A transformer core residual magnetism estimation method based on magnetic flux leakage detection

Abstract

The present application relates to the technical field of transformer, especially to a transformer core residual magnetism estimation method based on leakage magnetic detection. The method comprises the following steps: performing magnetic field scanning on the outer surface of the transformer oil tank, determining the edge and corner projection points of the core, and arranging measurement points at each corner projection point to form a symmetrical measurement array; collecting the magnetic field vector data at each measurement point in the symmetrical measurement array, extracting the normal magnetic field component, and performing fitting processing on the normal magnetic field component to obtain the leakage magnetic decay characteristic parameter; calculating the residual magnetism characteristic parameter based on the leakage magnetic decay characteristic parameter corresponding to each corner projection point, and determining the effectiveness of the residual magnetism characteristic parameter according to the fitting consistency of the leakage magnetic decay characteristic parameter. Through symmetrical modeling and consistency verification of the leakage magnetic decay law of multiple core corner regions, the present application realizes stable and quantitative estimation of the core residual magnetism.

Description

Technical Field

[0001] This invention relates to the field of transformer technology, and in particular to a method for estimating the residual magnetism of transformer cores based on leakage flux detection. Background Technology

[0002] During manufacturing, testing, and operation, transformer cores often retain a certain amount of residual magnetism due to DC bias, excitation impact, or improper closing operations. The presence of residual magnetism in the core can lead to problems such as increased inrush current, increased noise, and localized overheating when the transformer is put back into operation. In severe cases, it may even cause protection malfunctions or equipment damage. Therefore, detecting and evaluating the residual magnetism of the core before transformer maintenance or commissioning is a crucial technical step in ensuring the safe and stable operation of transformers.

[0003] Existing methods for detecting residual magnetism in iron cores mainly include: indirect judgment methods based on excitation current waveforms, demagnetization-reverse deduction methods by applying demagnetizing current to the windings, and methods that directly measure the magnetization state of the iron core in a disassembled state. These methods generally suffer from the following shortcomings: First, most methods require transformer de-energization, winding disconnection, or even unpacking, resulting in high on-site implementation costs and long cycles, making them unsuitable for the numerous in-service transformers in distribution networks; second, indirect judgment methods based on excitation current or voltage waveforms are significantly affected by load conditions, winding parameters, and operating conditions, making quantitative assessment difficult; third, existing methods typically focus on the overall magnetization state, making it difficult to identify the non-uniformity of residual magnetism distribution in different areas of the iron core. Summary of the Invention

[0004] Therefore, it is necessary for the present invention to provide a method for estimating the residual magnetism of transformer core based on leakage flux detection, so as to solve at least one of the above-mentioned technical problems.

[0005] To achieve the above objectives, a method for estimating the residual magnetism of a transformer core based on leakage flux detection includes the following steps:

[0006] Step S1: Perform a magnetic field scan on the outer surface of the transformer tank to determine the projection points of the iron core edge and corners, and arrange measurement points at each corner projection point to form a symmetrical measurement array;

[0007] Step S2: Collect magnetic field vector data at each measurement point in the symmetrical measurement array, extract the normal magnetic field component, and perform fitting processing on the normal magnetic field component to obtain leakage magnetic field attenuation characteristic parameters.

[0008] Step S3: Calculate the residual magnetism characteristic parameters based on the leakage magnetic flux attenuation characteristic parameters corresponding to each corner projection point, and determine the validity of the residual magnetism characteristic parameters based on the fitting consistency of the leakage magnetic flux attenuation characteristic parameters.

[0009] Step S4: Substitute the valid residual magnetism characteristic parameters into the pre-established residual magnetism calibration relationship to determine the residual magnetism estimation value of each corner region, and perform weighted fusion of each residual magnetism estimation value to obtain the residual magnetism estimation result of the transformer core.

[0010] This invention utilizes the spatial attenuation law of the external leakage magnetic field of a transformer core to achieve indirect inversion and quantitative estimation of the residual magnetism state inside the core. This significantly improves the accuracy and engineering feasibility of residual magnetism assessment while ensuring safe equipment operation. The method first focuses on the core corner, the most sensitive area for magnetic flux leakage. By collecting the normal magnetic field component along the measurement baseline at the corner projection point, an attenuation model of the magnetic field as a function of distance is constructed. The corresponding attenuation index, amplitude parameter, and fitting residual are extracted from both sides of the corner, fully characterizing the intensity and spatial distribution of the leakage magnetic field. By performing logarithmic transformation and least-squares fitting on the magnetic field data, the influence of measurement noise and local anomalies is weakened, allowing the attenuation index to stably reflect the overall attenuation trend of the magnetic field, and the amplitude parameter to characterize the comprehensive magnetization level at the corner projection point. Simultaneously, the fitting residual is introduced as a quantitative indicator of data reliability, ensuring that subsequent calculations no longer rely solely on ideal model assumptions but fully consider the consistency between measured data and theoretical models.

