Abnormity monitoring method, device and equipment of transformer and storage medium

Abstract

The invention relates to the technical field of transformer abnormity monitoring, in particular to a transformer abnormity monitoring method, device and equipment and a storage medium. The method comprises the following steps: acquiring multi-source state parameters of a transformer; carrying out time sequence load change analysis on the multi-source state parameters of the transformer, and carrying out deep correlation mining between features to generate a load-temperature change atlas; performing three-dimensional imaging scanning on the transformer, and marking a transformer assembly; performing center point space positioning calculation on the transformer assembly to obtain a plurality of assembly accurate position coordinates; three-dimensional form point cloud modeling is carried out on the load-temperature change atlas based on the precise position coordinates of the multiple assemblies, dynamic state attribute rendering is carried out, and therefore an operation state twin model is constructed; and performing multi-period state trend prediction on the operation state twin model, performing sequential serialization processing, and constructing a state trend prediction curve. According to the invention, efficient and accurate transformer abnormity monitoring is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of transformer abnormality monitoring, and in particular to a transformer abnormality monitoring method, device, equipment and storage medium. Background Art

[0002] As a key equipment in the power system, the stable and reliable operation of the transformer is of vital importance. However, in the long-term operation process, the transformer will be affected by various internal and external factors, such as load fluctuations, changes in environmental conditions, and component aging, which will have a certain degree of impact on the operating status of the transformer.

[0003] Traditional transformer condition monitoring methods mainly rely on regular manual inspections and fault diagnosis based on a single monitoring indicator. This method has problems such as limited monitoring range, low analysis accuracy, and low efficiency. With the increasing demand for intelligent management of power equipment, there is an urgent need for a new monitoring technology that can fully perceive the operating status of the transformer and intelligently analyze its abnormal changes. Based on this, an intelligent transformer abnormality monitoring method is needed. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a transformer abnormality monitoring method, device, equipment and storage medium to solve at least one of the above technical problems.

[0005] To achieve the above object, the present invention provides a transformer abnormality monitoring method, comprising the following steps: Step S1: Obtain multi-source state parameters of the transformer; perform time series load change analysis on the multi-source state parameters of the transformer, and perform deep correlation mining between features to generate a load-temperature change map; Step S2: perform three-dimensional imaging scanning on the transformer and mark the transformer components; perform spatial positioning calculation of the center points of the transformer components to obtain precise position coordinates of multiple components; Step S3: Based on the precise position coordinates of multiple components, a three-dimensional point cloud model is constructed for the load-temperature change map, and dynamic state attribute rendering is performed to construct an operation state twin model; Step S4: Predict the state trend of the operating state twin model in multiple time periods, perform sequential serialization processing, and construct a state trend prediction curve; Step S5: extracting curve outlier mutation points based on the state trend prediction curve; locating abnormal trends of the curve outlier mutation points, and marking abnormal state trend nodes; Step S6: Analyze the parameters of the components adjacent to the abnormal point for the abnormal state trend node, make an abnormal diagnosis decision, and build a transformer abnormal state diagnosis strategy.

[0006] The present invention also provides a transformer abnormality monitoring device, comprising: The deep correlation mining module is used to obtain the multi-source state parameters of the transformer; analyze the time series load changes of the multi-source state parameters of the transformer, and perform deep correlation mining between features to generate a load-temperature change map; The spatial positioning module is used to perform three-dimensional imaging scanning of the transformer and mark the transformer components; perform spatial positioning calculations on the center points of the transformer components to obtain the precise position coordinates of multiple components; The three-dimensional modeling module is used to perform three-dimensional point cloud modeling of the load-temperature change map based on the precise position coordinates of multiple components, and render dynamic state attributes to build a twin model of the operating state; The trend prediction module is used to predict the state trend of the operating state twin model in multiple periods, perform sequential serialization processing, and construct a state trend prediction curve; The abnormal trend positioning module is used to extract the curve outlier mutation points based on the state trend prediction curve; locate the abnormal trend of the curve outlier mutation points and mark the abnormal state trend nodes; The abnormal diagnosis decision module is used to analyze the parameters of the components adjacent to the abnormal state trend nodes, make abnormal diagnosis decisions, and build a transformer abnormal state diagnosis strategy.

[0007] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above-mentioned transformer abnormality monitoring methods when executing the computer program.

[0008] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any of the above-mentioned transformer abnormality monitoring methods are implemented.

