Winding deformation diagnosis method under transformer short circuit fault

Abstract

The invention discloses a winding deformation diagnosis method under a transformer short circuit fault, and belongs to the field of transformer fault diagnosis. Comprising the following steps: developing and constructing a three-dimensional electromagnetic structure finite element model, simulating interaction between a magnetic field and an electromagnetic force under a short circuit condition based on the three-dimensional electromagnetic structure finite element model, and establishing a nonlinear mapping relation between leakage magnetic field distribution and mechanical deformation; capturing the dynamic leakage magnetic field change on the transformer oil tank through the magnetic sensor array; and associating the magnetic flux density with the deformation amplitude based on a least square magnetic field inversion algorithm, and diagnosing the winding deformation to obtain a diagnosis result. According to the method, by integrating multi-physics field modeling, an optimized sensor array and an advanced algorithm, rapid and accurate diagnosis of winding deformation under the transformer short-circuit fault is achieved, and a powerful guarantee is provided for reliable operation of an intelligent power grid.

Description

Technical Field

[0001] The present invention belongs to the field of transformer fault diagnosis, and in particular relates to a method for diagnosing winding deformation under a transformer short-circuit fault. Background Art

[0002] Power transformers are critical devices for power transmission and voltage conversion in power grids. Their short-circuit resistance is determined by the mechanical stability of their windings. Winding deformation during short-circuit faults impacts the integrity of the equipment and the reliability of the power grid. During an SC fault, transient currents generate strong electromagnetic forces, generating mechanical stresses that exceed the yield strength of the insulation material, leading to irreversible winding damage such as interturn insulation failure, conductor misalignment, and axial collapse. However, traditional diagnostic methods face significant limitations. Dissolved gas analysis (DGA) monitors characteristic gases (such as C₂H₂ and H₂) to detect discharge or overheating, but gas diffusion delays and multi-field interference hinder accurate deformation quantification and location. Frequency response analysis (FRA) identifies structural anomalies using frequency-domain transfer functions (10 Hz to 2 MHz), but its accuracy is affected by capacitance effects and ground interference, requiring offline testing and unsuitable for dynamic monitoring. Similarly, the short-circuit impedance method (SCIM) assesses deformation through changes in leakage inductance, but relies on manual offline procedures and cannot capture transient SC responses. Summary of the Invention

[0003] To solve the above problems, the present invention provides a method for diagnosing winding deformation under a transformer short-circuit fault, comprising:

[0004] Develop and construct a three-dimensional electromagnetic structure finite element model, simulate the interaction between the magnetic field and the electromagnetic force under short-circuit conditions based on the three-dimensional electromagnetic structure finite element model, and establish a nonlinear mapping relationship between the leakage magnetic field distribution and the mechanical deformation;

[0005] Capturing the dynamic leakage magnetic field changes on the transformer tank through a magnetic sensor array;

[0006] Based on the least squares magnetic field inversion algorithm, the magnetic flux density is associated with the deformation amplitude, the winding deformation is diagnosed, and the diagnosis result is obtained.

[0007] Preferably, the three-dimensional electromagnetic structure finite element model is developed based on a 110 kV transformer to simulate the electromagnetic-mechanical interaction of the winding under short-circuit fault conditions.

[0008] Preferably, the process of establishing a nonlinear mapping relationship between leakage magnetic field distribution and mechanical deformation includes:

[0009] The electromagnetic force and short-circuit impulse current under short-circuit fault are obtained. Based on the three-dimensional electromagnetic structure finite element model and combined with the finite element discretization principle, the spatiotemporal distribution law of the winding magnetic flux density under the action of transient short-circuit current is obtained, and the dynamic electrodynamic characteristics are iteratively calculated using the electromagnetic force formula.

[0010] Preferably, the electromagnetic force The formula expression is:

[0011] ;

[0012] Where L is the effective conductor length, I is the current density, and B is the magnetic induction intensity value;

[0013] According to the Biot-Savart law, the magnetic induction intensity value B is expressed in terms of current I as:

[0014] ;

[0015] Where μ is the magnetic permeability, dl is the differential current element vector of the source, and e r is the unit vector of the current element pointing to the field point, and r is the straight-line distance between the current element and the observation point.