[0011] Based on this, by analyzing the ratio of the attenuation indices on both sides of the corner, only when the magnetic field distribution satisfies the mirror symmetry characteristic is the leakage magnetic field at that location considered to reflect the remanence of the iron core. This effectively distinguishes the remanence information from interference caused by environmental stray magnetic fields, structural inhomogeneities, or sensor bias. Furthermore, through finite element static magnetic field simulation, a calibration relationship between the true value of remanence and leakage magnetic characteristic parameters is constructed under the condition of known iron core geometry and material parameters. This allows the measured characteristic parameters to be mapped to remanence estimates with clear physical meaning, avoiding the uncertainty caused by relying on empirical judgment or relative comparison in traditional methods. Finally, by using the reciprocal of the square of the fitting residual as the weight, the remanence estimation results of multiple effective corner projection points are weighted and fused, so that locations with high fitting consistency and strong stability dominate the final result, thus reducing the overall impact of single measurement point errors on the estimation results. In summary, this method achieves non-intrusive measurement at the data acquisition level, integrates physical models and statistical analysis at the feature construction level, and introduces a credibility weighting mechanism at the result fusion level. Thus, it provides a reliable, robust, and easily deployable technical path for assessing the residual magnetism of transformer cores, which helps reduce the risk of inrush current and improve the safety and controllability of equipment operation. Attached Figure Description

[0012] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a schematic diagram of the steps in the transformer core residual magnetism estimation method based on leakage flux detection of the present invention; Figure 2 This is a schematic diagram showing the correspondence between the measurement area on the outer surface of the transformer tank and the core space according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the remanence calibration relationship curve according to an embodiment of the present invention. Detailed Implementation

[0013] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0014] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0015] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0016] To achieve the above objectives, please refer to Figures 1 to 3 This invention provides a method for estimating the residual magnetism of a transformer core based on leakage flux detection. The method includes the following steps:

[0017] Step S1: Perform a magnetic field scan on the outer surface of the transformer tank to determine the projection points of the iron core edge and corners, and arrange measurement points at each corner projection point to form a symmetrical measurement array;

[0018] Step S2: Collect magnetic field vector data at each measurement point in the symmetrical measurement array, extract the normal magnetic field component, and perform fitting processing on the normal magnetic field component to obtain leakage magnetic field attenuation characteristic parameters.

[0019] Step S3: Calculate the residual magnetism characteristic parameters based on the leakage magnetic flux attenuation characteristic parameters corresponding to each corner projection point, and determine the validity of the residual magnetism characteristic parameters based on the fitting consistency of the leakage magnetic flux attenuation characteristic parameters.

[0020] Step S4: Substitute the valid residual magnetism characteristic parameters into the pre-established residual magnetism calibration relationship to determine the residual magnetism estimation value of each corner region, and perform weighted fusion of each residual magnetism estimation value to obtain the residual magnetism estimation result of the transformer core.

[0021] Furthermore, step S1 includes the following steps:

[0022] Step S11: Use a single-axis magnetic field sensor to move along the surface of the oil tank and record the normal magnetic field components and vector magnitude at each position;

[0023] In one embodiment, when the transformer is de-energized, the operator places a single-axis magnetic field sensor close to the outer surface of the transformer tank, ensuring that the sensor's sensitive axis is always aligned with the normal direction of the tank surface. The sensor is then moved and scanned continuously or at equal intervals along the tank surface using manual or automatic sliding rails. During the scanning process, the acquisition system synchronously records the normal magnetic field component value and the corresponding magnetic field vector magnitude at each measurement location at a preset sampling frequency, thereby forming a spatial sequence of magnetic field data distributed along the tank surface.

[0024] For example, a linear scanning method with a step interval of 5mm can be used to move the sensor point by point along the horizontal direction of the oil tank surface. At each position, the normal magnetic field component and vector magnitude of the current point are collected and stored for subsequent edge feature recognition.

[0025] Step S12: When the sign of the normal magnetic field component of adjacent measurement points is reversed and the vector magnitude value has a local maximum, mark the position as the positioning point, and connect multiple positioning points to form the iron core edge projection line;

[0026] In one embodiment, the sign change of the normal magnetic field component at adjacent measurement points is detected based on continuous measurement data. When the sign reversal phenomenon of the normal component changing from positive to negative or from negative to positive is detected, the magnetic field vector magnitude at the corresponding position is combined with whether it exhibits a local maximum feature relative to the measurement points before and after. If both conditions are met, the measurement position is determined as a location point with significant core leakage magnetic characteristics. By spatially connecting multiple location points on the same scanning path, a continuous core edge projection line is formed to characterize the projection profile of the core on the outer surface of the tank.

