothing is used to reduce noise in the data, improve the smoothness and continuity of the data, and thus more realistically reflect the operating state of the power equipment.

[0074] It should be noted that directly using traditional preprocessing steps cannot meet the accuracy and timeliness requirements for analyzing the operating state of power equipment. Traditional preprocessing steps often only process a single type of data and do not take into account the complexity and diversity of the data on the operating state of power equipment. Therefore, in the embodiments of this application, customized first preprocessing steps are proposed for the first parameters after the first classification.

[0075] In the embodiments of this application, for the first dynamic parameters and the first static parameters, different preprocessing methods and strategies are selected according to their characteristics and analysis requirements. For example, for dynamic parameters that change rapidly over time, more refined time series analysis methods may be needed for preprocessing. For relatively stable static parameters, more attention may be paid to the integrity and accuracy of the data, and the preprocessing process may be relatively simple. Through such preprocessing steps, a higher-quality data foundation can be provided for subsequent analysis and modeling.

[0076] Specifically, the preprocessing of dynamic parameters includes: For the first dynamic parameters that change rapidly over time (such as power, frequency offset, etc.), their main characteristic is significant time dependence. Therefore, special attention needs to be paid to the time series characteristics of the data during the preprocessing process.

[0077] Outlier detection and handling: Use statistical methods or machine learning algorithms to identify and handle outliers, such as removing extreme values based on the 3σ principle.

[0078] Missing value filling: Adopt interpolation methods (linear interpolation, spline interpolation) or model prediction-based methods to fill in the missing data points.

[0079] Smoothing processing: Apply techniques such as moving average and exponential smoothing to reduce noise interference and highlight trend features.

[0080] Decomposition analysis: Use seasonal decomposition (such as STL decomposition) to separate the long-term trend, periodic components, and random fluctuation parts.

[0081] Feature extraction: Extract useful features from the time series, such as maximum value, minimum value, mean value, variance, slope change rate, etc.; It is also possible to calculate autocorrelation coefficients, cross-correlation coefficients, etc. to capture the relationships between parameters.

[0082] Standardization / Normalization: Convert data with different dimensions to the same scale for subsequent modeling.

[0083] Update the classification library: Based on the latest data analysis results, adjust the classification logic of dynamic parameters and update it to the parameter classification library.

[0084] The preprocessing of static parameters includes: Relatively stable static parameters (such as rated voltage, rated power, etc.), with the focus on ensuring the integrity and accuracy of the data, and the preprocessing process may be relatively simple.

[0085] Consistency check: Ensure the consistency of all records and avoid the occurrence of duplicate entries or incorrect inputs.

[0086] Verify data integrity: Check whether there is missing key information and, if necessary, complete the data through supplementary investigations or other means.

[0087] Format unification: Ensure that all data is stored in a consistent format for convenient subsequent processing.

[0088] Unit conversion: If there is data with different units, it needs to be uniformly converted to standard units.

[0089] Accuracy verification: Verify the accuracy of the data by comparing it with data from other reliable sources.

[0090] Reliability assessment: Evaluate the reliability of the data source and determine whether additional data support or correction is required.

[0091] Archiving management: Properly store the static parameters after cleaning and conversion, and establish a reliable database or archival system for long-term management and query.

[0092] Exemplarily, when analyzing the operating status of a power transformer and the dynamic parameter is the oil temperature: For the oil temperature, use the 3σ principle (i.e., data points outside the range of the mean ± 3 times the standard deviation are considered abnormal) to identify and eliminate outliers. Use linear interpolation to fill in the missing oil temperature data points. Apply a moving average filter (e.g., with a window size of 5 hours) to reduce the impact of short-term fluctuations and better display the long-term trend. Use the STL decomposition method to divide the oil temperature data into a trend component, a seasonal component, and a residual component for subsequent feature extraction. Calculate the hourly oil temperature change rate (i.e., the oil temperature at the current moment minus the oil temperature at the previous hour), as well as the daily maximum and minimum temperature differences. Scale all the oil temperature data to the range [0, 1] to ensure that data with different dimensions can be effectively processed by the model. According to the latest data analysis results, adjust the classification logic of the dynamic parameters and update it to the parameter classification library. For example, if it is found that the oil temperature change rate shows significant fluctuations within a specific time period, redefine it as an important dynamic feature and add it to the classification library (this step is set to facilitate the update of the above-mentioned second parameter classification judgment logic, and the parameters used in the update of the second parameter classification judgment logic can directly retrieve the data from this parameter classification library).