[0016] Preferably, the short-circuit impulse current is expressed as follows:

[0017] ;

[0018] in, is the instantaneous current, is the peak voltage on the high side, is the equivalent short-circuit impedance from the power supply to the fault point, is the initial phase angle when the short circuit occurs, is the time constant, t is the time, is the angular frequency.

[0019] Preferably, the process of capturing the dynamic leakage magnetic field change on the transformer tank by using the magnetic sensor array includes:

[0020] The magnetic sensor array is aligned parallel to the axial direction of the winding, symmetrically placed between the iron core and the winding system, with a placement range approximately equal to the winding length and installed on the transformer tank to capture the dynamic leakage magnetic field changes on the transformer tank.

[0021] Preferably, the magnetic flux density is correlated with the deformation amplitude based on the least squares magnetic field inversion algorithm to diagnose the winding deformation. The process of obtaining the diagnosis result includes:

[0022] Based on the least squares magnetic field inversion algorithm, the corresponding deformation rates are calculated according to the axial and radial deformation characteristics of the winding, and the deformation of the winding is quantitatively diagnosed according to the deformation rates.

[0023] Compared with the prior art, the present invention has the following advantages and technical effects:

[0024] This paper proposes a non-invasive, real-time diagnostic technique based on leakage magnetic field (LMF) analysis to quantify winding deformation during SC faults. Unlike traditional methods that rely on gas concentration or frequency-domain parameters, this method uses a highly sensitive magnetic sensor array to continuously acquire multidimensional LMF data, enabling simultaneous assessment of deformation severity and location. This technique, combined with a least-squares (LS) algorithm, achieves millisecond-level fault warning, shortening response time to hours. It also reconstructs the mechanical stress distribution through magnetic gradient analysis. This method overcomes the temporal and spatial limitations of offline diagnosis, improving measurement safety and diagnostic efficiency, and provides critical support for real-time condition monitoring and proactive fault prevention in smart grid systems.

[0025] This paper utilizes a multiphysics finite element analysis (FEA) model to establish a quantitative relationship between LMF distribution and mechanical deformation. This model was validated on a 110 kV transformer, thereby advancing transformer diagnostics. A highly sensitive sensor array strategically placed on the transformer tank captures dynamic LMF variations, while a least squares (LS)-based inversion algorithm correlates magnetic flux density with deformation amplitude. This method, employing field-circuit coupled finite element analysis to simulate electromagnetic-mechanical interactions, provides a strong theoretical foundation for real-time threshold monitoring and ensures compatibility with transformer health assessment frameworks.

[0026] This invention characterizes the vibration conditions of the windings under SC transient conditions, revealing the dynamic mechanical response to electromagnetic forces. It also elucidates the LMF distribution patterns associated with axial and radial deformation, enabling precise fault location. A deformation diagnosis system based on LMF characteristic parameters is developed, achieving diagnostic accuracy below 1.3% and millisecond-level response time. These advances provide a novel and cost-effective solution for real-time condition monitoring, supporting predictive maintenance, and improving the reliability of smart grid systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The accompanying drawings, which constitute part of this application, are intended to provide a further understanding of this application. The exemplary embodiments and descriptions of this application are intended to explain this application and do not constitute an improper limitation on this application. In the accompanying drawings:

[0028] Figure 1 Schematic diagram of a method flow in an embodiment of the present invention.

[0029] Figure 2Schematic diagram of simulation model of different types of 110kV power transformers according to an embodiment of the present invention.

[0030] Figure 3 This is a simulation result diagram of the short-circuit current output of low-voltage windings of different types of transformers changing with time according to an embodiment of the present invention.

[0031] Figure 4 1 is a diagram showing the distribution trend of magnetic leakage in the winding and the corresponding stress components according to an embodiment of the present invention.

[0032] Figure 5 Schematic diagram of the transformer-side sensor placement area according to an embodiment of the present invention.

[0033] Figure 6 1 is an axial and radial decomposition diagram of the winding deformation model of an embodiment of the present invention.

[0034] Figure 7 This is a diagram of magnetic induction intensity values ​​at different detection points of the winding under different axial deformation degrees according to an embodiment of the present invention.