[0009] The beneficial effects of the present invention are specifically as follows: by analyzing the time-series load changes of multi-source state parameters, the operation of the transformer under different loads can be understood, basic data can be provided for anomaly detection, a load-temperature change map can be generated through deep correlation mining between features, the potential relationship between state parameters can be revealed, and the operating status of the transformer can be monitored. The position information of the transformer components can be accurately obtained through three-dimensional imaging scanning and spatial positioning calculation, and accurate basic data can be provided for subsequent modeling and analysis. The component position coordinates are marked to establish an accurate three-dimensional morphological point cloud model of the transformer, providing visualization support for anomaly monitoring. The three-dimensional morphological point cloud modeling based on precise position coordinates can more realistically reflect the operating status of the transformer and improve the accuracy of anomaly detection. Dynamic state attribute rendering can monitor the operating status of the transformer in real time. , timely discover abnormal situations, predict the state trend of the operating status twin model to help predict the future operating status of the transformer, take measures in advance to avoid problems, build a state trend prediction curve to intuitively display the changing trend of the transformer state, and provide a reference for abnormal monitoring. Extract the curve outlier mutation points and locate the abnormal trend nodes to quickly identify abnormal situations in transformer operation and deal with problems in a timely manner. Mark the abnormal state trend nodes to help determine the time and location of the abnormality, and provide a basis for further abnormal diagnosis. Analyze the parameters of the components adjacent to the abnormal point of the abnormal state trend nodes to help find the cause of the abnormality, and propose corresponding diagnostic decisions. Construct a transformer abnormal state diagnosis strategy to systematically handle abnormal situations, improve operating efficiency and reduce losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A schematic diagram of the steps of a transformer abnormality monitoring method of the present invention; Figure 2 Detailed implementation flow chart of step S1; Figure 3 Detailed implementation flow chart of step S2; Figure 4 Detailed implementation flow chart of step S3. DETAILED DESCRIPTION

[0011] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0012] The present application example provides a transformer abnormality monitoring method, device, equipment and storage medium. The execution subjects of the transformer abnormality monitoring method, device, equipment and storage medium include but are not limited to the following: mechanical equipment, data processing platform, cloud server node, network upload device, etc. equipped with the system can be regarded as the general computing node of the present application, and the data processing platform includes but is not limited to: at least one of an audio image management system, an information management system, and a cloud data management system.

[0013] In the embodiment of the present invention, refer to Figure 1 , is a schematic flow chart of the steps of a transformer abnormality monitoring method of the present invention. In this example, the steps of the method include: Step S1: Obtain multi-source state parameters of the transformer; perform time series load change analysis on the multi-source state parameters of the transformer, and perform deep correlation mining between features to generate a load-temperature change map; In this embodiment, various sensor devices installed on the transformer are used to collect multi-dimensional state parameter data during the operation of the transformer. These parameters include load current, winding temperature, cooling system status, vibration noise, etc., covering multiple fields such as electrical, thermal, and mechanical. Time series analysis is performed on key parameters such as load current and winding temperature to extract their changing rules. Data mining methods are used to deeply analyze the correlation characteristics between load current and winding temperature. By establishing a mathematical model, an analysis map that can reflect the load-temperature change relationship of the transformer is formed, and trend analysis and periodic analysis are performed on the collected time series data such as load current and winding temperature. Etc., extract their changing rules, use time series analysis models, such as ARIMA model, Fourier analysis, etc., to mine the time evolution characteristics of these state parameters, use correlation analysis, regression analysis and other data mining methods, deeply explore the intrinsic correlation between load current and winding temperature, establish mathematical models that can reflect the relationship between the two, such as linear regression model, neural network model, etc., integrate the results of time series analysis and feature association analysis, and form an analysis map that can comprehensively describe the load-temperature change relationship of the transformer. This map can not only reflect the corresponding relationship between load and temperature, but also show their changing characteristics in the time dimension.

[0014] Step S2: perform three-dimensional imaging scanning on the transformer and mark the transformer components; perform spatial positioning calculation of the center points of the transformer components to obtain precise position coordinates of multiple components; In this embodiment, a three-dimensional scanning device, such as a laser scanner, a structured light scanner, etc., is used to perform high-precision three-dimensional modeling of the overall shape of the transformer, obtain point cloud data of the transformer surface, and construct a complete three-dimensional solid model. The various components of the transformer, such as windings, cores, insulators, etc., are manually or automatically labeled and segmented on the three-dimensional model to generate a three-dimensional transformer model marked with the position information of each component. The computer vision and geometric modeling methods are used to accurately calculate the three-dimensional coordinates of the center point of each component. The mathematical model is used to fit and geometrically analyze the surface point cloud data of the component to determine its precise spatial position. The three-dimensional point cloud data of the transformer obtained in the step is pre-processed by denoising and filtering to improve the data quality. For each segmented component point cloud data block, the geometric model fitting method is used to calculate its centroid coordinates. The principal component analysis, least squares fitting and other technologies are used to determine the three-dimensional spatial position of the center point of the component. The calculated coordinates of the center points of each component are integrated into a complete three-dimensional position data of the transformer component. The point cloud data is segmented and extracted at the component level using segmentation, clustering and other algorithms.