[0093] It should be noted that performing the first preprocessing on the first parameter after the first classification to obtain the second parameter can significantly improve the accuracy and efficiency of subsequent analysis and modeling. The second parameter after preprocessing, whether it is the second dynamic parameter or the second static parameter, is closer to the actual operating state of the power equipment, reducing the interference of noise and outliers. Such a data basis provides a strong guarantee for constructing an accurate analysis model. In addition, the preprocessing step also takes into account the different characteristics of dynamic parameters and static parameters and adopts targeted processing strategies, which makes the results of data preprocessing more in line with the actual needs of power equipment operating state analysis. Therefore, in subsequent steps, whether it is feature extraction, model training, or state prediction, more reliance can be placed on the preprocessed data, thereby improving the reliability and practicality of the entire analysis process.

[0094] S103, establish a first analysis model, the first analysis model includes two input ends, a four-layer network architecture, and an output end; In the embodiment of the present application, the four-layer network architecture includes a first feature judgment network architecture and a second feature judgment network architecture in parallel, a third state analysis network structure connected in series with the first feature judgment network architecture and the second feature judgment network architecture, and a fourth decision network structure connected in series with the third state analysis network structure, as Figure 4 shown.

[0095] In the embodiment of the present application, the first analysis model further includes: The two input ends are respectively the input ends of the first feature judgment network architecture and the second feature judgment network architecture; One of the first feature judgment network architectures and the second feature judgment network architecture is used to process the second dynamic data in the second parameter, and the other is used to process the second static data in the second parameter; The output of the first feature judgment network architecture is the first state analysis result; The output of the second feature judgment network architecture is the second state analysis result; The first state analysis result and the second state analysis result are the inputs of the third state analysis network structure.

[0096] In the embodiment of the present application, the first analysis model further includes: The third state analysis network structure is used to perform a third state analysis on the first state analysis result and the second state analysis result; A preset third state analysis logic, and the third state analysis network structure is established according to the third state analysis logic; The result of the third state analysis is the final analysis state of the target power equipment; Taking the final analysis state as an input and inputting it into the fourth decision network structure.

[0097] In the embodiment of the present application, the first analysis model further includes: The fourth decision network structure is used to decide a state processing strategy according to the final analysis state; The state processing strategy includes several processing strategies for different operating states of the target power equipment.

[0098] In the embodiment of the present application, each of the four-layer network architectures includes different network objective functions and network constraint conditions.

[0099] In an alternative embodiment, the first analysis model may adopt a deep learning network architecture, such as a convolutional neural network (CNN) or a recurrent neural network (RNN) and its variants, such as a long short-term memory network (LSTM) or a gated recurrent unit (GRU), to adapt to the analysis requirements of power equipment operating states with different complexities. CNN performs well in processing data with spatial structures (such as image data), and can extract local features and perform efficient classification. RNN and its variants are good at processing time series data and can capture the changing trends and periodic features of parameters over time, which is particularly important for the analysis of dynamic parameters. According to the characteristics and analysis requirements of the target power equipment, selecting a suitable deep learning network architecture can better mine the potential information in the data and improve the accuracy and efficiency of the analysis.

[0100] In an optional embodiment, the first analysis model can also adopt an ensemble learning method, such as random forest, gradient boosting decision tree (GBDT), or XGBoost, etc. By combining the prediction results of multiple weak learners, the ensemble learning method can significantly improve the generalization ability and robustness of the model. These methods perform excellently in dealing with high-dimensional data, non-linear relationships, and complex feature interactions, and are suitable for the comprehensive judgment of multiple parameters and features in the operation state analysis of power equipment. By training multiple base learners and integrating their prediction results, the ensemble learning model can more comprehensively capture the operation state characteristics of power equipment and improve the accuracy and reliability of the analysis.

