A method, system, device and storage medium for detecting internal abnormalities of a transformer

Abstract

The present invention relates to the technical field of transformer detection, and specifically to a method, system, device, and storage medium for detecting internal abnormalities of a transformer. To solve the technical problem of low accuracy in detecting transformer abnormalities in the prior art, the present invention first obtains ultrasonic data of a transformer in one week, and after filtering and alignment processing, obtains a number of aligned ultrasonic data of the transformer. Then, the aligned ultrasonic data of the transformer is converted into a point cloud and subjected to registration and splicing processing to obtain a three-dimensional point cloud model of the transformer. Finally, based on the three-dimensional point cloud model of the transformer, parameters reflecting the shape and position characteristics of the internal windings of the transformer are accurately and quickly obtained. After comparing with the corresponding standard parameters, the deformation degree of each winding can be quickly evaluated. And when the winding deformation degree is not large, the cumulative force on all internal windings of the transformer is calculated based on the diameter and radian of the winding, and the internal safety of the transformer is comprehensively analyzed to improve the accuracy and stability of detecting internal abnormalities of the transformer.

Description

Technical Field

[0001] The present invention relates to the technical field of transformer detection, and specifically provides a method, system, device, and storage medium for detecting internal abnormalities of a transformer. Background Art

[0002] When the internal windings of a transformer are deformed, electric field distortion will occur, leading to discharge phenomena and affecting the safety of the power grid. Therefore, the safe and accurate evaluation of transformers is the basis for the reliable operation of the power grid. Currently, existing technologies often use methods such as oil chromatography, partial discharge, infrared, and ultraviolet to detect abnormalities in transformers. However, the above methods all belong to the process detection after discharge or creepage occurs, and the detection timing is relatively lagged, lacking effective warning signals and time. At the same time, the above methods are easily affected by the operating environment, detection range, and sensitivity, resulting in poor accuracy and stability of the transformer abnormality detection results. Summary of the Invention

[0003] The purpose of the present invention is to provide a method, system, device, and storage medium for detecting internal abnormalities of a transformer.

[0004] The technical solution of the present invention is as follows:

[0005] A method for detecting internal abnormalities of a transformer includes the following operations:

[0006] S1. Arrange a number of ultrasonic sensors around the transformer to obtain ultrasonic data in real time, and obtain a number of transformer ultrasonic data. After filtering and aligning the number of transformer ultrasonic data respectively, a number of transformer ultrasonic aligned data are obtained;

[0007] S2. Convert the number of transformer ultrasonic aligned data into point cloud data respectively to obtain a number of transformer point cloud data; perform registration and splicing processing on the number of transformer point cloud data to obtain a three-dimensional point cloud model of the transformer;

[0008] S3. From the three-dimensional point cloud model of the transformer, obtain the diameter, height, radian, and inter-turn distance of the winding, calculate the differences from the standard diameter, standard height, standard radian, and standard inter-turn distance respectively, and obtain the diameter change amount, height change amount, radian change amount, and inter-turn distance change amount of the winding. If the diameter change amount, or height change amount, or radian change amount, or inter-turn distance change amount of the winding is greater than their respective change amount thresholds, there are abnormalities inside the transformer; if the diameter change amount, height change amount, radian change amount, and inter-turn distance change amount of the winding are not greater than their respective change amount thresholds, based on the diameter and height of the winding, obtain the internal safety value of the transformer; if the internal safety value of the transformer is less than the safety value threshold, there are abnormalities inside the transformer.

[0009] The operation of converting the current transformer ultrasonic alignment data into point cloud data in S2 is specifically as follows: Based on the position of each point in the current transformer ultrasonic alignment data, obtain the coordinates of each point in the current transformer ultrasonic alignment data to get the local coordinates of each point; Based on the distance between the ultrasonic sensor corresponding to the current transformer ultrasonic alignment data and the center of the transformer, convert the local coordinates of each point to the coordinate system with the center of the transformer as the origin to obtain the global coordinates of each point, thus forming the current transformer point cloud data.

