Vibration monitoring method and system for oil-immersed transformer

Abstract

The invention relates to the technical field of vibration monitoring, and provides a vibration monitoring method and system for an oil-immersed transformer, and the method comprises the steps: judging whether the pressing force of laminations is attenuated or not based on the vibration energy data between the laminations, based on strain data obtained by carrying out full-field strain analysis on the displacement data of the winding, judging whether the winding has a deformation or displacement risk or not; based on the strain data of the welding seam of the box body, predicting the fatigue crack risk of the welding seam, and based on the pre-tightening force data of the bolt of the box body, judging whether the bolt is loosened; based on acoustic emission signal data generated by oil flow impact, the abrasion state of the insulation structure is evaluated, and based on vibration mode data of the cooling fins, whether the cooling fins resonate or not is judged; and when any judgment result or evaluation result has an abnormal state, generating an early warning signal.

Description

Technical Field

[0001] This invention relates to the field of vibration monitoring technology, and in particular to a vibration monitoring method and system for oil-immersed transformers. Background Technology

[0002] During the transportation of oil-immersed transformers, the vibration environment is complex and intense, and existing monitoring methods have significant limitations: most existing monitoring points are concentrated on the outer wall of the tank and the bushing root, which can only capture the vibration amplitude and frequency of these exposed components. It is difficult to fully perceive the vibration impact on internal core components and vulnerable structures. For example, it is impossible to monitor in real time the insulation wear caused by the friction between the coil and the insulation material, coil displacement, increased losses and local overheating caused by the loosening of the core laminations, the risk of heat sink fracture due to resonance, the hidden danger of sealing failure caused by cracking of the outer shell weld and loosening of fixing bolts, and the hidden problems such as insulation aging caused by the increased impact of vibration on the oil flow. As a result, after the transformer is put into operation, there may be operational accidents such as performance degradation, short circuits, and leakage due to the structural defects accumulated in the early stage. There is a lack of systematic monitoring and risk prediction mechanism for the multi-dimensional vibration impact in the entire transportation process, making it difficult to detect vibration damage to components during transportation in a timely manner, and there is a problem of delayed early warning of potential faults. Summary of the Invention

[0003] To address the aforementioned shortcomings, the present invention aims to provide a vibration monitoring method and system for oil-immersed transformers, which comprehensively monitors the vibration of oil-immersed transformers during transportation, provides timely warnings of potential risks, and ensures the safety and reliability of the equipment.

[0004] To achieve this objective, the present invention adopts the following technical solution: A vibration monitoring method for an oil-immersed transformer, the oil-immersed transformer comprising a casing, core columns, windings, and heat sinks; The winding is mounted on the iron core column, the iron core column is composed of multiple layers of laminated sheets stacked sequentially, the winding is provided with insulating support bars, and the heat sink is connected to the housing through an oil circuit interface and communicates with the oil inside the housing; The gaps in the insulating support bar are provided with fiber optic sensors embedded in a spiral structure for acquiring displacement data of the winding. The vibration monitoring method for oil-immersed transformers includes the following steps: S1: Acquire the vibration energy data between the laminations, the displacement data of the windings, the vibration mode data of the heat sink, the strain data of the weld seam of the housing, the preload data of the bolts of the housing, and the acoustic emission signal data generated by the oil flow impact at the insulation structure. S2: Based on the vibration energy data between the laminations, determine whether the clamping force of the laminations has decreased; based on the strain data obtained by full-field strain analysis of the displacement data of the winding, determine whether the winding has deformation or displacement risk. S3: Based on the strain data of the weld of the box body, predict the fatigue crack risk of the weld, and based on the preload data of the bolts of the box body, determine whether the bolts are loose; S4: Based on the acoustic emission signal data generated by the oil flow impact, assess the wear state of the insulation structure, and based on the vibration mode data of the heat sink, determine whether the heat sink resonates; S5: When any judgment or evaluation result in steps S2 to S4 is abnormal, an early warning signal is generated.

[0005] Preferably, in step S2, determining whether the winding has a risk of deformation or displacement includes: The axial pressure deformation of the winding is sensed in real time, and the displacement field data of the winding is acquired when the displacement reaches or exceeds the preset displacement threshold. A two-dimensional rectangular coordinate system is established with the winding axis as the x-direction and the radial direction as the y-direction. The displacement field is spatially differentiated, and the displacement gradient is calculated using the central difference method. The following relation is satisfied: ; in, This represents the displacement component of the displacement field in the x-direction. This indicates the optimized step size set based on the spacing of the insulating support bars, where x and y represent the axial and radial position coordinates of the winding in the two-dimensional rectangular coordinate system, respectively. Based on the displacement gradient Substituting the components into the Cauchy equation to synthesize the full-field strain components, we obtain the normal strain in the x-direction. y-direction normal strain and shear strain in the xy plane The following relation is satisfied: ; in, Indicates the displacement field at The displacement components in the direction are represented, and the total strain is expressed by the strain tensor matrix. The following relation is satisfied: ; Among them, shear strain The correction is made by the angle between the spiral arrangement direction of the fiber optic sensor and the principal stress of the winding; Calculate the von Mises equivalent strain The following relation is satisfied: ; in, This indicates that the von Mises effect changes, when When the strain exceeds the preset strain threshold, a deformation warning is triggered. Curvature analysis is performed on the displacement field to identify regions where the second derivative of the displacement changes abruptly, and the curvature is... Satisfying the relation: ; in, Indicates curvature. This represents the displacement component of the displacement field in the vertical direction, when When the curvature exceeds the preset curvature threshold, the region where the second derivative of the displacement changes abruptly is marked as a deformation risk point.

[0006] Preferably, determining whether the clamping force of the stacked sheets has attenuated based on the vibration energy data between the stacked sheets includes: When the vibration energy entropy between the stacked pieces is greater than the preset vibration energy entropy threshold, it is determined that the clamping force between the stacked pieces is in a decaying state.

[0007] Preferably, determining whether the heat sink resonates based on its vibration mode data includes: Modal responses within a preset high-frequency range are selected from vibration modal data; Analyzing the frequency domain characteristics of the modal response, when the vibration energy is concentrated in a narrow frequency band with a width less than a preset bandwidth threshold and the amplitude of the vibration energy in the narrow frequency band exceeds a preset multiple threshold, it is determined that there is a sudden increase in narrowband vibration energy. Verify whether the sudden increase in the narrowband vibration energy conforms to the preset mode shape characteristics. If so, determine that the heat sink has resonated.

[0008] Preferably, step S3 includes: Calculate the cumulative amount of cyclic plastic strain at the weld. ,when When the cumulative value exceeds a preset threshold, it is determined that there is a risk of fatigue cracks. Calculate the preload decay rate at the bolt. The following relation is satisfied: ; in, This indicates the initial preload of the bolt. This indicates the remaining preload measured at the current moment. Based on the preload decay rate, the bolt loosening level is determined.