[0101] It should be noted that the model established by the above method cannot be directly associated with the subsequent processing operations for the power equipment state analysis, which will limit the application effect of the model. To solve this problem, the present application proposes an innovative strategy, that is, when designing the first analysis model, the decision logic associated with the subsequent processing operations is pre-embedded.

[0102] It should also be noted that the present application uses a convolutional neural network for design when establishing the first analysis model, and combines reinforcement learning for improvement. Such a design enables the model to have stronger feature extraction ability and state analysis ability when processing power equipment operation state data. The convolutional neural network (CNN) has achieved remarkable results in processing image data with its powerful feature learning ability. Applying it to the analysis of power equipment operation state data can automatically extract the key features in the data, avoiding the cumbersome process of manually designing features in traditional methods. At the same time, the improvement combined with reinforcement learning (RL) enables the model to optimize its decision-making ability in the continuous learning process, so as to more accurately formulate corresponding processing strategies according to the operation state of power equipment.

[0103] Specifically, the specific network objective functions and network constraint conditions of the four networks in the present application can be as follows (select the input of the first feature judgment network architecture as the second dynamic parameter): The first feature judgment network architecture is improved and designed through a convolutional neural network. Its objective is to judge the state of the target power equipment through dynamic parameters and obtain the first state analysis result; The second feature judgment network architecture is improved and designed through a convolutional neural network. Its objective is to judge the state of the target power equipment through static parameters and obtain the second state analysis result; The third state analysis network structure is improved and designed through a convolutional neural network. Its objective is to analyze the operation state of the final target equipment according to the first state analysis result and the second state analysis result, obtain the final operation state type, and provide an accurate input for the subsequent fourth decision network structure; The fourth decision-making network structure is improved and designed through reinforcement learning, and its goal is to determine which strategy to implement according to the types of final operating states.

[0104] Exemplarily, the specific network objective function and network constraint conditions used in this application can be as follows: The network objective function of the first feature judgment network (dynamic parameter processing) is as follows. The goal of this network is to extract the dynamic features of the device operating state through the second dynamic parameters (parameters that change with time such as voltage, current, temperature, etc. and have been preprocessed for the first time), and output the first state analysis result.

[0105] Objective function (loss function): , where: LCE represents the cross-entropy loss function, which measures the difference between the predicted state and the true state . is the regularization term (such as L2 regularization) to prevent overfitting. is the regularization coefficient. is the network weight parameter. N1 represents the number of training samples in the second dynamic parameter.

[0106] The network constraint conditions are as follows: Input constraint: The input is a dynamic parameter tensor , where T is the time step, and D 1 is the dynamic parameter feature dimension (such as the number of parameters such as voltage and current in this application).

[0107] Output constraint: The output is a state probability distribution , where C is the number of state categories (such as normal, overload, fault, etc. selected in this application).

[0108] Activation function constraint: The ReLU activation function is used in the convolutional layer, and the Softmax activation function is used in the fully connected layer.

[0109] Gradient constraint: Gradient clipping is used (such as the gradient norm is limited to in this application) to prevent gradient explosion.

[0110] It should be noted that the first feature judgment network is used to process dynamic parameters (such as time series data), adopts a 1D convolutional neural network (CNN) structure, and combines time series feature extraction.

[0111] The network objective function of the second feature judgment network (static parameter processing) is as follows. The goal of this network is to extract the static features of the device through static parameters (parameters that do not change rapidly with time such as rated power, ambient temperature, etc.), and output the second state analysis result.

[0112] Objective function (loss function): , where LCE is the cross - entropy loss function, measuring the difference between the predicted state and the true state . is the regularization term (such as L1 regularization), preventing overfitting. is the regularization coefficient. is the network weight parameter, and N2 represents the number of training samples in the second static parameter.