[0015] Step S3: Based on the precise position coordinates of multiple components, a three-dimensional point cloud model is constructed for the load-temperature change map, and dynamic state attribute rendering is performed to construct an operation state twin model; In this embodiment, 3D modeling software (such as Blender, 3ds Max or Unity) is used for point cloud modeling. The precise position coordinates of the components and the load-temperature variation map are imported into the modeling software. The corresponding 3D point cloud is generated according to the component position coordinates to ensure the correct position of each component in space. The generated point cloud is processed in detail, such as denoising and resampling, to improve the quality and accuracy of the model. A suitable rendering engine (such as Unity, Unreal Engine, etc.) is selected. Engine or Three.js) for dynamic attribute rendering, integrate the data of the load-temperature change map with the point cloud model, prepare for rendering, determine the dynamic attributes that need to be rendered, such as temperature change, load status and running time, map these attributes to the corresponding components of the point cloud model, use color, brightness or animation effects to represent status changes, dynamically update the display status of the point cloud model according to the load and temperature change data, simulate the performance of the transformer under different working conditions, set monitoring indicators, reflect the status changes of the transformer in real time, help with fault warning and performance evaluation, integrate the rendered dynamic point cloud model with the actual operation data of the transformer, build a complete operation status twin model, verify the twin model to ensure that it accurately reflects the actual operation status of the transformer, save and publish the constructed operation status twin model for subsequent monitoring and analysis.

[0016] Step S4: Predict the state trend of the operating state twin model in multiple time periods, perform sequential serialization processing, and construct a state trend prediction curve; In this embodiment, technical means such as time series prediction and dynamic simulation are used to predict the state change trend within a certain period of time in the future, and the development trend of multidimensional parameters such as load current, winding temperature, and cooling system status in the future time period is predicted. A data that can describe the future evolution trend of the overall state of the transformer is generated, which is integrated and arranged in chronological order to form a continuous state change sequence. These discrete state prediction points are connected into a smooth trend prediction curve using curve fitting and other methods to form a prediction curve that can intuitively reflect the future development trajectory of the overall state of the transformer. For key state parameters such as load current and winding temperature in the operating state twin model, time series analysis, machine learning and other methods are used to construct a prediction model for future state changes. These models are used to predict the future evolution trend of the state parameters. Based on the future prediction values ​​of each state parameter, they are integrated and arranged in chronological order to form a complete series of data on the change of the overall state of the transformer in the future time period. The discrete state prediction points are connected into a smooth trend prediction curve using curve fitting and other methods to generate a prediction curve that can intuitively reflect the future development trajectory of the overall state of the transformer.

[0017] Step S5: extracting curve outlier mutation points based on the state trend prediction curve; locating abnormal trends of the curve outlier mutation points, and marking abnormal state trend nodes; In this embodiment, methods such as outlier detection and mutation point analysis are used to identify outliers and mutation points on the curve. These outlier mutation points indicate that the transformer state has undergone abnormal changes at certain time points. Combined with the operating status twin model, the state change trend of these abnormal points is deeply analyzed to determine the specific time, duration, change amplitude and other characteristics of the abnormal state, and mark these abnormal state trend nodes on the transformer operating status twin model.

[0018] Step S6: Analyze the parameters of the components adjacent to the abnormal point for the abnormal state trend node, make an abnormal diagnosis decision, and build a transformer abnormal state diagnosis strategy.