[0010] The operation of registering and processing the current transformer point cloud data and the next set of transformer point cloud data in S2 is specifically as follows: In the next set of transformer point cloud data, obtain the corresponding point clouds of the characteristic point cloud of the current transformer point cloud data to get several pairs of characteristic point clouds; Based on several pairs of characteristic point clouds, construct several linear equations to form a system of linear equations; The system of linear equations is processed by the least squares method to obtain the rotation matrix and the translation vector; Based on the rotation matrix and the translation vector, perform a spatial transformation on the next set of transformer point cloud data to obtain the initial registration data of the next set of transformer point cloud data; The registered data of the next set of transformer point cloud and the current transformer point cloud data are processed by the iterative closest point method to obtain the registered data of the current transformer point cloud and the registered data of the next set of transformer point cloud.

[0011] The method for obtaining the characteristic point cloud of the current transformer point cloud data is specifically as follows: Obtain the neighborhood covariance matrix of each point cloud in the current transformer point cloud data; Use the point cloud corresponding to the eigenvalue ratio of the neighborhood covariance matrix greater than the eigenvalue ratio threshold as the characteristic point cloud.

[0012] The internal safety value of the transformer in S3 is obtained through the following calculation formula:

[0013] ,

[0014] ,

[0015] ,

[0016] is the internal safety value of the transformer, is the coefficient, is the i maximum allowable stress of the th winding, i is the mechanical stress of the th winding, i is the electromagnetic stress of the th winding, is the i elastic modulus of the th winding, i is the curvature of the is the diameter of the i th winding, is the radian of the i th winding, is the current density of the i th winding, is the magnetic field strength of the i th winding.

[0017] The alignment operation in S1 can be achieved by performing a spatial transformation on the ultrasonic filtering data of each transformer, converting the ultrasonic filtering data of each transformer from its respective local coordinate system to the global coordinate system.

[0018] S2 further includes downsampling a number of transformer point cloud data respectively to obtain a number of downsampled transformer point cloud data for performing subsequent registration operations; the downsampling operation is specifically: dividing the current transformer point cloud data into a number of point cloud units, retaining the centroid point cloud in each point cloud unit to obtain the current downsampled transformer point cloud data; the centroid point cloud is the average value of all point clouds in the corresponding point cloud unit.

[0019] A transformer internal anomaly detection system for implementing the above-mentioned transformer internal anomaly detection method, including:

[0020] A transformer ultrasonic alignment data generation module, which is used to arrange a number of ultrasonic sensors around the transformer to obtain ultrasonic data in real time to get a number of transformer ultrasonic data; after the a number of transformer ultrasonic data are respectively filtered, alignment processing is performed to obtain a number of transformer ultrasonic alignment data;

[0021] A transformer three-dimensional point cloud model generation module, which is used to convert a number of transformer ultrasonic alignment data into point cloud data respectively to obtain a number of transformer point cloud data; perform registration and splicing processing on the a number of transformer point cloud data to obtain a transformer three-dimensional point cloud model;

[0022] A transformer internal anomaly judgment module, which is used to obtain the diameter, height, radian and inter-turn distance of the winding from the transformer three-dimensional point cloud model, calculate the differences from the standard diameter, standard height, standard radian and standard inter-turn distance respectively to obtain the diameter change amount, height change amount, radian change amount and inter-turn distance change amount of the winding; if the diameter change amount, or height change amount, or radian change amount, or inter-turn distance change amount of the winding is respectively greater than their respective change amount thresholds, there is an anomaly inside the transformer; if the diameter change amount, height change amount, radian change amount and inter-turn distance change amount of the winding are all not greater than their respective change amount thresholds, based on the diameter and height of the winding, obtain the internal safety value of the transformer; if the internal safety value of the transformer is less than the safety value threshold, there is an anomaly inside the transformer.

[0023] An internal anomaly detection device for a transformer, comprising a processor and a memory. When the processor executes the computer program stored in the memory, the above-mentioned internal anomaly detection method for the transformer is implemented.

[0024] A computer-readable storage medium is used to store a computer program. When the computer program is executed by a processor, the above-mentioned internal anomaly detection method for the transformer is implemented.