[0027] For example, when processing data from a scanning path, if the normal magnetic field component of a certain measurement point changes abruptly from a positive value to a negative value, and the magnitude of the magnetic field vector corresponding to that point is significantly higher than that of the two sampling points before and after, then that point is marked as a positioning point, and a line is drawn between it and the positioning points on the adjacent path to gradually outline the projected shape of the iron core edge.

[0028] Step S13: Determine the corner projection point at the intersection of the projection line of the iron core edge with the outline of the iron core column and the outline of the iron yoke;

[0029] In one embodiment, after obtaining the projection line of the iron core edge, the projection line is spatially superimposed with the known or pre-calibrated iron core column outline and iron yoke outline. By calculating the intersection point, the corner projection point of the iron core column and iron yoke at the connection point on the outer surface of the oil tank is determined. The corner projection point is used to characterize the area where the magnetic flux turning point inside the iron core is most concentrated and is the reference point for the subsequent leakage magnetic flux symmetry measurement array arrangement.

[0030] For example, when the projection line of the iron core edge intersects the longitudinal iron core column outline and the transverse iron yoke outline at a certain position, the coordinates of the intersection point are recorded as a corner projection point and used as the center reference point of the subsequent measurement baseline.

[0031] Step S14: With the corner projection point as the center, establish a first measurement baseline along the normal direction perpendicular to the surface of the oil tank and pointing into the iron core. With the iron core edge projection line as the symmetry reference, establish a second measurement baseline parallel to the first measurement baseline at the mirror symmetric position of the corner projection point to form a symmetrical measurement array.

[0032] In one embodiment, the corner projection point is used as the center point of the symmetrical measurement array. First, the direction of the axis of symmetry is determined according to the tangential direction of the iron core edge projection line at that point, and the tangential direction is used as the symmetrical constraint reference of the measurement array. On this basis, a first measurement baseline and a second measurement baseline located on both sides of the axis of symmetry are established around the corner projection point, so that the two measurement baselines satisfy the mirror symmetry relationship in spatial position, direction and measurement point distribution, thereby constructing a symmetrical measurement array for leakage magnetic symmetry verification.

[0033] For example, the local tangent direction of the projection line of the iron core edge can be calculated at the corner projection point, and two measurement baselines can be arranged on both sides of the same direction with the same spacing and direction as the axis of symmetry, so as to provide a unified geometric reference for the subsequent placement of magnetic field sensors at the corresponding positions.

[0034] It should be noted that the core purpose of the established symmetrical measurement array is to provide a physical basis for the subsequent determination of the consistency of the attenuation index. Therefore, the direction of the symmetry axis is uniquely determined by the tangential direction of the projection line of the iron core edge, so as to avoid misjudgment of symmetry due to subjective selection of direction.

[0035] See Figure 2 This diagram illustrates the correspondence between the measurement area on the outer surface of the transformer tank and the internal core spatial structure. The core columns and yoke constitute the core edge, and their corners form corresponding corner projection points on the outer surface of the tank. The measurement area is located at these corner projection points. The transformer core consists of multiple core columns and a yoke located above them. The core is entirely encapsulated inside the tank, and the outer surface of the tank serves as the accessible measurement interface for leakage magnetic signals. In this embodiment, the measurement area corresponding spatially to the corner projection points of each core column and yoke is first determined on the outer surface of the tank based on the geometric arrangement of the core within the tank. The measurement area is defined with the corner at the core column-yoke connection as the core projection position, expanding along the outer surface of the tank to form several local measurement windows. The dashed box in the diagram represents the projection range of the measurement area on the outer surface of the tank, and the red markers indicate the corresponding points of the corner projection points on the outer surface of the tank. By establishing this correspondence, the leakage flux measurement locations on the outer surface of the oil tank can be correlated one-to-one with the key magnetic flux concentration areas inside the iron core without opening the oil tank or contacting the iron core itself. This provides a clear spatial basis for the subsequent deployment of symmetrical measurement arrays, extraction of leakage flux attenuation characteristics, and inversion of remanent magnetization parameters. It should be noted that this spatial correspondence does not require a strict proportional mapping between the outer surface of the oil tank and the geometric contour of the iron core. Instead, it is a functional correspondence based on the dominant influence of the iron core structure on the leakage flux distribution, to ensure that the leakage flux signal near the corner projection point has a stable physical source.

[0036] Furthermore, step S14 includes the following:

[0037] With the corner projection point as the center, establish the first measurement baseline along the normal direction perpendicular to the surface of the oil tank and pointing into the iron core. On the first measurement baseline, set three measurement points from the surface of the oil tank outward at geometric intervals.