[0019] In this embodiment, according to the structure of the transformer, components adjacent to the abnormal state trend node are identified, usually considering spatial position and functional similarity, collecting state parameters of adjacent components related to the abnormal state node, such as temperature, pressure, load, etc., extracting the state parameters of adjacent components from the real-time monitoring system or historical database, organizing the extracted parameters into a structured format to ensure the consistency and integrity of the data, selecting a suitable analysis method, such as correlation analysis, regression analysis or statistical test, evaluating the relationship between the abnormal state and the parameters of the adjacent components, analyzing the parameter changes of the adjacent components, identifying the parameter characteristics related to the abnormal state, formulating standards and strategies for abnormal diagnosis, including how to evaluate the abnormal state and its impact on the operation of the transformer, and determining key diagnostic indicators, such as The temperature rise, pressure anomaly, etc., select a suitable anomaly diagnosis model (such as decision tree, support vector machine or fuzzy logic system) according to the abnormal state characteristics and adjacent parameters, use historical data to train the diagnosis model to improve its prediction accuracy, apply the trained model to the newly identified abnormal state trend node, perform anomaly diagnosis, and evaluate the nature and potential causes of the abnormal state according to the results of the model output. According to the diagnosis results, formulate corresponding response measures, including maintenance, inspection or replacement of components, etc., establish an early warning mechanism for abnormal states to ensure timely response when an abnormality occurs, record the diagnosis results and strategies in the database for subsequent analysis and reference, establish a feedback mechanism, continuously monitor the effect after implementation, and adjust the diagnosis strategy according to actual conditions.

[0020] In this embodiment, refer to Figure 2 , is a flowchart of detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include: Step S11: monitor the transformer operating status in real time and obtain transformer multi-source status parameters; Step S12: performing load characteristic calculation on the transformer multi-source state parameters to obtain transformer load characteristic data; Step S13: performing a time series load change analysis on the transformer load characteristic data, thereby obtaining a time series load change characteristic; Step S14: performing temperature fluctuation identification on the transformer multi-source state parameters to obtain transformer temperature fluctuation data; Step S15: performing fluctuation distribution fitting on the transformer temperature fluctuation data to generate a temperature fluctuation distribution field; Step S16: Based on the time series load variation characteristics, deep correlation mining is performed on the temperature fluctuation distribution field to generate a load-temperature variation map.

[0021] In this embodiment, sensors are installed at key parts of the transformer (such as windings and oil tanks) to monitor state parameters such as current, voltage, temperature, and oil level. A data acquisition system is constructed to collect data from each sensor in real time to ensure the stability and real-time performance of data transmission. The collected multi-source state parameters are stored in a database for subsequent analysis. Load characteristic indicators such as rated load, actual load, and load factor are determined. Appropriate calculation methods (such as weighted average) are used to calculate load characteristics of multi-source state parameters. The calculated load characteristic data are organized into a structured format for subsequent analysis. The load characteristic data are organized into time series data according to timestamps. Appropriate time series analysis methods such as moving average and exponential smoothing are selected to analyze the time series data to identify trends and periodic characteristics of load changes. The time series load change characteristics are organized into a report for easy understanding and subsequent application. Temperature-related data are extracted from multi-source state parameters to ensure data integrity. Identification criteria for temperature fluctuations, such as fluctuations, are determined. The temperature fluctuation amplitude and frequency are identified by statistical methods (such as standard deviation calculation), and recorded as temperature fluctuation data. The temperature fluctuation data is organized into a structured format for subsequent analysis. An appropriate distribution fitting method is selected, such as normal distribution, Gamma distribution, etc. Statistical tools (such as Python's SciPy library) are used to fit the temperature fluctuation data to generate a probability distribution model for temperature fluctuation. A temperature fluctuation distribution field is generated based on the fitting results to represent the temperature fluctuation characteristics under different conditions. The results of the temperature fluctuation distribution field are organized into charts or data sets for subsequent analysis. An appropriate association analysis method is selected, such as correlation analysis, regression analysis, or machine learning model. The time series load change characteristics are integrated with the data of the temperature fluctuation distribution field to prepare for in-depth analysis. In-depth association mining is performed to identify the relationship between load changes and temperature fluctuations. A load-temperature change map is generated. The generated load-temperature change map is visualized to help understand the dynamic relationship between load and temperature.

[0022] In this embodiment, refer to Figure 3 , is a flowchart of detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include: Step S21: performing three-dimensional imaging scanning on the transformer to extract a three-dimensional image of the transformer; Step S22: performing multi-directional cutting processing on the three-dimensional image of the transformer, so as to obtain transformer images in multiple directions; Step S23: identifying components at different locations on the transformer images in multiple orientations, and marking the transformer components; Step S24: performing spatial positioning calculation of the center point of the transformer component to obtain the center position coordinates of multiple components; Step S25: performing three-dimensional position correction calculation on the center position coordinates of multiple components according to the transformer images in multiple orientations to obtain the precise position coordinates of the multiple components.