[0025] The beneficial effects of the present invention are as follows:

[0026] A method for detecting internal anomalies of a transformer provided by the present invention. First, a number of ultrasonic sensors are arranged around the transformer to obtain ultrasonic data in real time, and then the ultrasonic data is filtered and aligned to improve the quality of the ultrasonic data and the calculation efficiency, and a number of aligned ultrasonic data of the transformer are obtained; then, the number of aligned ultrasonic data of the transformer are respectively converted into three-dimensional point cloud data and then registered and spliced to obtain a three-dimensional point cloud model of the transformer that can intuitively reflect the internal space structure characteristics of the transformer; finally, based on the three-dimensional point cloud model of the transformer, parameters reflecting the shape characteristics (diameter, height, curvature) and position characteristics (inter-turn distance) of the internal windings of the transformer are accurately and quickly obtained. After comparing with the corresponding parameters of the standard three-dimensional point cloud model of the transformer, the deformation degree of each winding can be quickly evaluated; and when the deformation degree of the winding is not large, the cumulative force on all the windings inside the transformer will be calculated according to the diameter and curvature of the winding, and the internal safety of the transformer will be comprehensively analyzed to improve the accuracy and stability of the internal anomaly detection of the transformer. Specific embodiments

[0027] This embodiment provides a method for detecting internal anomalies of a transformer, including the following operations:

[0028] S1. Arrange a number of ultrasonic sensors around the transformer to obtain ultrasonic data in real time to obtain a number of ultrasonic data of the transformer; after the number of ultrasonic data of the transformer are respectively filtered, alignment processing is performed to obtain a number of aligned ultrasonic data of the transformer;

[0029] S2. Respectively convert the number of aligned ultrasonic data of the transformer into point cloud data to obtain a number of point cloud data of the transformer; perform registration and splicing processing on the number of point cloud data of the transformer to obtain a three-dimensional point cloud model of the transformer;

[0030] S3. Obtain the diameter, height, radian, and inter-turn distance of the winding from the three-dimensional point cloud model of the transformer, calculate the differences from the standard diameter, standard height, standard radian, and standard inter-turn distance respectively, and obtain the diameter change amount, height change amount, radian change amount, and inter-turn distance change amount of the winding. If the diameter change amount, or height change amount, or radian change amount, or inter-turn distance change amount of the winding is greater than their respective change amount thresholds, there is an abnormality inside the transformer. If the diameter change amount, height change amount, radian change amount, and inter-turn distance change amount of the winding are all not greater than their respective change amount thresholds, based on the diameter and height of the winding, obtain the internal safety value of the transformer. If the internal safety value of the transformer is less than the safety value threshold, there is an abnormality inside the transformer.

[0031] S1. Arrange a number of ultrasonic sensors around the transformer to obtain ultrasonic data in real time, and obtain a number of transformer ultrasonic data. After filtering the number of transformer ultrasonic data respectively, perform alignment processing to obtain a number of transformer ultrasonic aligned data.

[0032] By arranging a number of ultrasonic sensors around the transformer and performing filtering and alignment processing on the ultrasonic data obtained in real time, the quality of the ultrasonic data is improved, the calculation efficiency is improved, and a number of transformer ultrasonic aligned data are obtained.

[0033] First, arrange a number of ultrasonic sensors with strong penetration ability and not easily affected by the external environment around the transformer to ensure that the key parts inside the transformer, such as the winding and other areas, can be comprehensively covered, and at the same time, obtain the ultrasonic data of the internal structure of the transformer in real time, and obtain a number of transformer ultrasonic data for real-time monitoring of the transformer.

[0034] Then, after filtering the number of transformer ultrasonic data respectively (preferably wavelet filtering), while effectively processing the noise in the non-stationary signal, the detailed information of the ultrasonic data can also be well retained, and a number of transformer ultrasonic filtered data are obtained.

[0035] Finally, perform alignment processing on the number of transformer ultrasonic filtered data to obtain a number of transformer ultrasonic aligned data. The operation of the alignment processing can be realized by performing a spatial transformation on each transformer ultrasonic filtered data and converting each transformer ultrasonic filtered data from its respective local coordinate system to a unified global coordinate system.

[0036] S2. Convert the number of transformer ultrasonic aligned data into point cloud data respectively to obtain a number of transformer point cloud data. Perform registration and stitching processing on the number of transformer point cloud data to obtain a three-dimensional point cloud model of the transformer.

[0037] After converting several pieces of transformer ultrasonic alignment data into three-dimensional point cloud data respectively, registration and stitching processing are performed to obtain a three-dimensional point cloud model of the transformer that can intuitively reflect the internal space structure characteristics of the transformer, which is beneficial to quickly and accurately perform subsequent abnormal detection inside the transformer.

[0038] First, convert several pieces of transformer ultrasonic alignment data into point cloud data respectively to obtain several pieces of transformer point cloud data.