[0035] Figure 8 This is a diagram of magnetic induction intensity values ​​at different measuring points of the winding under different axial deformation degrees according to an embodiment of the present invention.

[0036] Figure 9 Schematic diagram of the winding deformation determination process according to an embodiment of the present invention. DETAILED DESCRIPTION

[0037] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0038] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0039] Example 1

[0040] like Figure 1 As shown, this embodiment provides a method for diagnosing winding deformation under a transformer short-circuit fault, including:

[0041] Develop and construct a three-dimensional electromagnetic structure finite element model, simulate the interaction between the magnetic field and electromagnetic force under short-circuit conditions based on the three-dimensional electromagnetic structure finite element model, and establish a nonlinear mapping relationship between the leakage magnetic field distribution and mechanical deformation;

[0042] Capturing the dynamic leakage magnetic field changes on the transformer tank through a magnetic sensor array;

[0043] Based on the least squares magnetic field inversion algorithm, the magnetic flux density is associated with the deformation amplitude, the winding deformation is diagnosed, and the diagnosis result is obtained.

[0044] Specifically, winding deformation during a power transformer short-circuit fault affects equipment integrity and grid reliability. Traditional diagnostic methods, such as dissolved gas analysis and frequency response analysis, rely on offline procedures and are unable to detect faults in a timely manner. This paper proposes a non-invasive, real-time diagnostic technique based on leakage magnetic field analysis to quantify axial and axial winding deformation. A three-dimensional electromagnetic structural finite element model of a 110 kV transformer was developed, and a nonlinear mapping relationship between leakage magnetic field (LMF) distribution and mechanical stress was established. Simulations verified that deformation increased 26-fold under fault conditions. A highly sensitive magnetic sensor array, strategically placed on the transformer tank, captures dynamic LMF changes, following an optimized layout principle to improve accuracy. Combined with a least-squares-based magnetic field inversion algorithm, the system achieves accurate deformation diagnosis with an error of less than 1.3%, enabling millisecond-level fault warning. Experiments validate the accuracy and feasibility of this method, providing an economical and secure solution for real-time condition monitoring in smart grids. This method overcomes the temporal and spatial limitations of traditional diagnostics, supports predictive maintenance, and paves the way for advanced fault warning systems.

[0045] Furthermore, a three-dimensional electromagnetic structural finite element model was developed based on a 110 kV transformer to simulate the electromagnetic-mechanical interaction of the winding under short-circuit fault conditions.

[0046] Furthermore, the process of establishing a nonlinear mapping relationship between leakage magnetic field distribution and mechanical deformation includes:

[0047] The electromagnetic force and short-circuit impulse current under short-circuit fault are obtained. Based on the three-dimensional electromagnetic structure finite element model and combined with the finite element discretization principle, the spatiotemporal distribution law of the winding magnetic flux density under the action of transient short-circuit current is obtained, and the dynamic electrodynamic characteristics are iteratively calculated using the electromagnetic force formula.

[0048] Specifically, under SC faults, power transformers experience strong electromagnetic-mechanical interactions, increasing the risk of winding deformation. Due to the limitations of the core material's magnetic permeability and geometric structure, part of the magnetic field escapes the core, forming a nonlinear LMF, which affects electromagnetic stability and exacerbates the fault condition. This embodiment establishes a quantitative mapping relationship between LMF distribution and mechanical deformation and proposes a detection method that converts LMF distortion into a measurable deformation index. This method links electromagnetic anomalies with mechanical conditions, providing a new approach for transformer fault diagnosis.

[0049] Furthermore, the electromagnetic force (F) generated by the interaction between the current density (I) and the LMF (B) leads to uneven stress distribution on the winding conductor, causing axial expansion and contraction in the radial direction; specifically, low-voltage axial elongation and radial inward contraction, and high-voltage axial contraction and radial outward expansion.

[0050] electromagnetic force The formula expression is:

[0051] (1);

[0052] Where L is the effective conductor length, I is the current density, and B is the magnetic induction intensity value;

[0053] According to the Biot-Savart law, the magnetic induction intensity value B is expressed in terms of current I as:

[0054] (2);

[0055] Where μ is the magnetic permeability, dl is the differential current element vector of the source, and e r is the unit vector of the current element pointing to the field point, and r is the straight-line distance between the current element and the observation point.