[0023] In this embodiment, a 3D imaging device (such as a laser scanner or a structured light scanner) is used for preparation to ensure that the device is calibrated correctly, a comprehensive scan of the transformer is performed, 3D point cloud data is collected, the external and internal structures of the transformer are recorded, and software (such as MeshLab or AutoCAD) is used to process the original point cloud data into a 3D image to generate a 3D model of the transformer, and the cutting orientation and cutting surface are determined to ensure that all key parts of the transformer are covered. The 3D image of the transformer is cut using 3D modeling software to generate transformer images in multiple orientations, and the multiple orientation images obtained by cutting are saved in different file formats (such as STL, OBJ) for subsequent processing, and the identification criteria of the components, such as shape, size, and material properties, are determined, and a suitable image is selected. Image processing algorithms (such as edge detection and morphological processing) are used to identify components. Images in multiple orientations are analyzed to identify and mark different parts of the transformer (such as windings, oil tanks, and insulators). The geometric center point calculation method of each component is determined, usually using the center of mass calculation. The coordinates of each marked component are calculated to obtain the position coordinates of its spatial center point. The calculated component center position coordinates are organized into structured data for subsequent correction. Appropriate correction methods are selected, such as rigid body transformation or similarity transformation, to ensure coordinate consistency. The center position coordinates of the component are corrected based on image data in multiple orientations to correct positioning errors caused by changes in viewing angles. The corrected component precise position coordinates are organized into a data set to record the final position of each component.

[0024] In this embodiment, refer to Figure 4 , is a flowchart of detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include: Step S31: performing state feature location inference on the load-temperature variation map based on the precise position coordinates of multiple components to obtain the matching position of each associated feature; Step S32: performing spatial layout analysis on the transformer components according to the precise position coordinates of the multiple components, thereby generating component spatial layout data; Step S33: mining the three-dimensional spatial topological relationship of the component spatial layout data to obtain a three-dimensional topological structure diagram of the component; Step S34: performing three-dimensional point cloud modeling on the three-dimensional topological structure diagram of the component to generate a three-dimensional point cloud model of the transformer; Step S35: Use the matching position of each associated feature to render the dynamic state attributes of the transformer three-dimensional point cloud model using the load-temperature change map, thereby constructing an operating state twin model.

[0025] In this embodiment, key features in the load-temperature variation map are identified, such as temperature peaks, load peaks, etc., and a suitable matching algorithm (such as nearest neighbor matching or interpolation) is selected to associate the map features with the component positions. For each identified feature, its matching position in the component space is inferred and recorded as a state feature positioning result. A suitable spatial analysis method is selected, such as cluster analysis or geometric shape analysis, and the precise position coordinates of multiple components are organized into a data set to prepare for spatial layout analysis. The spatial positions of the components are analyzed to identify the relative positional relationships between the components, and the analysis results are organized into component spatial layout data to describe the layout features of the components, determine the topological relationship standards between the components, such as adjacency and connection relationships, and select Select a suitable topology mining algorithm (such as graph theory algorithm) to analyze the relationship between components, generate a three-dimensional topology structure diagram based on the layout data, represent the connection and relative position between the components, save the generated three-dimensional topology structure diagram for subsequent visualization and analysis, use three-dimensional modeling software (such as CloudCompare or MeshLab) for point cloud modeling, convert the spatial layout of the components into point cloud data according to the three-dimensional topology structure diagram, generate a three-dimensional point cloud model of the transformer, optimize the point cloud model in detail, such as denoising, resampling, etc., to improve the model quality, save the generated three-dimensional point cloud model of the transformer in a usable format (such as PLY, OBJ), select a suitable rendering tool (such as Unity, Unreal Engine) for dynamic attribute rendering, integrate the state characteristics of the matching position with the three-dimensional point cloud model, prepare for rendering, and render the dynamic state attributes of the three-dimensional point cloud model of the transformer according to the characteristics of the load-temperature change map to display the operating status, and save the rendered model as a dynamic operating state twin model for real-time monitoring and analysis.

[0026] In this embodiment, step S4 includes the following steps: Step S41: performing state characteristic trend evolution on the operation state twin model to generate operation state trend characteristic data; Step S42: performing sliding window segmentation processing on the running status trend characteristic data, thereby obtaining a plurality of trend characteristic segmentation data; Step S43: performing deep time series evolution learning on multiple trend feature segment data to generate transformer trend feature evolution law; Step S44: using the transformer trend feature evolution law to perform multi-period state trend prediction on the operation state twin model to generate state trend prediction data for multiple time periods; Step S45: sequentially serialize the state trend prediction data of multiple time periods to construct a state trend prediction curve.