[0009] Preferably, in step S4, assessing the wear condition of the insulation structure based on acoustic emission signal data generated by oil flow impact includes: A CFD-FEM joint model of the enclosure-oil-insulation structure was established. The CFD-FEM joint model includes a fluid domain and a structural domain, wherein: The fluid domain is described by the transient incompressible Navier-Stokes equations, satisfying the following relationship: ; in, Indicates the density of the oil. Represents the oil velocity vector. Indicates time, Indicates oil pressure, Indicates the dynamic viscosity of the oil. Represents the gravitational acceleration vector. This represents the volume force transformed from the vibrational spectrum; The fluid domain uses the VOF model to track the oil-gas interface, satisfying the following relationship: ;in, This indicates the volume fraction of the oil phase, and ; The fluid domain adopts The turbulence model satisfies the following relationship: ; in, Represents turbulent kinetic energy. Indicates the turbulent specific dissipation rate. Represents the turbulence generation term. and Represents the constants of the turbulence model. Indicates turbulent eddy viscosity. Indicates the kinematic viscosity of a fluid; The structural domain is described by elastic dynamics equations to represent the insulation structure response, satisfying the following relationship: ; in, Indicates the density of structural materials. Represents the structural displacement vector. Represents the structural damping matrix. Represents the stiffness matrix of the insulation structure. This represents the force vector exerted by the fluid on the insulating structure. This indicates a vibration load that acts directly on the insulating structure; The constitutive model of the insulation structure is orthotropic and satisfies the following relation: ; in, Represents the components of the Cauchy stress tensor. The stiffness tensor represents the stiffness of the insulation structure. Represents the components of the strain tensor; Data exchange between the fluid domain and the structural domain is achieved through a bidirectional fluid-structure interaction (FSI) interface, where: The fluid-to-solid coupling is transmitted to the surface of the insulating structure through integrated fluid pressure and shear stress, satisfying the following relationship: ,in, This represents the fluid-structure interaction interface. Represents the unit normal vector of the interface; The solid-to-fluid coupling updates the fluid mesh boundary through the displacement of the insulating structure, satisfying the following relationship: And the ALE mesh deformation method is adopted. Indicates the velocity of the fluid; The measured vibration spectrum was preprocessed into volume force. and structural boundary loads Among them, volume force satisfy: ,in , Represents the acceleration power spectral density. This represents the phase randomization function.

[0010] Preferably, assessing the wear condition of the insulation structure includes: The signal components within a preset characteristic frequency band are extracted from the acoustic emission signal data using a CFD-FEM joint model, and the pulse count rate N and amplitude integral V of the acoustic emission signal within the characteristic frequency band are calculated. The wear level of the insulation structure is determined based on the combined characteristics of the pulse count rate N and the amplitude integral V. If N is less than the first count rate threshold and V is less than the first amplitude threshold, the wear level is determined to be safe; if N is greater than the second count rate threshold and V is greater than the second amplitude threshold, the wear level is determined to be high-risk; otherwise, the wear level of the insulation structure is determined to be warning. Among them, the second count rate threshold is greater than the first count rate threshold, and the second amplitude threshold is greater than the first amplitude threshold.

[0011] A vibration monitoring system for an oil-immersed transformer, the oil-immersed transformer comprising a housing, core columns, windings, and heat sinks; The winding is mounted on the iron core column, the iron core column is composed of multiple layers of laminated sheets stacked sequentially, the winding is provided with insulating support bars, and the heat sink is connected to the housing through an oil circuit interface and communicates with the oil inside the housing; The gaps in the insulating support bar are provided with fiber optic sensors embedded in a spiral structure for acquiring displacement data of the winding. The vibration monitoring system for oil-immersed transformers includes: The data acquisition module is used to acquire vibration energy data between the laminations, displacement data of the windings, vibration mode data of the heat sink, strain data of the weld seam of the housing, preload data of the bolts of the housing, and acoustic emission signal data generated by oil flow impact at the insulation structure. The first monitoring module is used to determine whether the clamping force of the laminations has decayed based on the vibration energy data between the laminations, and to determine whether there is a risk of deformation or displacement of the winding based on the strain data obtained by full-field strain analysis of the displacement data of the winding. The second monitoring module is used to predict the fatigue crack risk of the weld based on the strain data of the weld of the box body, and to determine whether the bolts are loose based on the preload data of the bolts of the box body. The third monitoring module is used to assess the wear state of the insulation structure based on the acoustic emission signal data generated by the oil flow impact, and to determine whether the heat sink resonates based on the vibration mode data of the heat sink. The early warning signal generation module is used to generate an early warning signal when any judgment result or evaluation result is in an abnormal state.

[0012] Preferably, the first monitoring module is further configured to: The axial pressure deformation of the winding is sensed in real time, and the displacement field data of the winding is acquired when the displacement reaches or exceeds the preset displacement threshold. A two-dimensional rectangular coordinate system is established with the winding axis as the x-direction and the radial direction as the y-direction. The displacement field is spatially differentiated, and the displacement gradient is calculated using the central difference method. The following relation is satisfied: ; in, This represents the displacement component of the displacement field in the x-direction. This indicates the optimized step size set based on the spacing of the insulating support bars, where x and y represent the axial and radial position coordinates of the winding in the two-dimensional rectangular coordinate system, respectively. Based on the displacement gradient Substituting the components into the Cauchy equation to synthesize the full-field strain components, we obtain the normal strain in the x-direction. y-direction normal strain and shear strain in the xy plane The following relation is satisfied: ; in, Indicates the displacement field at The displacement components in the direction are represented, and the total strain is expressed by the strain tensor matrix. The following relation is satisfied: ; Among them, shear strain The correction is made by the angle between the spiral arrangement direction of the fiber optic sensor and the principal stress of the winding; Calculate the von Mises equivalent strain The following relation is satisfied: ; in, This indicates that the von Mises effect changes, when When the strain exceeds the preset strain threshold, a deformation warning is triggered. Curvature analysis is performed on the displacement field to identify regions where the second derivative of the displacement changes abruptly, and the curvature is... Satisfying the relation: ; in, Indicates curvature. This represents the displacement component of the displacement field in the vertical direction, when When the curvature exceeds the preset curvature threshold, the region where the second derivative of the displacement changes abruptly is marked as a deformation risk point.

[0013] Preferably, the third monitoring module is further used for: A CFD-FEM joint model of the enclosure-oil-insulation structure was established. The CFD-FEM joint model includes a fluid domain and a structural domain, wherein: The fluid domain is described by the transient incompressible Navier-Stokes equations, satisfying the following relationship: ; in, Indicates the density of the oil. Represents the oil velocity vector. Indicates time, Indicates oil pressure, Indicates the dynamic viscosity of the oil. Represents the gravitational acceleration vector. This represents the volume force transformed from the vibrational spectrum; The fluid domain uses the VOF model to track the oil-gas interface, satisfying the following relationship: ;in, This indicates the volume fraction of the oil phase, and ; The fluid domain adopts The turbulence model satisfies the following relationship: ; in, Represents turbulent kinetic energy. Indicates the turbulent specific dissipation rate. Represents the turbulence generation term. and Represents the constants of the turbulence model. Indicates turbulent eddy viscosity. Indicates the kinematic viscosity of a fluid; The structural domain is described by elastic dynamics equations to represent the insulation structure response, satisfying the following relationship: ; in, Indicates the density of structural materials. Represents the structural displacement vector. Represents the structural damping matrix. Represents the stiffness matrix of the insulation structure. This represents the force vector exerted by the fluid on the insulating structure. This indicates a vibration load that acts directly on the insulating structure; The constitutive model of the insulation structure is orthotropic and satisfies the following relation: ; in, Represents the components of the Cauchy stress tensor. The stiffness tensor represents the stiffness of the insulation structure. Represents the components of the strain tensor; Data exchange between the fluid domain and the structural domain is achieved through a bidirectional fluid-structure interaction (FSI) interface, where: The fluid-to-solid coupling is transmitted to the surface of the insulating structure through integrated fluid pressure and shear stress, satisfying the following relationship: ,in, This represents the fluid-structure interaction interface. Represents the unit normal vector of the interface; The solid-to-fluid coupling updates the fluid mesh boundary through the displacement of the insulating structure, satisfying the following relationship: And the ALE mesh deformation method is adopted. Indicates the velocity of the fluid; The measured vibration spectrum was preprocessed into volume force. and structural boundary loads Among them, volume force satisfy: ,in , Represents the acceleration power spectral density. This represents the phase randomization function.