[0056] Furthermore, the formula for the short-circuit impulse current is:

[0057] (3);

[0058] in, is the instantaneous current, is the peak voltage on the high side, is the equivalent short-circuit impedance from the power supply to the fault point, is the initial phase angle when the short circuit occurs, is the time constant, t is the time, is the angular frequency.

[0059] The short-circuit current can be divided into two parts: the transient component (decaying to zero over time) and the steady-state component (changing in a sinusoidal manner). Therefore, according to formula (3), when and When the transient and steady-state components reach their maximum values, their directions are opposite. When a 50Hz transformer is connected, the short-circuit current reaches its maximum value at t=0.01s. This short-circuit current can typically reach 10-30 times the normal current.

[0060] Substituting Equation (2) into Equation (1) yields the expression for the electromagnetic force F removing the magnetic field (Equation 4), where the current term is in square form. Consequently, when a transformer experiences a sudden short-circuit fault, the short-circuit current surges, generating severe impact loads in the windings and significantly increasing the risk of mechanical deformation.

[0061] ( );

[0062] can be determined by the voltage U, the system and the short-circuit impedance of the winding express:

[0063] ( );

[0064] Furthermore, this example establishes a three-dimensional electromagnetic-structural coupling model of a 110kV oil-immersed power transformer to study the interaction between magnetic fields and electromagnetic forces under short-circuit (SC) conditions (parameters see Table 1). Leveraging the electromagnetic module of finite element simulation software and the mechanical analysis tools of Statics Tructure, this model integrates the core and winding geometry, nonlinear material properties, and hysteresis effects, following the Hyun-MoAhn framework. Based on the finite element discretization principle, the Maxwell equations (Equations 6–9) are solved to obtain the spatiotemporal distribution of the winding magnetic flux density under transient short-circuit current. The dynamic electrodynamic characteristics are then iteratively calculated using the electromagnetic force formula (Equation 1). Furthermore, this modeling strategy optimizes the winding boundary conditions to significantly improve computational efficiency while ensuring magnetic field calculation accuracy. A three-dimensional simulation model was specifically constructed to analyze the asymmetric electromagnetic-mechanical coupling caused by winding deformation. This method effectively balances computational complexity with the integrity of the physical phenomenon representation, providing a reliable numerical analysis tool for transformer short-circuit capability assessment.

[0065] Table 1 110kV power transformer parameters

[0066]

[0067] ( );

[0068] ( );

[0069] ( );

[0070] ( );

[0071] This embodiment establishes a multi-scale simulation model of a double-winding three-phase transformer and a double-winding single-phase transformer, such as Figure 2As shown in the figure, both complex and simple models are set for the two types of models. The complex model uses a layered disc winding structure to accurately characterize the gradient distribution of the leakage magnetic field. The simple model uses a hollow cylinder equivalent winding disc structure to effectively reduce the complexity of the electromagnetic field solution. Both models introduce hysteresis and saturation effects to construct nonlinear equivalent magnetic circuits, thereby enabling electromagnetic-mechanical cross-scale coupled analysis.

[0072] FEM analysis verifies the accuracy of the theoretical model, and the numerical calculation results are consistent with the first peak value of the short-circuit current at t = 0.01s derived from formula (3). The distributed air gap design of the multi-stage pancake winding changes the structure of the leakage magnetic path. Compared with the traditional lumped hollow cylinder model, the increase in equivalent leakage reactance reduces the amplitude of the short-circuit current on the low-voltage side, such as Figure 3 At the same time, both structures exhibit typical exponential decay oscillation characteristics.

[0073] Field-circuit coupled finite element analysis shows significant magnetic flux density (MFD) gradients, such as Figure 4 As shown in the figure, the high-permeability layered structure of the core confines most of the magnetic flux, while a small amount of flux escapes the core structure, resulting in an exponentially decreasing axial MFD with distance. In the xz plane, along a 4000mm test path at a distance of 300mm from the winding along the z-axis, the peak MFD near the winding edge is 0.03151T, which decays to below 0.00158T with increasing distance. The LMF exhibits a bimodal distribution, with the field strength increasing toward the winding edge and decreasing toward the center, forming a localized high-flux region that is crucial for deformation diagnosis.