[0039] Taking the current transformer ultrasonic alignment data as an example, the operation of converting the current transformer ultrasonic alignment data into point cloud data specifically is as follows: Based on the position of each point in the current transformer ultrasonic alignment data, obtain the coordinates of each point in the current transformer ultrasonic alignment data to get the local coordinates of each point; Based on the distance between the ultrasonic sensor corresponding to the current transformer ultrasonic alignment data and the center of the transformer, convert the local coordinates of each point to the coordinate system with the center of the transformer as the origin to obtain the global coordinates of each point, forming the current transformer point cloud data; Store the current transformer point cloud data in a three-dimensional point cloud file format (such as saving in PLY, XYZ, etc. formats).

[0040] Perform the above operations on the remaining transformer ultrasonic alignment data respectively to obtain the remaining transformer point cloud data, which can intuitively reflect the three-dimensional structure inside the transformer and is beneficial to improving the subsequent detection accuracy.

[0041] In addition, to improve the calculation efficiency, downsample several pieces of transformer point cloud data respectively to obtain several pieces of downsampled transformer point cloud data for performing subsequent registration operations. The operation of downsampling specifically is: Divide the current transformer point cloud data into several point cloud units, and retain the centroid point cloud in each point cloud unit to obtain the current transformer point cloud downsampled data. The centroid point cloud is the average value of all point clouds in the corresponding point cloud unit.

[0042] Then, perform registration processing on several pieces of transformer point cloud data or several pieces of downsampled transformer point cloud data to obtain several pieces of registered transformer point cloud data.

[0043] Taking the current transformer point cloud data (or the current downsampled transformer point cloud data) and the next set of transformer point cloud data (or the next set of downsampled transformer point cloud data) as an example, the operation of registration processing is specifically as follows. The positions of the two ultrasonic sensors corresponding to the current transformer point cloud data and the next set of transformer point cloud data are adjacent.

[0044] Step 1: In the next set of transformer point cloud data (or the next set of downsampled transformer point cloud data), obtain the corresponding point clouds of the feature point clouds of the current transformer point cloud data (or the current downsampled transformer point cloud data) to obtain several pairs of feature point clouds.

[0045] Among them, the method for obtaining the characteristic point cloud of the current transformer point cloud data is specifically as follows: obtain the neighborhood covariance matrix of each point cloud in the current transformer point cloud data (obtained based on the coordinate differences between each point cloud and the point clouds within the corresponding neighborhood range); use the point cloud corresponding to the eigenvalue ratio of the neighborhood covariance matrix being greater than the eigenvalue ratio threshold as the characteristic point cloud.

[0046] In addition, in a pair of characteristic point clouds, the cosine similarity between the characteristic point cloud belonging to the next set of transformer point cloud data and the characteristic point cloud belonging to the current transformer point cloud data is the maximum value of the cosine similarities between all the point clouds in the next set of transformer point cloud data and the characteristic point cloud of the current transformer point cloud data respectively.

[0047] Step 2: Based on several pairs of characteristic point clouds, construct several linear equations to form a system of linear equations; the system of linear equations is processed by the least squares method to obtain a rotation matrix and a translation vector; based on the rotation matrix and the translation vector, perform a spatial transformation on the next set of transformer point cloud data (or the downsampled data of the next set of transformer point clouds) (which can be achieved by multiplying the next set of transformer point cloud data by the transformation matrix obtained based on the rotation matrix and the translation vector) to obtain the initial registration data of the next set of transformer point clouds.

[0048] Step 3: To further improve the registration accuracy, the next set of transformer point cloud registration data and the current transformer point cloud data (or the downsampled data of the current transformer point clouds) are processed by the Iterative Closest Point (ICP) method to obtain the current transformer point cloud registration data and the next set of transformer point cloud registration data.

[0049] After all the transformer point cloud data (or all the downsampled data of the transformer point clouds) have completed the operations in Step 1, Step 2, and Step 3, several transformer point cloud registration data are obtained.

[0050] Finally, for the overlapping regions between two adjacent transformer point cloud registration data among the several transformer point cloud registration data, perform point cloud weighted average fusion, and for the non-overlapping regions, perform point cloud merging to achieve point cloud stitching, and obtain a three-dimensional point cloud model of the transformer that can intuitively reflect the internal spatial structure characteristics of the transformer.