[0027] In this embodiment, according to the transformer operating state twin model, the dynamic change trend of key operating state characteristics such as load and temperature is analyzed, and the change law and development trend of these state characteristics in the time series are extracted. The transformer operating state twin model, the dynamic change trend of key operating state characteristics such as load and temperature is analyzed, and the change law and development trend of these state characteristics in the time series are extracted. A deep learning time series analysis model, such as a recurrent neural network, is used to perform deep time series learning on it, and the evolution law of the overall operating state of the transformer in the time dimension is excavated to form a mathematical model that can describe the development trend of the transformer state. Based on the evolution law obtained by learning, a state trend prediction model is constructed, and the constructed prediction model is used to perform multi-period state trend prediction on the operating state twin model, and state trend prediction data for multiple time periods is generated. A suitable serialization method (such as timestamp serialization) is selected to convert the state trend prediction data into a sequence, and a state trend prediction curve is drawn according to the serialized data. It is usually implemented using a chart tool (such as Matplotlib or Excel), and the constructed state trend prediction curve is visualized to ensure that the information is clear and easy to understand, and the trend prediction curve and related data are saved for subsequent reference and decision support.

[0028] In this embodiment, step S5 includes the following steps: Step S51: Identify trend outlier mutation points on the state trend prediction curve and extract the curve outlier mutation points; Step S52: Analyze the abnormal trend change of the outlier mutation point of the curve to obtain abnormal state trend data; Step S53: locate the abnormal trend of the operating status twin model according to the abnormal status trend data and mark the abnormal status trend node.

[0029] In this embodiment, relevant data points are extracted from the state trend prediction curve, an analyzable numerical array is constructed, and a suitable outlier detection algorithm is selected, such as the Z-score method, the IQR (interquartile range) method, or the LOF (local outlier factor) algorithm. The selected algorithm is applied to detect the curve data to identify outlier mutation points, that is, data points that significantly deviate from the normal trend. The identified outlier mutation points are organized into a data set, and their positions and values ​​are recorded for subsequent analysis. The characteristics of the abnormal trend, such as rapid rise, sharp drop, or increased fluctuation, are determined, and a suitable analysis method, such as time series analysis, trend analysis, or Statistical tests are performed to conduct detailed analysis of outlier mutation points, identify the causes and impacts behind them, record the changing characteristics of abnormal states, organize the abnormal state trend data obtained from the analysis into a structured format for subsequent use, determine the marking standards for abnormal trend nodes, such as abnormal amplitude, duration or impact range, and mark the abnormal trend nodes in the operating state twin model based on the abnormal state trend data to ensure that they are easy to identify. The marked abnormal nodes are visualized in the operating state twin model to improve visibility, and the marked abnormal state trend nodes and their characteristics are recorded in the database for subsequent monitoring and analysis.

[0030] In this embodiment, step S6 includes the following steps: Step S61: performing abnormal type identification on the abnormal state trend node to generate a node abnormal type; Step S62: Visualize the abnormal trend of the abnormal state trend node to generate an abnormal state trend visualization model; Step S63: Analyze the parameters of the components adjacent to the abnormal point on the abnormal state trend visualization model to obtain the parameters of the components adjacent to the abnormal point; Step S64: performing abnormal state correlation analysis on abnormal state trend nodes based on adjacent component parameters, thereby obtaining component abnormality correlation data; Step S65: Based on the node abnormality type and component abnormality associated data, an abnormality diagnosis decision is made on the abnormal state trend visualization model to construct a transformer abnormal state diagnosis strategy.

[0031] In this embodiment, the types of different abnormal states are determined, such as overload, temperature abnormality, vibration abnormality, etc., relevant feature data such as amplitude change, duration and frequency features are extracted from the abnormal state trend node, a suitable classification algorithm (such as decision tree, random forest or support vector machine) is selected to identify the abnormal type, the model is trained and applied to the abnormal state trend node, and the abnormal type label of each node is generated. A suitable visualization tool (such as Tableau, Matplotlib or D3.js) is selected for data visualization, and the data of the abnormal state trend node and its abnormal type are sorted to ensure that the format is suitable for visualization. An abnormal state trend visualization model is generated in the selected tool, and the abnormal node and its characteristic attributes are displayed to optimize the visualization effect to ensure that the information is clear and easy to understand, and to facilitate analysis and decision-making. The components adjacent to the abnormal node are determined, usually based on spatial position or functional similarity, and the status data of the adjacent components, such as temperature, load and vibration parameters, are collected from the monitoring system. A suitable analysis method (such as correlation analysis or regression analysis) is selected to analyze the parameters of the adjacent components. The parameters of the adjacent components are analyzed to identify the parameter characteristics related to the abnormal node, and a suitable association analysis method (such as Pearson correlation or Cohen's d) to evaluate the relationship between abnormal states, integrate the abnormal state trend nodes and their adjacent component parameters into an analysis data set, analyze the integrated data, identify the correlation between abnormal states and adjacent components, organize the analysis results into component abnormality association data, record the relationship between each component and the abnormal node, determine the standards and strategies for abnormal diagnosis, including how to make decisions based on the abnormal type and associated data, select appropriate decision support algorithms (such as fuzzy logic, expert system or machine learning model) for abnormal diagnosis, apply the selected algorithm, make abnormal diagnosis decisions based on the node abnormality type and component abnormality association data, generate a transformer abnormal state diagnosis strategy based on the diagnosis results, and record countermeasures and suggestions.