[0113] The network constraint conditions are as follows: Input constraint: The input is a static parameter vector , where D 2 is the feature dimension of the static parameter (such as the number of parameters like rated voltage, ambient humidity, etc.).

[0114] Output constraint: The output is a state probability distribution , where C is the number of state categories (such as normal, overload, fault, etc. selected in this application).

[0115] Activation function constraint: The ReLU activation function is used in the fully - connected layer, and the Softmax is used in the output layer.

[0116] Data normalization constraint: The input data needs to be standardized (such as mean = 0, variance = 1).

[0117] It should be noted that the second feature judgment network processes static parameters (such as scalar features), using a fully - connected neural network (FCNN) without a convolutional layer.

[0118] The network objective function of the third state analysis network is as follows. The goal of this network is to fuse the analysis results of dynamic and static features and output the final device state.

[0119] Objective function (loss function): , where is the true final state label, is the predicted final state probability distribution, is the regularization term (such as L2 regularization), preventing overfitting. is the regularization coefficient. is the network weight parameter, and N3 represents the number of training samples in the output first state analysis result and the second state analysis result.

[0120] The network constraint conditions are as follows: Input constraint: The input is the concatenation result of the outputs of the previous two networks .

[0121] Fusion constraint: Use weighted average or fully connected layer to fuse dynamic and static features: , wherein, is a learnable parameter.

[0122] Output constraint: Output the final state .

[0123] It should be noted that the output of the first two networks is fused by the third state analysis network, which must include a fully connected layer and an attention mechanism.

[0124] The network objective function in the fourth decision network structure is to determine what strategy to implement according to the type of the final operating state, which can be transformed into the discounted cumulative reward that maximizes the expectation of the agent, specifically as follows: The goal of the agent is to maximize the expected discounted cumulative reward: , wherein, represents a trajectory (the trajectory is generated by the policy ). represents the discount factor (such as ), which controls the weight of the long-term reward. T represents the termination time of the trajectory (such as equipment failure or maintenance completion).

[0125] wherein, the immediate reward r at each time step t t is composed of the following items , wherein, represents the task completion reward, is the penalty cost, is the exploration reward, is the constraint penalty.

[0126] In the structure of the reinforcement learning in this application, the environment is the power equipment operation system, which provides states and rewards. The agent is a policy network based on PPO / SAC (PPO (Proximal Policy Optimization): suitable for discrete action spaces and has stable training; SAC (Soft Actor-Critic): suitable for continuous action spaces and automatically balances exploration and exploitation), which selects maintenance actions. The state is a vector that fuses dynamic and static features and includes historical dependencies. The action is a discrete or continuous maintenance strategy (such as shutdown, alarm). The reward is to quantify the action effect and drive the agent to learn the optimal strategy. Specifically, such as Figure 3As shown, the environment provides the current state to the agent and gives a reward according to the action result after the agent takes an action. The agent receives the state information from the environment, decides what action to take through its internal algorithm (such as PPO or SAC), and feeds this action back to the environment. The state is the data obtained from the environment, containing information about the operation of the power equipment, and this information is input into the agent to assist in decision-making. The action is an operation selected by the agent based on the current state (such as shutting down the machine, alarming, etc.), and this operation acts on the environment. The reward is the corresponding reward or punishment given by the environment according to the action taken by the agent, which is used to guide the agent to learn the optimal strategy.

[0127] The data flow is as follows: The environment provides the state to the agent (state flow), the agent decides and executes an action based on the received state (action flow), after the action is executed, the environment gives the agent a reward according to the effect of the action (reward flow), and the agent uses the reward to update its policy model to make better decisions in the future.

[0128] Specifically, in this application is used to reward the agent for completing critical tasks (such as avoiding failures).

[0129] .

[0130] In this application is used to punish unnecessary maintenance actions (such as frequent shutdowns resulting in increased costs).

[0131] .

[0132] In this application is used to encourage the agent to explore new actions (such as trying unexecuted strategies).

[0133] , where represents the exploration coefficient (such as 0.1), and N a represents the historical execution times of action a.

[0134] In this application is used to ensure that the action conforms to physical constraints (such as the load cannot exceed the rated power).