[0051] S3. Obtain the diameter, height, radian, and inter-turn distance of the winding from the three-dimensional point cloud model of the transformer. Calculate the differences from the standard diameter, standard height, standard radian, and standard inter-turn distance respectively to obtain the diameter change amount, height change amount, radian change amount, and inter-turn distance change amount of the winding. If the diameter change amount, or height change amount, or radian change amount, or inter-turn distance change amount of the winding is greater than their respective change amount thresholds, there is an abnormality inside the transformer. If the diameter change amount, height change amount, radian change amount, and inter-turn distance change amount of the winding are all not greater than their respective change amount thresholds, obtain the internal safety value of the transformer based on the diameter and radian of the winding. If the internal safety value of the transformer is less than the safety value threshold, there is an abnormality inside the transformer.

[0052] Based on the three-dimensional point cloud model of the transformer, accurately and quickly obtain the parameters reflecting the shape characteristics (diameter, height, radian) and position characteristics (inter-turn distance) of the internal winding of the transformer. After comparing with the corresponding parameters of the standard three-dimensional point cloud model of the transformer, the deformation degree of each winding can be quickly evaluated. And when the deformation degree of the winding is not large, the cumulative force of all windings inside the transformer will be calculated based on the diameter and radian of the winding, and the internal safety of the transformer will be comprehensively analyzed to improve the accuracy of detecting abnormalities inside the transformer.

[0053] First, obtain the diameter, height, radian, and inter-turn distance of the winding from the three-dimensional point cloud model of the transformer.

[0054] Then, calculate the differences between the diameter, height, radian, and inter-turn distance of the winding in the three-dimensional point cloud model of the transformer and the standard diameter, standard height, standard radian, and standard inter-turn distance in the standard three-dimensional point cloud model of the transformer respectively to obtain the diameter change amount, height change amount, radian change amount, and inter-turn distance change amount of the winding in the three-dimensional point cloud model of the transformer, which are used to analyze the changes of the winding in the horizontal, vertical, and both sides.

[0055] The winding diameter is the cross-sectional diameter on the region of interest of the winding (preferably the middle region of the winding) in the three-dimensional point cloud model of the transformer. The winding height is the vertical height of the winding in the three-dimensional point cloud model of the transformer. The winding radian can be obtained by analyzing the vertical curve of the winding in the three-dimensional point cloud model of the transformer. The inter-turn distance of the winding is the minimum distance between adjacent windings in the three-dimensional point cloud model of the transformer.

[0056] Finally, based on the diameter change amount, height change amount, radian change amount, inter-turn distance change amount, and change amount, judge whether the deformation of the winding will cause an abnormality inside the transformer.

[0057] If the diameter change amount, or height change amount, or radian change amount, or inter-turn distance change amount of the winding is greater than their respective change amount thresholds, it proves that the deformation of the winding is serious and there is an abnormality inside the transformer.

[0058] If the variation amounts of the diameter, height, radian, and inter-turn distance of the winding are all not greater than their respective variation thresholds, for comprehensive analysis, based on the diameter and radian of the winding, the winding stress condition is evaluated to obtain the internal safety value of the transformer.

[0059] The internal safety value of the transformer is obtained through the following calculation formula:

[0060] ,

[0061] ,

[0062] ,

[0063] is the internal safety value of the transformer, is a coefficient, is the i maximum allowable stress of the is the i mechanical stress of the is the i electromagnetic stress of the is the total number of windings, is the i elastic modulus of the is the i curvature of the is the i diameter of the is the i radian of the is the i current density of the is the i magnetic field strength of the

[0064] If the internal safety value of the transformer is less than the safety value threshold, it proves that the greater the force on the internal winding of the transformer, the more likely it is to deform, and it is very easy to be dangerous inside the transformer, then there is an abnormality inside the transformer.

[0065] This embodiment also provides a transformer internal abnormality detection system for implementing the above-mentioned transformer internal abnormality detection method, including:

[0066] A transformer ultrasonic alignment data generation module, which is used to arrange several ultrasonic sensors around the transformer to obtain ultrasonic data in real time to obtain several transformer ultrasonic data; after the several transformer ultrasonic data are respectively filtered, alignment processing is performed to obtain several transformer ultrasonic alignment data;

[0067] The 3D point cloud model generation module of the transformer is used to convert several ultrasonic alignment data of the transformer into point cloud data respectively to obtain several transformer point cloud data; perform registration and splicing processing on the several transformer point cloud data to obtain the 3D point cloud model of the transformer;