[0032] In this embodiment, the present invention further provides a transformer abnormality monitoring device, comprising: The deep correlation mining module is used to obtain the multi-source state parameters of the transformer; analyze the time series load changes of the multi-source state parameters of the transformer, and perform deep correlation mining between features to generate a load-temperature change map; The spatial positioning module is used to perform three-dimensional imaging scanning of the transformer and mark the transformer components; perform spatial positioning calculations on the center points of the transformer components to obtain the precise position coordinates of multiple components; The three-dimensional modeling module is used to perform three-dimensional point cloud modeling of the load-temperature change map based on the precise position coordinates of multiple components, and render dynamic state attributes to build a twin model of the operating state; The trend prediction module is used to predict the state trend of the operating state twin model in multiple periods, perform sequential serialization processing, and construct a state trend prediction curve; The abnormal trend positioning module is used to extract the curve outlier mutation points based on the state trend prediction curve; locate the abnormal trend of the curve outlier mutation points and mark the abnormal state trend nodes; The abnormal diagnosis decision module is used to analyze the parameters of the components adjacent to the abnormal state trend nodes, make abnormal diagnosis decisions, and build a transformer abnormal state diagnosis strategy.

[0033] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements any of the above-mentioned transformer abnormality monitoring methods when executing the computer program.

[0034] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the transformer abnormality monitoring method described in any one of the above items are implemented.

[0035] Those skilled in the art clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0036] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it is stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or partly or all or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media for storing program codes such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks or optical disks. Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0037] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A transformer abnormality monitoring method, characterized in that: The following steps are involved: Step S1: Obtain multi-source state parameters of the transformer; perform time series load change analysis on the multi-source state parameters of the transformer, and perform deep correlation mining between features to generate a load-temperature change map; Step S2: perform three-dimensional imaging scanning on the transformer and mark the transformer components; perform spatial positioning calculation of the center points of the transformer components to obtain precise position coordinates of multiple components; Step S3: Based on the precise position coordinates of multiple components, a three-dimensional point cloud model is constructed for the load-temperature change map, and dynamic state attribute rendering is performed to construct an operation state twin model; Step S4: Predict the state trend of the operating state twin model in multiple time periods, perform sequential serialization processing, and construct a state trend prediction curve; Step S5: extracting curve outlier mutation points based on the state trend prediction curve; locating abnormal trends of the curve outlier mutation points, and marking abnormal state trend nodes; Step S6: Analyze the parameters of the components adjacent to the abnormal point for the abnormal state trend node, make an abnormal diagnosis decision, and build a transformer abnormal state diagnosis strategy.

2. The transformer abnormality monitoring method according to claim 1, characterized in that: The specific steps of step S1 are: Step S11: monitor the transformer operating status in real time and obtain transformer multi-source status parameters; Step S12: performing load characteristic calculation on the transformer multi-source state parameters to obtain transformer load characteristic data; Step S13: performing a time series load change analysis on the transformer load characteristic data, thereby obtaining a time series load change characteristic; Step S14: performing temperature fluctuation identification on the transformer multi-source state parameters to obtain transformer temperature fluctuation data; Step S15: performing fluctuation distribution fitting on the transformer temperature fluctuation data to generate a temperature fluctuation distribution field; Step S16: Based on the time series load variation characteristics, deep correlation mining is performed on the temperature fluctuation distribution field to generate a load-temperature variation map.