[0014] One of the above technical solutions has the following advantages or beneficial effects: This invention achieves full-area vibration monitoring of the core components and vulnerable structures inside an oil-immersed transformer at key locations. By comparing vibration energy entropy thresholds, it directly reflects the attenuation state of the lamination clamping force, enabling early identification of potential core lamination loosening and preventing localized overheating and increased losses due to lamination friction. By acquiring winding displacement data and calculating the full-field strain components using the center difference method and Cauchy equation, combined with von Mises equivalent strain and curvature analysis, it accurately assesses whether there is a risk of winding deformation or displacement, enabling early detection of localized warping or displacement problems and avoiding short-circuit hazards caused by winding deformation. Furthermore, by monitoring the strain components of the tank welds… This system uses the cumulative amount of cyclic plastic strain to predict the risk of weld cracking, and judges the degree of bolt loosening based on the preload decay rate. It can provide early warning of weld cracking and bolt loosening risks, avoiding oil leakage due to seal failure. By analyzing frequency domain characteristics, it can identify the phenomenon of sudden increase in narrowband vibration energy, and verify whether the heat sink is resonating by combining modal vibration characteristics. It can accurately detect the resonance risk of the heat sink and avoid fatigue fracture caused by resonance. By combining the CFD-FEM joint model to analyze the pulse count rate and amplitude integral of the acoustic emission signal, it can quantitatively assess the wear level of the insulation structure, and detect insulation aging problems in advance, avoiding short circuit risks caused by insulation failure. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0016] Figure 1 This is a flowchart of a vibration monitoring method for oil-immersed transformers provided in an embodiment of the present invention; Figure 2 This is a first structural schematic diagram of an oil-immersed transformer provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the second structure of the oil-immersed transformer provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the third structure of the oil-immersed transformer provided in an embodiment of the present invention; Figure 5 This is a schematic diagram of the fourth structure of the oil-immersed transformer provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of the vibration monitoring system for an oil-immersed transformer provided in an embodiment of the present invention; The components include: housing 1, iron core column 2, laminate 20, thin-film PVDF piezoelectric sensor 23, winding 3, insulating support bar 31, heat sink 4, heat sink 43, and MEMS triaxial accelerometer 44. Detailed Implementation

[0017] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0018] In this invention, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0019] A vibration monitoring method for oil-immersed transformers, such as Figure 2-5 As shown, the oil-immersed transformer includes a housing 1, a core column 2, windings 3, and heat sinks 43; The winding 3 is mounted on the iron core column 2. The iron core column 2 is composed of multiple layers of laminated sheets 20 stacked sequentially. The winding 3 is provided with insulating support strips 31. The radiator 4 is connected to the surface of the housing 1. The heat sink 43 on the radiator 4 is connected to the housing 1 through the oil circuit interface and is in communication with the oil inside the housing 1. An optical fiber sensor for acquiring displacement data of the winding 3 is embedded in the gap of the insulating support bar 31 in a spiral structure. In one embodiment, such as Figure 2-5 As shown, the oil-immersed transformer has a piezoresistive washer sensor at the bolt of the housing 1, an acoustic sensor in the insulating support bar 31 and the rest of the insulating structure of the oil-immersed transformer, a thin film PVDF piezoelectric sensor 23 at the joint between the laminations, a MEMS triaxial accelerometer 44 attached to the root or welding point of the heat sink 43, and a metal foil strain gauge array at the weld of the housing. like Figure 1 As shown, the vibration monitoring method for oil-immersed transformers includes the following steps: S1: Acquire the vibration energy data between the laminations, the displacement data of the windings, the vibration mode data of the heat sink, the strain data of the weld seam of the housing, the preload data of the bolts of the housing, and the acoustic emission signal data generated by the oil flow impact at the insulation structure. It should be noted that the vibration energy data between laminations refers to the vibration energy information between the laminations of the iron core collected by a thin-film PVDF piezoelectric sensor, and its function is to reflect the compaction state of the laminations through the energy entropy value; the displacement data of the winding refers to the winding displacement vector obtained by a fiber optic sensor (such as FBG fiber) embedded in the gap of the insulating support bar with a helical structure, and its function is to provide input for full-field strain analysis; the vibration modal data of the heat sink refers to the vibration frequency, amplitude and phase information of the heat sink collected by a MEMS triaxial accelerometer, and its function is to identify resonance risk through modal response; the strain data of the weld seam of the housing refers to the strain distribution on the weld seam surface measured by a metal foil strain gauge grid array, and its function is to predict fatigue cracks through the cumulative amount of cyclic plastic strain; the preload data of the bolts refers to the bolt tightening force change measured in real time by a piezoresistive washer sensor, and its function is to judge the degree of loosening through the attenuation rate; the acoustic emission signal data refers to the sound wave signal generated by the oil flow impact at the insulating structure captured by an acoustic sensor (such as a miniature acoustic emission sensor), and its function is to assess the insulation wear state through pulse counting and amplitude integration. The technologies employed include high-frequency sampling sensors (such as PVDF arrays with a sampling rate ≥10kHz), optical measurements (such as FBG wavelength resolution ≤1pm), and acoustic emission analysis (frequency range 50k-1MHz).

[0020] The purpose of step S1 is to synchronously collect vibration-related data of key components of the transformer by deploying sensors in multiple locations, so as to provide comprehensive input for subsequent analysis.

[0021] S2: Based on the vibration energy data between the laminations, determine whether the clamping force of the laminations has decreased; based on the strain data obtained by full-field strain analysis of the displacement data of the winding, determine whether the winding has deformation or displacement risk. The full-field strain analysis includes calculating displacement gradients (such as the central difference method), synthesizing strain components using the Cauchy equation, and calculating von Mises equivalent strain and curvature. The techniques employed involve spatial differential algorithms and strain tensor calculations. Step S2 aims to identify potential core lamination loosening and winding deformation hazards early through vibration energy entropy and displacement-strain analysis. Vibration energy entropy threshold comparison directly indicates clamping force attenuation, preventing overheating due to lamination friction. Strain analysis of displacement data can locate high-strain regions, and equivalent strain and curvature thresholds provide early warnings of winding warping or displacement, thereby addressing the problem of delayed early warning of potential faults caused by internal component vibration damage.

[0022] S3: Based on the strain data of the weld of the box body, predict the fatigue crack risk of the weld, and based on the preload data of the bolts of the box body, determine whether the bolts are loose; Understandably, the purpose of step S3 is to prevent structural fatigue and sealing failure by monitoring weld strain accumulation and bolt preload changes. Threshold comparisons of accumulated strain can provide early warning of weld cracks, preventing oil leaks; grading of preload decay rates quantifies bolt loosening, enabling targeted maintenance and addressing sealing hazards caused by external structural vibration damage. This step enhances overall structural reliability by continuously tracking vulnerable points.