[0135] , where is the constraint penalty coefficient (taken as 50 in this application).

[0136] It should be noted that the reward function in this application can be further updated. By considering the sparsity effect, key events (faults, maintenance) are given high rewards to avoid the learning difficulties caused by sparse rewards. It is also possible to increase the weights of exploration rewards and dense rewards in the early stage (exploration stage) to encourage the agent to try various actions, and strengthen the task rewards in the later stage (convergence stage) while sparsifying non-critical rewards.

[0137] It should also be noted that the fourth decision network structure in this application maximizes the discounted cumulative reward , and the agent will learn to adopt optimal maintenance strategies (such as shutdown, alarm) under device states (such as overload, precursor to failure), while avoiding the cost losses caused by over-maintenance.

[0138] It should be noted that establishing the first analysis model can make full use of the advantages of deep learning technology to comprehensively and accurately analyze the operating state of power equipment. By selecting appropriate deep learning network architectures, such as RNN and its variants, convolutional neural networks, and improved designs combined with reinforcement learning, not only the accuracy and efficiency of the analysis are improved, but also the generalization ability and robustness of the model are enhanced. In addition, by pre-embedding decision logic associated with subsequent processing operations when designing the model, the model can directly guide practical applications, avoiding the problem of limited application effects of the model. This innovative power equipment operating state analysis system will provide strong guarantees for the safe and stable operation of the power system.

[0139] S104. Train the first analysis model according to the second parameter to obtain the first analysis model after training is completed, denoted as the second analysis model; It should be noted that training the already established first analysis model according to the second parameter can further optimize the performance of the model to make it more suitable for the actual analysis requirements of the operating state of power equipment. The second parameter is divided into a training set, a validation set, and a test set. The training set is used for model training, the validation set is used to adjust hyperparameters and prevent overfitting, and the test set is used to evaluate the performance of the final model.

[0140] In the embodiment of this application, according to the established objective function and constraint conditions above, the first analysis model is globally trained in combination with the second parameter. After the model training is completed (denoted as the second analysis model), it is evaluated using an independent test set to check the generalization ability of the model on unseen data. Once the second analysis model passes the test, it can be deployed into practical applications to be used for real-time analysis of the operating state of the target power equipment. This process not only improves the accuracy and robustness of the model, but also ensures that it can adapt to different power equipment and operating environments, thus providing strong guarantees for the safe and stable operation of the power system.

[0141] S105. In response to the real-time data of the target power device being input into the second analysis model, an analysis result of the operating state of the power device is obtained.

[0142] It should be noted that data needs to be collected in real time from the target power device. These data include, but are not limited to, dynamic parameters such as voltage, current, power, temperature, and static parameters such as rated power and rated voltage. And the corresponding data is subjected to the first classification and the first preprocessing, and the data after the first classification and the first preprocessing is input into the second analysis model. The second analysis model will comprehensively analyze the operating state of the target power device based on these data through its internal first feature judgment network architecture, second feature judgment network architecture, third state analysis network structure, and fourth decision network structure. Specifically, the first feature judgment network architecture will process the dynamic parameters and extract the dynamic features of the device operating state; the second feature judgment network architecture will process the static parameters and extract the static features of the device; the third state analysis network structure will fuse the analysis results of the dynamic and static features and output the final device state; and the fourth decision network structure will determine what strategy to implement according to the type of the final operating state. This process will be carried out in real time to ensure that it can quickly and accurately reflect the actual operating state of the target power device and provide strong support for the safe and stable operation of the power system.