[0074] There are such Figure 4 The specific current distribution shown can be used to determine the direction of the electromagnetic force based on the left-hand rule. For ease of analysis, the electromagnetic force can be decomposed into axial With axial component , and then make corresponding corrections to formula (1):

[0075] (10);

[0076] (11);

[0077] Where, is the component of the leakage magnetic induction intensity value in the x-axis direction; is the component of the leakage magnetic induction intensity value in the z-axis direction;

[0078] Electromagnetic force analysis using equations (7) and (8) clearly demonstrates the mechanical behavior during a short circuit. The low-voltage (LV) winding is subjected to axial tension and axial centrifugal force, causing it to stretch axially and expand radially. Conversely, the high-voltage (HV) winding faces axial compression and axial centripetal force, resulting in axial buckling and radial contraction. These opposing deformations occur because the phase difference between the winding currents during transient energy conversion causes the direction of the Lorentz force to reverse.

[0079] A multiphysics finite element model based on electromagnetic-mechanical coupling was developed to simulate these characteristics of a 110 kV transformer. The Maxwell stress tensor maps the winding current density to the MFD, which is calculated using finite element simulation software. The resulting force density was applied as a boundary load to the three-dimensional structural model in the multiphysics coupling simulation software, discretized, and the harmonic response was solved. The simulation results confirmed axial extension / compression and axial displacement patterns, consistent with theoretical predictions. Under SC conditions, the maximum displacement increased 26 times compared to normal operation, significantly increasing the risk of deformation. Repeated SC events cause cumulative damage due to material nonlinearity and residual stresses. Cyclic electromagnetic loading increases the dislocation density in the copper conductor, while the insulating spacers exhibit nonlinear compressive deformation, reducing the axial pretension and axial stiffness. The accumulated plastic strain compromises the stability of the winding and, if not addressed, can lead to complete collapse. These findings highlight the need for real-time diagnostics to mitigate irreversible damage in smart grid applications.

[0080] Furthermore, the process of capturing the dynamic leakage magnetic field changes on the transformer tank by using the magnetic sensor array includes:

[0081] The magnetic sensor array is aligned parallel to the axial direction of the winding, symmetrically placed between the iron core and the winding system, with a placement range approximately equal to the winding length and installed on the transformer tank to capture the dynamic leakage magnetic field changes on the transformer tank.

[0082] Furthermore, based on LMF attenuation analysis, sensors were placed close to the windings to maximize detection sensitivity and balance the feasibility of online monitoring, operational safety, and structural integrity. The array was installed on the side wall of the transformer tank, and multi-sensor data fusion mitigated local LMF distortion caused by single-point measurement. Finite element method (FEM) simulations were performed as follows: Figure 5 As shown, a bimodal axial LMF distribution is revealed, revealing three layout principles:

[0083] (1) Aligned parallel to the winding axis.

[0084] (2) Placed symmetrically between the core and the winding system.

[0085] (3) Coverage range similar to the winding length.

[0086] Five equally spaced nodes are symmetrically distributed along the axis, with a node spacing of 200 mm and a total node coverage area of ​​1000 mm, ensuring the accuracy of deformation detection and engineering reliability. Figure 5 This configuration supports the diagnostic system by capturing dynamic LMF changes.

[0087] The high-precision finite element model characterizes the axial and radial winding deformations, such as Figure 6 As shown in Figure 2, the axial model quantifies the compression ratio through variable disc height, while the axial model captures ID contraction and OD expansion through radius adjustment. These parametric models normalize the inputs to the mechanical response analysis and provide accurate deformation feature extraction for the diagnostic algorithm.

[0088] The finite element method is used to analyze the changes of LMF at five axial detection points (point A, point B, point C, point D and point E) outside the winding affected by multi-physical field coupling, as shown in the following figure: Figure 7 shown.

[0089] When the winding undergoes axial deformation, the leakage magnetic flux density in the x-axis direction shows a significant change characteristic with the deformation rate (0.00%-0.08%). Through observation at 5 equally spaced test points, it was found that: except for the central measuring point C, the leakage magnetic induction intensity of the other measuring points (points A, B, D and E) all showed a monotonically increasing trend with the increase of the deformation rate. Specifically, when the deformation rate increases by 0.01%, the magnetic induction intensity increments of the four measuring points A, B, D and E all exceed 150 T. In contrast, the gradient of the central measuring point C is only 29 T, which is significantly lower than other measuring points.