[0023] S4: Based on the acoustic emission signal data generated by the oil flow impact, assess the wear state of the insulation structure, and based on the vibration mode data of the heat sink, determine whether the heat sink resonates; Understandably, the purpose of step S4 is to detect wear of the insulating material and resonance of the heat sink through acoustic signal and vibration modal analysis. Acoustic emission characteristics can reflect the degree of wear of the insulating structure under the impact of oil flow, preventing insulation aging; vibration modal analysis can identify the resonant frequency, avoiding fatigue fracture of the heat sink, thereby solving the hidden problems caused by fluid-structure interaction.

[0024] S5: When any judgment or evaluation result in steps S2 to S4 is abnormal, an early warning signal is generated.

[0025] Preferably, in step S2, determining whether the winding has a risk of deformation or displacement includes: The axial pressure deformation of the winding is sensed in real time, and the displacement field data of the winding is acquired when the displacement reaches or exceeds the preset displacement threshold. Axial pressure deformation refers to the dimensional change in the length direction of the winding caused by axial mechanical force during transformer transportation. This is monitored using a micro-bend-sensitive FBG fiber optic sensor, with a measurement accuracy of up to 0.1 mm. The displacement threshold is a safety limit set based on the mechanical properties of the winding insulation material, typically 0.1 mm. Displacement field data refers to the three-dimensional displacement distribution dataset on the winding surface acquired through a distributed fiber optic sensor network. This step utilizes high-precision fiber optic sensing technology to monitor minute deformations of the winding in real time, solving the problem of traditional methods failing to detect potential winding displacement issues promptly. In one embodiment, 4-6 measuring points are arranged on the high-voltage and low-voltage sides of each phase winding, respectively. The sensors are embedded in the gaps of the insulation support bars with a spiral structure, and the sampling frequency is set to 100 Hz. When any measuring point detects an axial displacement exceeding 0.1 mm, the system immediately activates all 128 sensing points in the network for synchronous data acquisition. The wavelength offset is converted into displacement using an optical demodulator, generating a three-dimensional displacement field matrix including timestamps.

[0026] A two-dimensional rectangular coordinate system is established with the winding axis as the x-direction and the radial direction as the y-direction. The displacement field is spatially differentiated, and the displacement gradient is calculated using the central difference method. The following relation is satisfied: ; in, This represents the displacement component of the displacement field in the x-direction. This indicates the optimized step size set based on the spacing of the insulating support bars, where x and y represent the axial and radial position coordinates of the winding in the two-dimensional rectangular coordinate system, respectively. In this invention, a displacement field is used to describe the movement of each point. In a continuous structure such as a transformer coil, every point undergoes displacement. and It is in space Each point in the equation has defined physical quantities, such as... It's about location. The function, refer to From the formula, we can obtain .

[0027] Spatial differentiation refers to the calculation of the rate of change of displacement field data in a spatial direction; the central difference method is a numerical differentiation method that calculates the gradient by dividing the difference in displacement values ​​of adjacent measuring points by the distance between measuring points; the displacement gradient represents the rate of change of the displacement vector in a specific direction and reflects the characteristics of strain distribution.

[0028] This paper addresses the challenge of directly measuring strain distribution by extracting gradient information from the displacement field using numerical calculation methods. Compared to forward or backward differencing, the central difference method offers higher computational accuracy and can accurately capture strain concentration at the gaps in the insulating struts, providing reliable input for subsequent strain analysis.

[0029] Based on the displacement gradient Substituting the components into the Cauchy equation to synthesize the full-field strain components, we obtain the normal strain in the x-direction. y-direction normal strain and shear strain in the xy plane The following relation is satisfied: ; in, Indicates the displacement field at The displacement components in the direction are represented, and the total strain is expressed by the strain tensor matrix. The following relation is satisfied: ; Among them, shear strain The correction is made by the angle between the spiral arrangement direction of the fiber optic sensor and the principal stress of the winding; The Cauchy equation is a geometric equation describing the relationship between displacement gradient and strain components; normal strain represents the rate of change of the length of a material element along the coordinate axis; shear strain represents the change of the angle of the material element, reflecting the degree of shear deformation. By converting the displacement gradient into strain components using the Cauchy equation, the problem of converting geometric deformation into mechanical response is solved. Calculating the three strain components allows for a comprehensive description of the deformation characteristics of the winding in different directions, providing complete strain field information for accurately assessing the mechanical stress state.

[0030] Calculate the von Mises equivalent strain The following relation is satisfied: ; in, This indicates that the von Mises effect changes, when When the strain exceeds the preset strain threshold, a deformation warning is triggered. The von Mises equivalent strain is a strength theory that synthesizes multi-dimensional strain components into a single equivalent value to assess the yield risk of materials. The preset strain threshold is set based on the yield strength of the winding insulation material, with an optional value of 60 microstrain. By simplifying complex stress states to a single criterion through equivalent strain, the system solves the problem of comprehensive evaluation of multi-dimensional strain. The von Mises criterion accurately reflects the material response under composite stress states. When the equivalent strain exceeds the threshold, it indicates that the insulation material may have entered the plastic deformation stage, requiring immediate warning. For example, in a certain region, the calculated positive strain is 55 microstrain, the negative strain is 30 microstrain, and the shear strain is 25 microstrain. The equivalent strain calculated using the von Mises formula is 68 microstrain, exceeding the threshold of 8 microstrain, and the system immediately generates a high-level warning signal.

[0031] Curvature analysis is performed on the displacement field to identify regions where the second derivative of the displacement changes abruptly, and the curvature is... Satisfying the relation: ; in, Indicates curvature. This represents the displacement component of the displacement field in the vertical direction, when When the curvature exceeds the preset curvature threshold, the region where the second derivative of the displacement changes abruptly is marked as a deformation risk point.

[0032] Curvature analysis is a method for calculating the degree of local bending using the second derivative of the displacement field. The second derivative of displacement reflects the rate of change of curvature in the displacement field. Deformation risk points refer to critical locations where concentrated local deformation may lead to insulation damage. Curvature analysis enhances the ability to identify local bending deformation and can solve the problem of equivalent strain being insensitive to bending. As a second-order characteristic quantity of the displacement field, curvature can effectively detect micro-bending deformation at the gaps in the insulation struts, identifying potential structural risk points in advance. In one embodiment, three adjacent sensor measurement points are selected, and the curvature value is calculated using a second-order difference formula. A curvature threshold of 0.01 per millimeter is set. When the monitored value exceeds the threshold, a comprehensive risk assessment is performed based on the equivalent strain in that area. The system generates a risk distribution map, marking high-risk areas and their risk levels.

[0033] Preferably, determining whether the clamping force of the stacked sheets has attenuated based on the vibration energy data between the stacked sheets includes: When the vibration energy entropy between the stacked pieces is greater than the preset vibration energy entropy threshold, it is determined that the clamping force between the stacked pieces is in a decaying state.

[0034] It should be noted that the vibration energy entropy threshold is the Shannon entropy value calculated by performing intrinsic mode function decomposition on the vibration signal collected by the PVDF piezoelectric sensor using the Hilbert-Huang Transform. Its physical meaning characterizes the degree of disorder in vibration energy. The preset vibration energy entropy threshold is a critical value determined based on the characteristics of the core lamination material and long-term experimental data, with a typical setting range of 0.32-0.38 J. Clamping force attenuation refers to the state where the mechanical preload between the laminations is lower than the design value, usually caused by misalignment of the silicon steel sheets due to vibration or aging of the insulating pads.