[0143] In summary, the present invention proposes a method for analyzing the operating state of a power device, which includes obtaining the first parameters of the target power device and performing the first classification on the first parameters; performing the first preprocessing on the first parameters after the first classification to obtain the second parameters, where the second parameters include second dynamic parameters and second static parameters; establishing a first analysis model, where the first analysis model includes two input ends, a four-layer network architecture, and an output end; training the first analysis model according to the second parameters to obtain the first analysis model after training, denoted as the second analysis model; and in response to the real-time data of the target power device being input into the second analysis model, obtaining an analysis result of the operating state of the power device. This method fully considers the complexity and diversity of the operating state of the power device, and through fine parameter classification and preprocessing, as well as an advanced four-layer network architecture analysis model, realizes the accurate analysis and real-time monitoring of the operating state of the power device. Compared with the traditional power device monitoring methods, the method of the present invention not only improves the analysis efficiency, but also can timely discover potential problems, avoiding economic losses and social impacts caused by equipment failures or power outages. In addition, this method can also dynamically adjust the analysis model according to the historical data and real-time data of the power device, improving the accuracy and reliability of the analysis. Therefore, the present invention has broad application prospects and important practical significance in the technical field of analyzing the operating state of power devices.

[0144] Embodiment 2. In this embodiment, a power equipment operation state analysis system is further provided, including: A data acquisition and classification module, configured to acquire first parameters of a target power equipment and perform a first classification on the first parameters; The first classification is used to classify the first parameters into first dynamic parameters and first static parameters; A preprocessing module, configured to perform a first preprocessing on the first parameters after the first classification to obtain second parameters, where the second parameters include second dynamic parameters and second static parameters; A model establishment module, configured to establish a first analysis model, where the first analysis model includes two input ends, a four-layer network architecture, and an output end; The four-layer network architecture includes a first feature judgment network architecture and a second feature judgment network architecture connected in parallel, a third state analysis network structure connected in series with the first feature judgment network architecture and the second feature judgment network architecture, and a fourth decision network structure connected in series with the third state analysis network structure; A model training module, configured to train the first analysis model according to the second parameters to obtain a first analysis model after training, denoted as a second analysis model; An analysis module, configured to obtain a power equipment operation state analysis result in response to real-time data of the target power equipment being input into the second analysis model.

[0145] The above-mentioned unit modules can be embedded in the processor of the computer device in a hardware form or be independent of it, or can be stored in the memory of the computer device in a software form, so as to facilitate the processor to call and execute the operations corresponding to the above-mentioned various modules.

[0146] This embodiment further provides a computer device, which may be a terminal, and its internal structure diagram may be as Figure 5As shown in the figure. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a method for analyzing the operating state of a power device. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the casing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0147] This embodiment also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented: Obtain the first parameter of the target power device and perform a first classification on the first parameter; The first classification is used to divide the first parameter into a first dynamic parameter and a first static parameter; Perform a first preprocessing on the first parameter after the first classification to obtain a second parameter, and the second parameter includes a second dynamic parameter and a second static parameter; Establish a first analysis model, and the first analysis model includes two input ends, a four-layer network architecture, and an output end; The four-layer network architecture includes a first feature judgment network architecture and a second feature judgment network architecture connected in parallel, a third state analysis network structure connected in series with the first feature judgment network architecture and the second feature judgment network architecture, and a fourth decision network structure connected in series with the third state analysis network structure; Train the first analysis model according to the second parameter to obtain the first analysis model after training, denoted as the second analysis model; In response to the real-time data of the target power device being input into the second analysis model, obtain the analysis result of the operating state of the power device.

[0148] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

[0149] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The solutions in the embodiments of the present application can be implemented in various computer languages.

[0150] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the specified functions in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0151] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device that implements the specified functions in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0152] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide steps for implementing the specified functions in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0153] Although the preferred embodiments of the present application have been described, those skilled in the art can make additional changes and modifications to these embodiments once they learn the basic creative concepts. Therefore, the appended claims are intended to be construed as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0154] Obviously, those skilled in the art can make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalent technologies, this application is also intended to include these modifications and variations.