[0090] The evolution law of magnetic induction intensity collected by the sensor array is as follows: Figure 8 As shown in the figure, except for the sensing node C, the leakage magnetic induction intensity of the other nodes (points A, B, D, and E) shows a monotonically decreasing characteristic as the winding deformation rate increases (0.00%→0.08%). In contrast, the magnetic induction intensity of the sensing node C shows an increasing response. Among them, when the winding deformation rate increases by 0.01%, the leakage magnetic induction intensity values ​​of the sensing points a and e decrease by 4 T, sensing point b, d decreases by 2 T, but the sensor node C only increases by 0.17 T. This characteristic change pattern provides a key diagnostic basis for reconstructing mechanical deformation through magnetic field inversion algorithms.

[0091] Analysis of the spatial magnetic field distribution characteristics of the winding under axial and axial deformation conditions reveals that sensors located at the winding ends (Sensor A and Sensor E) exhibit greater sensitivity to changes in leakage magnetic flux density than other measurement points. Therefore, Sensor A or Sensor E, located at the winding ends, can be selected as the deformation assessment point for evaluating the severity of winding deformation. In this embodiment, Sensor A is designated as the reference for deformation assessment because it has optimal response characteristics for capturing changes in electromagnetic parameters caused by mechanical displacement. This selection is based on the electromagnetic-mechanical coupling mechanism, which states that the end regions exhibit stronger magnetic field gradient changes when the structure deforms.

[0092] When the axial deformation rate of the winding is 0.01%, the leakage magnetic flux density increases by 0.17 mT, while the same radial deformation rate only causes a decrease of 0.004 mT, a sensitivity difference of 42.5 times. This characteristic indicates that under the combined deformation condition, the change in the leakage magnetic flux density at sensing point A mainly reflects the cumulative effect of the axial deformation.

[0093] In order to eliminate the mutual influence between the axial and radial directions, this embodiment adds sensor point C, and through multi-sensor comprehensive judgment, the accuracy of the axial and radial composite deformation judgment is improved. The specific judgment process is as follows:

[0094] The leakage magnetic induction intensity of each sensor point A and sensor point C is monitored and recorded as , Substitute the values ​​of these two points into the deformation state judgment formula to diagnose the axial deformation and radial deformation :

[0095] (12);

[0096] Among them, the shape variable calculation parameters , can be obtained in the following ways: Under the premise of safety, obtain the transformer axial ( ) or radial ( ) Under small deformation, the axial deformation leakage magnetic induction intensity of points A and C and , radial deformation leakage magnetic induction intensity and According to the test results, the least square method is used to substitute into formula (13), formula (14), formula (15), formula (16) (axial and radial The relationship between the deformation rate and the leakage magnetic induction intensity at points A and C is used for calculation:

[0097] (13);

[0098] (14);

[0099] (15);

[0100] (16);

[0101] in, and is the flux density caused by the axial or axial deformation of the sensor position A; and are the flux density caused by axial or axial deformation at the sensor position C, respectively; and is a constant coefficient.

[0102] Calculating different winding deformation rates, the axial error is generally smaller than the radial error, but the judgment error is less than 1.3%. It is sufficient to accurately identify the axial and radial deformation of the winding. The statistical analysis of the winding deformation error is shown in Table 2. The judgment process is as follows Figure 9 shown.

[0103] Table 2 Statistical analysis of winding deformation error

[0104]

[0105] This example develops a non-invasive, real-time diagnosis method for transformer winding deformation during short-circuit (SC) faults, leveraging leakage magnetic field (LMF) analysis to improve smart grid reliability. By integrating multi-physics modeling, an optimized sensor array, and advanced algorithms, this method achieves precise deformation quantification with millisecond-level response, overcoming the limitations of traditional offline diagnosis.