3. The transformer abnormality monitoring method according to claim 1, characterized in that: The specific steps of step S2 are: Step S21: performing three-dimensional imaging scanning on the transformer to extract a three-dimensional image of the transformer; Step S22: performing multi-directional cutting processing on the three-dimensional image of the transformer, so as to obtain transformer images in multiple directions; Step S23: identifying components at different locations on the transformer images in multiple orientations, and marking the transformer components; Step S24: performing spatial positioning calculation of the center point of the transformer component to obtain the center position coordinates of multiple components; Step S25: performing three-dimensional position correction calculation on the center position coordinates of multiple components according to the transformer images in multiple orientations to obtain the precise position coordinates of the multiple components.

4. The transformer abnormality monitoring method according to claim 1, characterized in that: The specific steps of step S3 are: Step S31: performing state feature location inference on the load-temperature variation map based on the precise position coordinates of multiple components to obtain the matching position of each associated feature; Step S32: performing spatial layout analysis on the transformer components according to the precise position coordinates of the multiple components, thereby generating component spatial layout data; Step S33: mining the three-dimensional spatial topological relationship of the component spatial layout data to obtain a three-dimensional topological structure diagram of the component; Step S34: performing three-dimensional point cloud modeling on the three-dimensional topological structure diagram of the component to generate a three-dimensional point cloud model of the transformer; Step S35: Use the matching position of each associated feature to render the dynamic state attributes of the transformer three-dimensional point cloud model using the load-temperature change map, thereby constructing an operating state twin model.

5. The transformer abnormality monitoring method according to claim 1, characterized in that: The specific steps of step S4 are: Step S41: performing state characteristic trend evolution on the operation state twin model to generate operation state trend characteristic data; Step S42: performing sliding window segmentation processing on the running status trend characteristic data, thereby obtaining a plurality of trend characteristic segmentation data; Step S43: performing deep time series evolution learning on multiple trend feature segment data to generate transformer trend feature evolution law; Step S44: using the transformer trend feature evolution law to perform multi-period state trend prediction on the operation state twin model to generate state trend prediction data for multiple time periods; Step S45: sequentially serialize the state trend prediction data of multiple time periods to construct a state trend prediction curve.

6. The transformer abnormality monitoring method according to claim 1, characterized in that: The specific steps of step S5 are: Step S51: Identify trend outlier mutation points on the state trend prediction curve and extract the curve outlier mutation points; Step S52: Analyze the abnormal trend change of the outlier mutation point of the curve to obtain abnormal state trend data; Step S53: locate the abnormal trend of the operating status twin model according to the abnormal status trend data and mark the abnormal status trend node.

7. The transformer abnormality monitoring method according to claim 1, characterized in that: The specific steps of step S6 are: Step S61: performing abnormal type identification on the abnormal state trend node to generate a node abnormal type; Step S62: Visualize the abnormal trend of the abnormal state trend node to generate an abnormal state trend visualization model; Step S63: Analyze the parameters of the components adjacent to the abnormal point on the abnormal state trend visualization model to obtain the parameters of the components adjacent to the abnormal point; Step S64: performing abnormal state correlation analysis on abnormal state trend nodes based on adjacent component parameters, thereby obtaining component abnormality correlation data; Step S65: Based on the node abnormality type and component abnormality associated data, an abnormality diagnosis decision is made on the abnormal state trend visualization model to construct a transformer abnormal state diagnosis strategy.

8. A transformer abnormality monitoring device, characterized in that: The method for performing abnormality monitoring of a transformer according to claim 1 comprises: The deep correlation mining module is used to obtain the multi-source state parameters of the transformer; analyze the time series load changes of the multi-source state parameters of the transformer, and perform deep correlation mining between features to generate a load-temperature change map; The spatial positioning module is used to perform three-dimensional imaging scanning of the transformer and mark the transformer components; perform spatial positioning calculations on the center points of the transformer components to obtain the precise position coordinates of multiple components; The three-dimensional modeling module is used to perform three-dimensional point cloud modeling of the load-temperature change map based on the precise position coordinates of multiple components, and render dynamic state attributes to build a twin model of the operating state; The trend prediction module is used to predict the state trend of the operating state twin model in multiple periods, perform sequential serialization processing, and construct a state trend prediction curve; The abnormal trend positioning module is used to extract the curve outlier mutation points based on the state trend prediction curve; locate the abnormal trend of the curve outlier mutation points and mark the abnormal state trend nodes; The abnormal diagnosis decision module is used to analyze the parameters of the components adjacent to the abnormal state trend nodes, make abnormal diagnosis decisions, and build a transformer abnormal state diagnosis strategy.

9. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the transformer abnormality monitoring method according to 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 the transformer abnormality monitoring method according to any one of claims 1 to 7 are implemented.

Images