[0035] When the clamping force of the stacked plates is normal, the vibration energy distribution is relatively concentrated and orderly, with a low entropy value. When the clamping force decays, nonlinear collisions and friction occur between the stacked plates, and the vibration energy distribution tends to be dispersed and disordered, with a significant increase in entropy value. By setting a reasonable entropy threshold, the normal tightening state and the abnormal decay state can be effectively distinguished, solving the problem that traditional amplitude monitoring methods are not sensitive to early loosening.

[0036] Preferably, determining whether the heat sink resonates based on its vibration mode data includes: Modal responses within a preset high-frequency range are selected from vibration modal data; Analyzing the frequency domain characteristics of the modal response, when the vibration energy is concentrated in a narrow frequency band with a width less than a preset bandwidth threshold and the amplitude of the vibration energy in the narrow frequency band exceeds a preset multiple threshold, it is determined that there is a sudden increase in narrowband vibration energy. Verify whether the sudden increase in the narrowband vibration energy conforms to the preset mode shape characteristics. If so, determine that the heat sink has resonated.

[0037] It should be noted that vibration modal data refers to the structural dynamic characteristic parameters obtained through modal analysis of the heat sink vibration signals collected by a MEMS triaxial accelerometer, including information such as frequency, damping ratio, and mode shape. The preset high-frequency range is a monitoring frequency band set according to the inherent characteristics of the heat sink structure. Narrow-band vibration energy surge refers to the phenomenon where vibration energy is concentrated in a frequency band with a bandwidth of less than 5Hz and the amplitude exceeds 10 times the background noise. Modal shape characteristics refer to the bending, torsional, or combined mode shapes exhibited by the heat sink at a specific resonant frequency. It can be understood that when the external excitation frequency is close to the natural frequency of the heat sink, a narrow-band vibration response with highly concentrated energy will be generated. Through the dual criteria of frequency domain characteristic analysis and mode shape verification, random vibration and dangerous resonance in the transportation environment can be accurately distinguished, solving the problem of inaccurate identification of resonance characteristics by traditional vibration monitoring methods and avoiding the risk of heat sink fatigue fracture due to resonance.

[0038] Preferably, step S3 includes: Calculate the cumulative amount of cyclic plastic strain at the weld. ,when When the cumulative value exceeds a preset threshold, it is determined that there is a risk of fatigue cracks. Calculate the preload decay rate at the bolt. The following relation is satisfied: ; in, This indicates the initial preload of the bolt. This indicates the remaining preload measured at the current moment. Based on the preload decay rate, the bolt loosening level is determined.

[0039] It should be noted that the cumulative amount of cyclic plastic strain This refers to the cumulative damage amount of weld strain amplitude counted using the rainflow counting method, calculated using Miner's linear cumulative damage theory, and used to assess low-cycle fatigue life; the preset cumulative threshold is determined based on the weld material's SN curve, with a typical value of 1.0; preload attenuation rate. This refers to the initial preload of the bolt. With real-time measurement values The percentage change was measured using a piezoresistive washer sensor; initial preload. Design the bolt tightening force value, remaining preload. The axial pressure value is monitored in real time during transportation; for example, the loosening level can be divided into three levels: <15% is considered normal, 15% ≤ A percentage below 25% warrants a warning. ≥25% indicates looseness.

[0040] By employing a strain cumulative damage model and preload grading assessment, this study addresses the quantitative assessment of two structural hazards caused by transportation vibration: weld fatigue crack propagation and bolt connection failure. Miner's linear cumulative damage theory is used to transform random vibration loads into equivalent fatigue damage, enabling early prediction of weld crack initiation. Graded alarms based on preload attenuation rates distinguish between normal wear and emergency faults, preventing over-maintenance or missed alarms and ensuring the structural integrity of the transformer tank.

[0041] Preferably, in step S4, assessing the wear condition of the insulation structure based on acoustic emission signal data generated by oil flow impact includes: A CFD-FEM joint model of the enclosure-oil-insulation structure was established. The CFD-FEM joint model includes a fluid domain and a structural domain, wherein: The fluid domain is described by the transient incompressible Navier-Stokes equations, satisfying the following relationship: ; in, Indicates the density of the oil. Represents the oil velocity vector. Indicates time, Indicates oil pressure, Indicates the dynamic viscosity of the oil. Represents the gravitational acceleration vector. This represents the volume force transformed from the vibrational spectrum; The fluid domain uses the VOF model to track the oil-gas interface, satisfying the following relationship: ;in, This indicates the volume fraction of the oil phase, and ; The fluid domain adopts The turbulence model satisfies the following relationship: ; in, Represents turbulent kinetic energy. Indicates the turbulent specific dissipation rate. Represents the turbulence generation term. and Represents the constants of the turbulence model. Indicates turbulent eddy viscosity. Indicates the kinematic viscosity of a fluid; The structural domain is described by elastic dynamics equations to represent the insulation structure response, satisfying the following relationship: ; in, Indicates the density of structural materials. Represents the structural displacement vector. Represents the structural damping matrix. Represents the stiffness matrix of the insulation structure. This represents the force vector exerted by the fluid on the insulating structure. This indicates a vibration load that acts directly on the insulating structure; The constitutive model of the insulation structure is orthotropic and satisfies the following relation: ; in, Represents the components of the Cauchy stress tensor. The stiffness tensor represents the stiffness of the insulation structure. Represents the components of the strain tensor; Data exchange between the fluid domain and the structural domain is achieved through a bidirectional fluid-structure interaction (FSI) interface, where: The fluid-to-solid coupling is transmitted to the surface of the insulating structure through integrated fluid pressure and shear stress, satisfying the following relationship: ,in, This represents the fluid-structure interaction interface. Represents the unit normal vector of the interface; The solid-to-fluid coupling updates the fluid mesh boundary through the displacement of the insulating structure, satisfying the following relationship: And the ALE mesh deformation method is adopted. Indicates the velocity of the fluid; The measured vibration spectrum was preprocessed into volume force. and structural boundary loads Among them, volume force satisfy: ,in , Represents the acceleration power spectral density. This represents the phase randomization function.

[0042] It should be noted that the CFD-FEM joint model refers to a multiphysics simulation model that couples computational fluid dynamics and the finite element method. The fluid domain represents the oil flow region inside the transformer (using Eulerian description), and the structural domain represents the solid region of the insulating material (using Lagrangian description). Key parameters involved in this model include: oil density. (Typical value 870 kg / m³) Characterizes the inertial properties of the oil, oil velocity vector. Reflects the flow state, oil pressure Includes hydrostatic and hydrodynamic components, and the dynamic viscosity of the oil. (0.016 Pa·s at 40℃) affects viscous force and gravitational acceleration vector. Considering the influence of gravity, vibrational body force Transforming external vibrations into loads within the fluid domain; oil phase volume fraction in the VOF model. (Value range 0-1) is used to track dynamic changes in the oil-gas interface, from the formula. It can be concluded that the oil-gas interface is a component of volume fraction The range defined by cells between 0 and 1; Turbulent kinetic energy in turbulence models Characterizing turbulence intensity, turbulence specific dissipation rate Reflects dissipation rate, turbulence generation term Generated by the velocity gradient, model constant (0.09) and (1.0) Ensure model stability, turbulent eddy viscosity Reflects turbulent transport capacity; material density within the structural domain (Typical value of insulating paper: 950 kg / m³) affects inertial force and displacement vector. Describes structural deformation, damping matrix Rayleigh damping model is adopted, stiffness matrix Considering orthotropic anisotropy, fluid forces Including pressure and viscous forces, vibration loads External excitation; Cauchy stress in the constitutive model Stiffness tensor reflects internal stress state Includes 21 independent components, strain tensor Describe the degree of deformation; fluid-structure interaction interface To achieve data exchange, unit normal vector Defining direction; acceleration time history in vibration spectrum preprocessing The power spectral density is obtained through the inverse Fourier transform. Phase randomization function based on measured data Preserving spectral characteristics, structural boundary load With volume force This was used to construct transport vibration conditions in a CFD-FEM joint model, under which the oil flow impact effect was characterized by acoustic emission signal data to assess the wear state of the insulation structure.