[0106] This embodiment uses the finite element method to establish a three-dimensional electromagnetic-structural coupling model to quantify the nonlinear relationship between superconducting current and mechanical deformation. Driven by opposite electromagnetic forces, the low-voltage winding exhibits axial expansion and axial elongation, while the high-voltage winding exhibits axial contraction and axial buckling. The optimized magnetic sensor array is arranged along the winding axis and covers its length to capture dynamic LMF changes. The least squares-based magnetic field inversion algorithm decouples the axial and axial deformations, achieving an axial error of ≤0.028%, an axial error of ≤1.21%, and an overall error of <1.3%. This method surpasses traditional methods such as dissolved gas analysis and frequency response analysis, providing rapid fault detection and predictive maintenance capabilities.

[0107] The method of this embodiment provides an economical and safe alternative to invasive diagnostics, enabling millisecond-level early warning using micromagnetic sensors. By accurately identifying phase change patterns, it supports predictive maintenance, improving the reliability of smart grid equipment and reducing the risk of power outages. Its non-invasive design facilitates integration into existing transformer systems, meeting the needs of advanced condition monitoring.

[0108] Example 2

[0109] This embodiment further discloses a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method described in the first embodiment.

[0110] Example 3

[0111] This embodiment further discloses a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described in the first embodiment are implemented.

[0112] Example 4

[0113] This embodiment further discloses a computer program product, including a computer program, which implements the steps of the method described in the first embodiment when executed by a processor.

[0114] The above are merely preferred embodiments of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A method for diagnosing winding deformation under a transformer short-circuit fault, characterized in that: include: Develop and construct a three-dimensional electromagnetic structure finite element model, simulate the interaction between the magnetic field and the electromagnetic force under short-circuit conditions based on the three-dimensional electromagnetic structure finite element model, and establish a nonlinear mapping relationship between the leakage magnetic field distribution and the mechanical deformation; Capturing the dynamic leakage magnetic field changes on the transformer tank through a magnetic sensor array; Based on the least squares magnetic field inversion algorithm, the magnetic flux density is associated with the deformation amplitude, the winding deformation is diagnosed, and the diagnosis result is obtained.

2. The method according to claim 1, characterized in that The three-dimensional electromagnetic structure finite element model is developed based on a 110 kV transformer to simulate the electromagnetic-mechanical interaction of the winding under short-circuit fault conditions.

3. The method according to claim 1, characterized in that The process of establishing a nonlinear mapping relationship between leakage magnetic field distribution and mechanical deformation includes: The electromagnetic force and short-circuit impulse current under short-circuit fault are obtained. Based on the three-dimensional electromagnetic structure finite element model and combined with the finite element discretization principle, the spatiotemporal distribution law of the winding magnetic flux density under the action of transient short-circuit current is obtained, and the dynamic electrodynamic characteristics are iteratively calculated using the electromagnetic force formula.

4. The method according to claim 3, characterized in that The electromagnetic force The formula expression is: ； Where L is the effective conductor length, I is the current density, and B is the magnetic induction intensity value; According to the Biot-Savart law, the magnetic induction intensity value B is expressed in terms of current I as: ； Where μ is the magnetic permeability, dl is the differential current element vector of the source, and e r is the unit vector of the current element pointing to the field point, and r is the straight-line distance between the current element and the observation point.

5. The method according to claim 3, characterized in that The formula for the short-circuit impulse current is: ； in, is the instantaneous current, is the peak voltage on the high side, is the equivalent short-circuit impedance from the power supply to the fault point, is the initial phase angle when the short circuit occurs, is the time constant, t is the time, is the angular frequency.

6. The method according to claim 1, wherein The process of capturing the dynamic leakage magnetic field changes on the transformer tank through the magnetic sensor array includes: The magnetic sensor array is aligned parallel to the axial direction of the winding and symmetrically placed between the iron core and the winding system. The placement range includes the winding length and is installed on the transformer tank to capture the dynamic leakage magnetic field changes on the transformer tank.

7. The method according to claim 1, characterized in that Based on the least squares magnetic field inversion algorithm, the magnetic flux density is correlated with the deformation amplitude to diagnose the winding deformation. The process of obtaining the diagnosis result includes: Based on the least squares magnetic field inversion algorithm, the corresponding deformation rates are calculated according to the axial and radial deformation characteristics of the winding, and the deformation of the winding is quantitatively diagnosed according to the deformation rates.

Images