[0068] The abnormal condition judgment module inside the transformer is used to obtain the diameter, height, radian and inter-turn distance of the winding from the 3D point cloud model of the transformer, calculate the differences from the standard diameter, standard height, standard radian and standard inter-turn distance respectively to obtain the diameter change amount, height change amount, radian change amount and inter-turn distance change amount of the winding; if the diameter change amount, or height change amount, or radian change amount, or inter-turn distance change amount of the winding is greater than their respective change amount thresholds, there is an abnormality inside the transformer; if the diameter change amount, height change amount, radian change amount and inter-turn distance change amount of the winding are not greater than their respective change amount thresholds, based on the diameter and height of the winding, obtain the internal safety value of the transformer; if the internal safety value of the transformer is less than the safety value threshold, there is an abnormality inside the transformer.

[0069] This embodiment also provides a transformer internal abnormality detection device, including a processor and a memory. Among them, when the processor executes the computer program stored in the memory, the above-mentioned transformer internal abnormality detection method is implemented.

[0070] This embodiment also provides a computer-readable storage medium for storing a computer program. Among them, when the computer program is executed by the processor, the above-mentioned transformer internal abnormality detection method is implemented.

[0071] A transformer internal abnormality detection method provided in this embodiment first arranges several ultrasonic sensors around the transformer to obtain ultrasonic data in real time and then perform filtering and alignment processing to improve the quality of ultrasonic data and the calculation efficiency to obtain several transformer ultrasonic alignment data; then, convert the several transformer ultrasonic alignment data into 3D point cloud data respectively and perform registration and splicing processing to obtain the 3D point cloud model of the transformer that can intuitively reflect the internal space structure characteristics of the transformer; finally, based on the 3D point cloud model of the transformer, accurately and quickly obtain the parameters reflecting the shape characteristics (diameter, height, radian) and position characteristics (inter-turn distance) of the internal winding of the transformer. After comparing with the corresponding parameters of the standard 3D point cloud model of the transformer, the deformation degree of each winding can be quickly evaluated; and when the winding deformation degree is not large, the cumulative force condition of all windings inside the transformer will be calculated according to the diameter and radian of the winding, and the internal safety of the transformer will be comprehensively analyzed to improve the accuracy and stability of the transformer internal abnormality detection.

Claims

1. A method for detecting internal anomalies of a transformer, characterized in that, Including the following operations: S1. Arrange a number of ultrasonic sensors around the transformer to obtain ultrasonic data in real time, and obtain a number of transformer ultrasonic data; after filtering the number of transformer ultrasonic data respectively, perform alignment processing to obtain a number of transformer ultrasonic alignment data; S2. Convert the number of transformer ultrasonic alignment data into point cloud data respectively to obtain a number of transformer point cloud data; perform registration and stitching processing on the number of transformer point cloud data to obtain a three-dimensional point cloud model of the transformer; S3. From the three-dimensional point cloud model of the transformer, obtain the diameter, height, radian and inter-turn distance of the winding, calculate the differences from the standard diameter, standard height, standard radian and standard inter-turn distance respectively, and obtain the diameter change amount, height change amount, radian change amount and inter-turn distance change amount of the winding; If the diameter change amount, or height change amount, or radian change amount, or inter-turn distance change amount of the winding is greater than their respective change amount thresholds, there is an abnormality inside the transformer; If the diameter change amount, height change amount, radian change amount and inter-turn distance change amount of the winding are not greater than their respective change amount thresholds, based on the diameter and height of the winding, obtain the internal safety value of the transformer; if the internal safety value of the transformer is less than the safety value threshold, there is an abnormality inside the transformer.

2. The internal anomaly detection method of a transformer according to claim 1, characterized in that, In the above S2, the operation of converting the current transformer ultrasonic alignment data into point cloud data is specifically as follows: Based on the position of each point in the current transformer ultrasonic alignment data, obtain the coordinates of each point in the current transformer ultrasonic alignment data to obtain the local coordinates of each point; based on the distance between the ultrasonic sensor corresponding to the current transformer ultrasonic alignment data and the center of the transformer, convert the local coordinates of each point to the coordinate system with the center of the transformer as the origin to obtain the global coordinates of each point, thus forming the current transformer point cloud data.