[0043] Specifically, by establishing a high-precision fluid-structure interaction numerical model, the problem that traditional single-physics simulation cannot accurately reflect the interaction between oil flow impact and insulation structure vibration is solved. The CFD-FEM co-simulation can simultaneously capture the turbulent characteristics of the oil and the dynamic response of the structure. Through a two-way coupling mechanism, it achieves accurate simulation of energy transfer between physical fields, providing more realistic physical field data for insulation wear assessment and significantly improving the accuracy and predictability of condition assessment.

[0044] Preferably, assessing the wear condition of the insulation structure includes: The signal components within a preset characteristic frequency band are extracted from the acoustic emission signal data using a CFD-FEM joint model, and the pulse count rate N and amplitude integral V of the acoustic emission signal within the characteristic frequency band are calculated. The wear level of the insulation structure is determined based on the combined characteristics of the pulse count rate N and the amplitude integral V. If N is less than the first count rate threshold and V is less than the first amplitude threshold, the wear level is determined to be safe; if N is greater than the second count rate threshold and V is greater than the second amplitude threshold, the wear level is determined to be high-risk; otherwise, the wear level of the insulation structure is determined to be warning. Among them, the second count rate threshold is greater than the first count rate threshold, and the second amplitude threshold is greater than the first amplitude threshold.

[0045] It should be noted that the pulse count rate N refers to the number of pulses exceeding a set threshold in the acoustic emission signal per unit time within the characteristic frequency band, reflecting the frequency of microcracks in the insulating material, and is measured in pulses per second; the amplitude integral V refers to the integral area of ​​the signal envelope within the characteristic frequency band, characterizing the degree of accumulation of acoustic emission energy, and is measured in mV·ms; the first count rate threshold (typically 50 pulses / second) and the second count rate threshold (typically 150 pulses / second) are grading criteria set according to the damage mechanism of the insulating material; the first amplitude threshold (typically 100 mV·ms) and the second amplitude threshold (typically 300 mV·ms) correspond to the energy release levels of different damage degrees; the wear level is divided into three levels: safe, warning, and high risk, each corresponding to different maintenance strategies.

[0046] By extracting the time-frequency features of acoustic emission signals and fusing multiple parameters, the challenge of quantitatively assessing the wear state of insulation structures has been solved. Based on the sensitivity of acoustic emission signals to microscopic damage in materials, the pulse count rate is used to reflect the damage frequency, and the amplitude integral is used to reflect the damage intensity. The dual-parameter combination judgment overcomes the risk of misjudgment by a single parameter, realizing full-process monitoring of insulation wear from quantitative to qualitative changes, and providing accurate basis for preventive maintenance.

[0047] For example, a transformer detects acoustic emission signals with characteristic data of N=182 times / second and V=347mV·ms in the 126kHz frequency band, which exceed the second count rate threshold and the second amplitude threshold. The system determines that the insulation wear level is high-risk, immediately generates a red alarm, and suggests checking the condition of the winding insulation paper.

[0048] A vibration monitoring system for an oil-immersed transformer, the oil-immersed transformer comprising a housing, core columns, windings, and heat sinks; The winding is mounted on the iron core column, the iron core column is composed of multiple layers of laminated sheets stacked sequentially, the winding is provided with insulating support bars, and the heat sink is connected to the housing through an oil circuit interface and communicates with the oil inside the housing; The gaps in the insulating support bar are provided with fiber optic sensors embedded in a spiral structure for acquiring displacement data of the winding. The vibration monitoring system for oil-immersed transformers includes: The data acquisition module is used to acquire vibration energy data between the laminations, displacement data of the windings, vibration mode data of the heat sink, strain data of the weld seam of the housing, preload data of the bolts of the housing, and acoustic emission signal data generated by oil flow impact at the insulation structure. The first monitoring module is used to determine whether the clamping force of the laminations has decayed based on the vibration energy data between the laminations, and to determine whether there is a risk of deformation or displacement of the winding based on the strain data obtained by full-field strain analysis of the displacement data of the winding. The second monitoring module is used to predict the fatigue crack risk of the weld based on the strain data of the weld of the box body, and to determine whether the bolts are loose based on the preload data of the bolts of the box body. The third monitoring module is used to assess the wear state of the insulation structure based on the acoustic emission signal data generated by the oil flow impact, and to determine whether the heat sink resonates based on the vibration mode data of the heat sink. The early warning signal generation module is used to generate an early warning signal when any judgment result or evaluation result is in an abnormal state.

[0049] Preferably, the first monitoring module is further configured to: The axial pressure deformation of the winding is sensed in real time, and the displacement field data of the winding is acquired when the displacement reaches or exceeds the preset displacement threshold. A two-dimensional rectangular coordinate system is established with the winding axis as the x-direction and the radial direction as the y-direction. The displacement field is spatially differentiated, and the displacement gradient is calculated using the central difference method. The following relation is satisfied: ; in, This represents the displacement component of the displacement field in the x-direction. This indicates the optimized step size set based on the spacing of the insulating support bars, where x and y represent the axial and radial position coordinates of the winding in the two-dimensional rectangular coordinate system, respectively. Based on the displacement gradient Substituting the components into the Cauchy equation to synthesize the full-field strain components, we obtain the normal strain in the x-direction. y-direction normal strain and shear strain in the xy plane The following relation is satisfied: ; in, Indicates the displacement field at The displacement components in the direction are represented, and the total strain is expressed by the strain tensor matrix. The following relation is satisfied: ; Among them, shear strain The correction is made by the angle between the spiral arrangement direction of the fiber optic sensor and the principal stress of the winding; Calculate the von Mises equivalent strain The following relation is satisfied: ; in, This indicates that the von Mises effect changes, when When the strain exceeds the preset strain threshold, a deformation warning is triggered. Curvature analysis is performed on the displacement field to identify regions where the second derivative of the displacement changes abruptly, and the curvature is... Satisfying the relation: ; in, Indicates curvature. This represents the displacement component of the displacement field in the vertical direction, when When the curvature exceeds a preset threshold, regions where the second derivative of the displacement abruptly changes are marked as deformation risk points.