Claims

1. A method for analyzing the operating status of electric power equipment, characterized in that: include: Acquire a first parameter of a target electric device, and perform a first classification on the first parameter; The first classification is used to divide the first parameter into a first dynamic parameter and a first static parameter; Performing a first preprocessing on the first parameter after the first classification to obtain a second parameter, where the second parameter includes a second dynamic parameter and a second static parameter; Establishing a first analysis model, wherein the first analysis model includes two input terminals, a four-layer network architecture, and an output terminal; The four-layer network architecture includes a first feature judgment network architecture and a second feature judgment network architecture connected in parallel, a third state analysis network structure connected in series with the first feature judgment network architecture and the second feature judgment network architecture, and a fourth decision network structure connected in series with the third state analysis network structure; Training the first analysis model according to the second parameter to obtain a trained first analysis model, which is recorded as a second analysis model; In response to the real-time data of the target electric power equipment being input into the second analysis model, an analysis result of the operating status of the electric power equipment is obtained.

2. A method for analyzing the operating status of electric power equipment according to claim 1, characterized in that: The first classification of the first parameter comprises: The first parameter includes a first internal parameter and a second external parameter; The first category includes: Preset the first parameter classification library and the second parameter classification judgment logic; Acquire classified parameters and unclassified parameters in the first internal parameter and the second external parameter according to the first parameter classification library; Performing classification judgment on the unclassified parameters according to the second parameter classification judgment logic to obtain a classification judgment result; A first dynamic parameter and a first static parameter are obtained according to the classified parameters and the classification judgment result.

3. A method for analyzing the operating status of electric power equipment according to claim 2, characterized in that: The first category also includes: The initial first parameter classification library is a blank library; The initial first classification is performed only through the second parameter classification judgment logic, and the initial first classification is performed based on the second parameter classification judgment logic. According to the parameters in the initial classification judgment result, a non-blank first parameter classification library is established.

4. A method for analyzing the operating status of electric power equipment according to claim 3, characterized in that: The first analysis model also includes: The two input ends are input ends of the first feature judgment network architecture and the second feature judgment network architecture respectively; One of the first feature judgment network architecture and the second feature judgment network architecture is used to process the second dynamic data in the second parameter, and the other is used to process the second static data in the second parameter; The output of the first feature judgment network architecture is a first state analysis result; The output of the second feature judgment network architecture is a second state analysis result; The first state analysis result and the second state analysis result are inputs of the third state analysis network structure.

5. A method for analyzing the operating status of electric power equipment according to claim 4, characterized in that: The first analysis model also includes: The third state analysis network structure is used to perform a third state analysis on the first state analysis result and the second state analysis result; Preset a third state analysis logic, and the third state analysis network structure is established according to the third state analysis logic; The result of the third state analysis is the final analysis state of the target power equipment; The final analysis state is input as input to the fourth decision network structure.

6. A method for analyzing the operating status of electric power equipment according to claim 5, characterized in that: The first analysis model also includes: The fourth decision network structure is used to decide the state processing strategy according to the final analysis state; The state processing strategy includes several processing strategies for different target power equipment operating states.

7. A method for analyzing the operating status of electric power equipment according to claim 6, characterized in that: The four-layer network architecture includes different network objective functions and network constraints.

8. A power equipment operating status analysis system, using the method according to any one of claims 1 to 7, characterized in that: include: A data acquisition and classification module, used to acquire a first parameter of a target power device and perform a first classification on the first parameter; The first classification is used to divide the first parameter into a first dynamic parameter and a first static parameter; A preprocessing module, used for performing a first preprocessing on the first parameter after the first classification to obtain a second parameter, where the second parameter includes a second dynamic parameter and a second static parameter; A model building module, used to build a first analysis model, wherein the first analysis model includes two input terminals, a four-layer network architecture and an output terminal; The four-layer network architecture includes a first feature judgment network architecture and a second feature judgment network architecture connected in parallel, a third state analysis network structure connected in series with the first feature judgment network architecture and the second feature judgment network architecture, and a fourth decision network structure connected in series with the third state analysis network structure; A model training module, used for training the first analysis model according to the second parameter to obtain a trained first analysis model, which is recorded as a second analysis model; The analysis module is used to obtain the analysis result of the operation status of the power equipment in response to the real-time data of the target power equipment input into the second analysis model.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of a method for analyzing the operating status of an electric power device as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a method for analyzing the operating status of an electric power device according to any one of claims 1 to 7 are implemented.

Images