3. The method for detecting internal abnormalities of a transformer according to claim 1, characterized in that, In the above S2, the operation of performing registration processing on the current transformer point cloud data and the next set of transformer point cloud data is specifically as follows: In the next set of transformer point cloud data, obtain the corresponding point clouds of the feature point clouds of the current transformer point cloud data to obtain a number of feature point cloud pairs; Based on the number of feature point cloud pairs, construct a number of linear equations to form a system of linear equations; the system of linear equations is processed by the least squares method to obtain a rotation matrix and a translation vector; based on the rotation matrix and the translation vector, perform a spatial transformation on the next set of transformer point cloud data to obtain the initial registration data of the next set of transformer point cloud data; The next set of transformer point cloud registration data and the current transformer point cloud data are processed by the iterative closest point method to obtain the current transformer point cloud registration data and the next set of transformer point cloud registration data.

4. The internal abnormal detection method of a transformer according to claim 3, wherein, The method for obtaining the feature point cloud of the current transformer point cloud data is specifically as follows: obtain the neighborhood covariance matrix of each point cloud in the current transformer point cloud data; use the point cloud corresponding to the eigenvalue ratio of the neighborhood covariance matrix greater than the eigenvalue ratio threshold as the feature point cloud.

5. The method for detecting internal abnormalities of a transformer according to claim 1, wherein In the above S3, the internal safety value of the transformer is obtained through the following calculation formula: ， ， ， is the internal safety value of the transformer, is the coefficient, is the maximum allowable stress of the i th winding, is the mechanical stress of the i th winding, is the electromagnetic stress of the i th winding, is the total number of windings, is the elastic modulus of the i th winding, is the curvature of the i th winding, is the diameter of the i th winding, is the radian of the i th winding, is the current density of the i th winding, is the magnetic field strength of the i th winding.

6. The internal abnormal detection method of a transformer according to claim 1, characterized in that In the above S1, the alignment process can be achieved by performing a spatial transformation on the ultrasonic filtering data of each transformer, and converting the ultrasonic filtering data of each transformer from its respective local coordinate system to the global coordinate system.

7. The method for detecting internal anomalies of a transformer according to claim 1, wherein The S2 further includes downsampling a plurality of transformer point cloud data respectively to obtain a plurality of downsampled transformer point cloud data for performing subsequent registration operations. The specific operation of the downsampling process is as follows: dividing the current transformer point cloud data into a plurality of point cloud units, and retaining the centroid point cloud in each point cloud unit to obtain the downsampled current transformer point cloud data. The centroid point cloud is the average value of all the point clouds in the corresponding point cloud unit.

8. A transformer internal anomaly detection system for implementing the transformer internal anomaly detection method described in claim 1, characterized in that, It includes: A transformer ultrasonic alignment data generation module, which is used to arrange a plurality of ultrasonic sensors around the transformer to obtain ultrasonic data in real time, and obtain a plurality of transformer ultrasonic data; after the plurality of transformer ultrasonic data are respectively filtered, alignment processing is performed to obtain a plurality of transformer ultrasonic alignment data. A transformer three-dimensional point cloud model generation module, which is used to convert the plurality of transformer ultrasonic alignment data into point cloud data respectively to obtain a plurality of transformer point cloud data; perform registration and splicing processing on the plurality of transformer point cloud data to obtain a transformer three-dimensional point cloud model. A transformer internal anomaly judgment module, which is used to obtain the diameter, height, radian, and inter-turn distance of the winding from the transformer three-dimensional point cloud model, calculate the differences from the standard diameter, standard height, standard radian, and standard inter-turn distance respectively, and obtain the diameter change amount, height change amount, radian change amount, and inter-turn distance change amount of the winding; if the diameter change amount, or height change amount, or radian change amount, or inter-turn distance change amount of the winding is respectively greater than their respective change amount thresholds, there is an anomaly inside the transformer; if the diameter change amount, height change amount, radian change amount, and inter-turn distance change amount of the winding are all not greater than their respective change amount thresholds, based on the diameter and height of the winding, obtain the internal safety value of the transformer; if the internal safety value of the transformer is less than the safety value threshold, there is an anomaly inside the transformer.

9. An abnormal detection device inside a transformer, characterized in that, It includes a processor and a memory. Among them, when the processor executes the computer program stored in the memory, it implements the transformer internal anomaly detection method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that It is used to store a computer program. Among them, when the computer program is executed by the processor, it implements the transformer internal anomaly detection method according to any one of claims 1-7.

Images