[0050] Preferably, the third monitoring module is further used for: A CFD-FEM joint model of the enclosure-oil-insulation structure was established. The CFD-FEM joint model includes a fluid domain and a structural domain, wherein: The fluid domain is described by the transient incompressible Navier-Stokes equations, satisfying the following relationship: ; in, Indicates the density of the oil. Represents the oil velocity vector. Indicates time, Indicates oil pressure, Indicates the dynamic viscosity of the oil. Represents the gravitational acceleration vector. This represents the volume force transformed from the vibrational spectrum; The fluid domain uses the VOF model to track the oil-gas interface, satisfying the following relationship: ;in, This indicates the volume fraction of the oil phase, and ; The fluid domain adopts The turbulence model satisfies the following relationship: ; in, Represents turbulent kinetic energy. Indicates the turbulent specific dissipation rate. Represents the turbulence generation term. and Represents the constants of the turbulence model. Indicates turbulent eddy viscosity. Indicates the kinematic viscosity of a fluid; The structural domain is described by elastic dynamics equations to represent the insulation structure response, satisfying the following relationship: ; in, Indicates the density of structural materials. Represents the structural displacement vector. Represents the structural damping matrix. Represents the stiffness matrix of the insulation structure. This represents the force vector exerted by the fluid on the insulating structure. This indicates a vibration load that acts directly on the insulating structure; The constitutive model of the insulation structure is orthotropic and satisfies the following relation: ; in, Represents the components of the Cauchy stress tensor. The stiffness tensor represents the stiffness of the insulation structure. Represents the components of the strain tensor; Data exchange between the fluid domain and the structural domain is achieved through a bidirectional fluid-structure interaction (FSI) interface, where: The fluid-to-solid coupling is transmitted to the surface of the insulating structure through integrated fluid pressure and shear stress, satisfying the following relationship: ,in, This represents the fluid-structure interaction interface. Represents the unit normal vector of the interface; The solid-to-fluid coupling updates the fluid mesh boundary through the displacement of the insulating structure, satisfying the following relationship: And the ALE mesh deformation method is adopted. Indicates the velocity of the fluid; The measured vibration spectrum was preprocessed into volume force. and structural boundary loads Among them, volume force satisfy: ,in , Represents the acceleration power spectral density. This represents the phase randomization function.

[0051] This embodiment implements a vibration monitoring method and process for oil-immersed transformers. Please refer to the above embodiments for details, which will not be repeated here.

[0052] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0053] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A vibration monitoring method for oil-immersed transformers, characterized in that, The oil-immersed transformer includes a casing, core columns, windings, and heat sinks; The winding is mounted on the iron core column, the iron core column is composed of multiple layers of laminated sheets stacked sequentially, the winding is provided with insulating support bars, and the heat sink is connected to the housing through an oil circuit interface and communicates with the oil inside the housing; The gaps in the insulating support bar are provided with fiber optic sensors embedded in a spiral structure for acquiring displacement data of the winding. The vibration monitoring method for oil-immersed transformers includes the following steps: S1: Acquire the vibration energy data between the laminations, the displacement data of the windings, the vibration mode data of the heat sink, the strain data of the weld seam of the housing, the preload data of the bolts of the housing, and the acoustic emission signal data generated by the oil flow impact at the insulation structure. S2: Based on the vibration energy data between the laminations, determine whether the clamping force of the laminations has decreased; based on the strain data obtained by full-field strain analysis of the displacement data of the winding, determine whether the winding has deformation or displacement risk. S3: Based on the strain data of the weld of the box body, predict the fatigue crack risk of the weld, and based on the preload data of the bolts of the box body, determine whether the bolts are loose; S4: Based on the acoustic emission signal data generated by the oil flow impact, assess the wear state of the insulation structure, and based on the vibration mode data of the heat sink, determine whether the heat sink resonates; S5: When any judgment or evaluation result in steps S2 to S4 is abnormal, an early warning signal is generated.

2. The vibration monitoring method for oil-immersed transformers according to claim 1, characterized in that, In step S2, determining whether the winding has a risk of deformation or displacement includes: The axial pressure deformation of the winding is sensed in real time, and the displacement field data of the winding is acquired when the displacement reaches or exceeds the preset displacement threshold. A two-dimensional rectangular coordinate system is established with the winding axis as the x-direction and the radial direction as the y-direction. The displacement field is spatially differentiated, and the displacement gradient is calculated using the central difference method. The following relation is satisfied: ； in, This represents the displacement component of the displacement field in the x-direction. This indicates the optimized step size set based on the spacing of the insulating support bars, where x and y represent the axial and radial position coordinates of the winding in the two-dimensional rectangular coordinate system, respectively. Based on the displacement gradient Substituting the components into the Cauchy equation to synthesize the full-field strain components, we obtain the normal strain in the x-direction. y-direction normal strain and shear strain in the xy plane The following relation is satisfied: ； in, Indicates the displacement field at The displacement components in the direction are represented, and the total strain is expressed by the strain tensor matrix. The following relation is satisfied: ； Among them, shear strain The correction is made by the angle between the spiral arrangement direction of the fiber optic sensor and the principal stress of the winding; Calculate the von Mises equivalent strain The following relation is satisfied: ； in, This indicates that the von Mises effect changes, when When the strain exceeds the preset strain threshold, a deformation warning is triggered. Curvature analysis is performed on the displacement field to identify regions where the second derivative of the displacement changes abruptly, and the curvature is... Satisfying the relation: ； in, Indicates curvature. This represents the displacement component of the displacement field in the vertical direction, when When the curvature exceeds the preset curvature threshold, the region where the second derivative of the displacement changes abruptly is marked as a deformation risk point.

3. The vibration monitoring method for oil-immersed transformers according to claim 1, characterized in that, Based on the vibration energy data between the stacked plates, determining whether the clamping force of the stacked plates has attenuated includes: When the vibration energy entropy between the stacked pieces is greater than the preset vibration energy entropy threshold, it is determined that the clamping force between the stacked pieces is in a decaying state.

4. The vibration monitoring method for oil-immersed transformers according to claim 1, characterized in that, Determining whether the heat sink resonates based on its vibration mode data includes: Modal responses within a preset high-frequency range are selected from vibration modal data; Analyzing the frequency domain characteristics of the modal response, when the vibration energy is concentrated in a narrow frequency band with a width less than a preset bandwidth threshold and the amplitude of the vibration energy in the narrow frequency band exceeds a preset multiple threshold, it is determined that there is a sudden increase in narrowband vibration energy. Verify whether the sudden increase in the narrowband vibration energy conforms to the preset mode shape characteristics. If so, determine that the heat sink has resonated.

5. The vibration monitoring method for oil-immersed transformers according to claim 1, characterized in that, Step S3 includes: Calculate the cumulative amount of cyclic plastic strain at the weld. ,when When the cumulative value exceeds a preset threshold, it is determined that there is a risk of fatigue cracks. Calculate the preload decay rate at the bolt. The following relation is satisfied: ; in, This indicates the initial preload of the bolt. This indicates the remaining preload measured at the current moment. Based on the preload decay rate, the bolt loosening level is determined.

6. The vibration monitoring method for oil-immersed transformers according to claim 1, characterized in that, In step S4, the wear condition of the insulation structure is evaluated based on the acoustic emission signal data generated by the oil flow impact, including: A CFD-FEM joint model of the enclosure-oil-insulation structure was established. The CFD-FEM joint model includes a fluid domain and a structural domain, wherein: The fluid domain is described by the transient incompressible Navier-Stokes equations, satisfying the following relationship: ； in, Indicates the density of the oil. Represents the oil velocity vector. Indicates time, Indicates oil pressure, Indicates the dynamic viscosity of the oil. Represents the gravitational acceleration vector. This represents the volume force transformed from the vibrational spectrum; The fluid domain uses the VOF model to track the oil-gas interface, satisfying the following relationship: ;in, This indicates the volume fraction of the oil phase, and ; The fluid domain adopts The turbulence model satisfies the following relationship: ； in, Represents turbulent kinetic energy. Indicates the turbulent specific dissipation rate. Represents the turbulence generation term. and Represents the constants of the turbulence model. Indicates turbulent eddy viscosity. Indicates the kinematic viscosity of a fluid; The structural domain is described by elastic dynamics equations to represent the insulation structure response, satisfying the following relationship: ； in, Indicates the density of structural materials. Represents the structural displacement vector. Represents the structural damping matrix. Represents the stiffness matrix of the insulation structure. This represents the force vector exerted by the fluid on the insulating structure. This indicates a vibration load that acts directly on the insulating structure; The constitutive model of the insulation structure is orthotropic and satisfies the following relation: ； in, Represents the components of the Cauchy stress tensor. The stiffness tensor represents the stiffness of the insulation structure. Represents the components of the strain tensor; Data exchange between the fluid domain and the structural domain is achieved through a bidirectional fluid-structure interaction (FSI) interface, where: The fluid-to-solid coupling is transmitted to the surface of the insulating structure through integrated fluid pressure and shear stress, satisfying the following relationship: ,in, This represents the fluid-structure interaction interface. Represents the unit normal vector of the interface; The solid-to-fluid coupling updates the fluid mesh boundary through the displacement of the insulating structure, satisfying the following relationship: And the ALE mesh deformation method is adopted. Indicates the velocity of the fluid; The measured vibration spectrum was preprocessed into volume force. and structural boundary loads Among them, volume force satisfy: ,in , Represents the acceleration power spectral density. This represents the phase randomization function.

7. The vibration monitoring method for oil-immersed transformers according to claim 6, characterized in that, Assessing the wear condition of the insulation structure includes: The signal components within a preset characteristic frequency band are extracted from the acoustic emission signal data using a CFD-FEM joint model, and the pulse count rate N and amplitude integral V of the acoustic emission signal within the characteristic frequency band are calculated. The wear level of the insulation structure is determined based on the combined characteristics of the pulse count rate N and the amplitude integral V. If N is less than the first count rate threshold and V is less than the first amplitude threshold, the wear level is determined to be safe; if N is greater than the second count rate threshold and V is greater than the second amplitude threshold, the wear level is determined to be high-risk; otherwise, the wear level of the insulation structure is determined to be warning. Among them, the second count rate threshold is greater than the first count rate threshold, and the second amplitude threshold is greater than the first amplitude threshold.

8. A vibration monitoring system for oil-immersed transformers, characterized in that, The oil-immersed transformer includes a casing, core columns, windings, and heat sinks; The winding is mounted on the iron core column, the iron core column is composed of multiple layers of laminated sheets stacked sequentially, the winding is provided with insulating support bars, and the heat sink is connected to the housing through an oil circuit interface and communicates with the oil inside the housing; The gaps in the insulating support bar are provided with fiber optic sensors embedded in a spiral structure for acquiring displacement data of the winding. The vibration monitoring system for oil-immersed transformers includes: The data acquisition module is used to acquire vibration energy data between the laminations, displacement data of the windings, vibration mode data of the heat sink, strain data of the weld seam of the housing, preload data of the bolts of the housing, and acoustic emission signal data generated by oil flow impact at the insulation structure. The first monitoring module is used to determine whether the clamping force of the laminations has decayed based on the vibration energy data between the laminations, and to determine whether there is a risk of deformation or displacement of the winding based on the strain data obtained by full-field strain analysis of the displacement data of the winding. The second monitoring module is used to predict the fatigue crack risk of the weld based on the strain data of the weld of the box body, and to determine whether the bolts are loose based on the preload data of the bolts of the box body. The third monitoring module is used to assess the wear state of the insulation structure based on the acoustic emission signal data generated by the oil flow impact, and to determine whether the heat sink resonates based on the vibration mode data of the heat sink. The early warning signal generation module is used to generate an early warning signal when any judgment result or evaluation result is in an abnormal state.

9. The vibration monitoring system for oil-immersed transformers according to claim 8, characterized in that, The first monitoring module is also used for: The axial pressure deformation of the winding is sensed in real time, and the displacement field data of the winding is acquired when the displacement reaches or exceeds the preset displacement threshold. A two-dimensional rectangular coordinate system is established with the winding axis as the x-direction and the radial direction as the y-direction. The displacement field is spatially differentiated, and the displacement gradient is calculated using the central difference method. The following relation is satisfied: ； in, This represents the displacement component of the displacement field in the x-direction. This indicates the optimized step size set based on the spacing of the insulating support bars, where x and y represent the axial and radial position coordinates of the winding in the two-dimensional rectangular coordinate system, respectively. Based on the displacement gradient Substituting the components into the Cauchy equation to synthesize the full-field strain components, we obtain the normal strain in the x-direction. y-direction normal strain and shear strain in the xy plane The following relation is satisfied: ； in, Indicates the displacement field at The displacement components in the direction are represented, and the total strain is expressed by the strain tensor matrix. The following relation is satisfied: ； Among them, shear strain The correction is made by the angle between the spiral arrangement direction of the fiber optic sensor and the principal stress of the winding; Calculate the von Mises equivalent strain The following relation is satisfied: ； in, This indicates that the von Mises effect changes, when When the strain exceeds the preset strain threshold, a deformation warning is triggered. Curvature analysis is performed on the displacement field to identify regions where the second derivative of the displacement changes abruptly, and the curvature is... Satisfying the relation: ； in, Indicates curvature. This represents the displacement component of the displacement field in the vertical direction, when When the curvature exceeds the preset curvature threshold, the region where the second derivative of the displacement changes abruptly is marked as a deformation risk point.

10. The vibration monitoring system for oil-immersed transformers according to claim 8, characterized in that, The third monitoring module is also used for: A CFD-FEM joint model of the enclosure-oil-insulation structure was established. The CFD-FEM joint model includes a fluid domain and a structural domain, wherein: The fluid domain is described by the transient incompressible Navier-Stokes equations, satisfying the following relationship: ； in, Indicates the density of the oil. Represents the oil velocity vector. Indicates time, Indicates oil pressure, Indicates the dynamic viscosity of the oil. Represents the gravitational acceleration vector. This represents the volume force transformed from the vibrational spectrum; The fluid domain uses the VOF model to track the oil-gas interface, satisfying the following relationship: ;in, This indicates the volume fraction of the oil phase, and ; The fluid domain adopts Turbulence model, satisfying the following relationship: ； in, Represents turbulent kinetic energy. Indicates the turbulent specific dissipation rate. Represents the turbulence generation term. and Represents the constants of the turbulence model. Indicates turbulent eddy viscosity. Indicates the kinematic viscosity of a fluid; The structural domain is described by elastic dynamics equations to represent the insulation structure response, satisfying the following relationship: ； in, Indicates the density of structural materials. Represents the structural displacement vector. Represents the structural damping matrix. Represents the stiffness matrix of the insulation structure. This represents the force vector exerted by the fluid on the insulating structure. This indicates a vibration load that acts directly on the insulating structure; The constitutive model of the insulation structure is orthotropic and satisfies the following relation: ； in, Represents the components of the Cauchy stress tensor. The stiffness tensor represents the stiffness of the insulation structure. Represents the components of the strain tensor; Data exchange between the fluid domain and the structural domain is achieved through a bidirectional fluid-structure interaction (FSI) interface, where: The fluid-to-solid coupling is transmitted to the surface of the insulating structure through integrated fluid pressure and shear stress, satisfying the following relationship: ,in, This represents the fluid-structure interaction interface. Represents the unit normal vector of the interface; The solid-to-fluid coupling updates the fluid mesh boundary through the displacement of the insulating structure, satisfying the following relationship: And the ALE mesh deformation method is adopted. Indicates the velocity of the fluid; The measured vibration spectrum was preprocessed into volume force. and structural boundary loads Among them, volume force satisfy: ,in , Represents the acceleration power spectral density. This represents the phase randomization function.

Images