[0038] In one embodiment, the corner projection point is used as the spatial reference origin. Based on the geometric normal direction of the outer surface of the transformer tank, a first measurement baseline is established from the surface of the tank to the inside of the core, so that the measurement baseline is perpendicular to the surface of the tank. Subsequently, three measurement points are set on the first measurement baseline in order from near to far according to a preset geometric series spacing. The distances of each measurement point to the surface of the tank are in a fixed ratio relationship, so that the measurement points cover the attenuation region of leakage magnetic flux from strong to weak in space.

[0039] For example, the measurement point closest to the surface of the oil tank can be set at a distance of 10 mm from the surface of the oil tank, and the next two measurement points can be set at 20 mm and 40 mm respectively, so that the distance between the measurement points increases by a factor of 2, so as to take into account the acquisition of near-field and far-field leakage magnetic characteristics with a limited number of measurement points.

[0040] Determine the center line of the projection line of the iron core edge, and establish a second measurement baseline at the mirror symmetrical position of the corner projection point with the center line as the axis of symmetry. Set three measurement points on the second measurement baseline with the same spacing as the first measurement baseline.

[0041] In one embodiment, based on the determined position of the first measurement baseline, a geometric analysis is performed on the projection line of the iron core edge to determine the direction of its center line near the corner projection point, and this center line is used as the axis of symmetry. Subsequently, using this axis of symmetry as a mirror reference, a second measurement baseline is established at the symmetrical position of the corner projection point, so that the second measurement baseline is completely consistent with the first measurement baseline in terms of direction, length, and relative position to the surface of the tank. On the second measurement baseline, three measurement points are set out sequentially from the surface of the tank outward according to the same geometric series spacing as the first measurement baseline, thereby ensuring that the measurement points on the two measurement baselines form a one-to-one mirror relationship in spatial distribution.

[0042] For example, when the first measurement baseline is located on one side of the center line of the projection line of the iron core edge, a second measurement baseline can be generated on the other side by mirroring its spatial position relative to the center line, and three corresponding measurement points can be arranged at the same height position so that the two sets of measurement points are completely consistent in terms of the distance from the surface of the oil tank.

[0043] A symmetrical measurement array is formed by placing axially aligned triaxial magnetic field sensors at six measurement points.

[0044] In one embodiment, triaxial magnetic field sensors are installed at six measurement points determined by the first and second measurement baselines. During installation, the coordinate axes of each sensor are aligned so that one axis of each sensor is aligned with the normal direction of the oil tank surface, while the other two axes are distributed along the tangential direction of the oil tank surface. Through the above axial alignment, the magnetic field vector data collected from different measurement points have a consistent direction reference, thereby forming a symmetrical measurement array in which both the spatial structure and measurement direction satisfy a mirror symmetry relationship.

[0045] For example, the sensor can be calibrated using a calibration fixture before installation so that its Z-axis points uniformly to the normal direction of the outer surface of the oil tank. After calibration, it can be fixedly installed at the six measurement points in sequence to form a measurement array with consistent structure.

[0046] Of particular importance is the specific method of placing axially aligned triaxial magnetic field sensors at six measurement points:

[0047] Place all triaxial magnetic field sensors in an open area away from the transformer to measure the geomagnetic field vector and record the direction of the geomagnetic field vector measured by each sensor;

[0048] Using the measurement results of the first sensor as a reference, calculate the angle between the geomagnetic field direction measured by other sensors and the reference direction;

[0049] Adjust the physical orientation of each sensor so that its three measurement axes are parallel to the corresponding axes of the reference sensor, until the angle between the directions of the geomagnetic field measured by all sensors is less than one degree; mark uniform direction indicator lines on the housing of each sensor.

[0050] When arranging sensors at six measurement points, ensure that the direction indicator lines of all sensors point in the same direction to form a symmetrical measurement array.

[0051] Furthermore, step S2 includes the following steps:

[0052] Step S21: Synchronously trigger the six triaxial magnetic field sensors in the symmetrical measurement array to acquire magnetic field vector data, wherein the magnetic field vector data includes magnetic field components in three orthogonal directions output by each sensor;

[0053] In one embodiment, after the arrangement of the symmetrical measurement array and the axial alignment of the triaxial magnetic field sensors are completed, the six triaxial magnetic field sensors in the symmetrical measurement array are synchronously triggered by a unified trigger control module, so that the six sensors start to collect magnetic field data at the same time. During the acquisition process, each triaxial magnetic field sensor outputs magnetic field component data along its own three orthogonal coordinate axes, thereby obtaining complete magnetic field vector information at the six measurement points on the same time section.

[0054] For example, a main control acquisition unit can send sampling commands to six sensors simultaneously, enabling them to complete magnetic field data acquisition within the same sampling period, thus avoiding the impact of external magnetic field fluctuations caused by time asynchrony on the measurement results.

[0055] It should be noted that the purpose of synchronous triggering is to ensure that the magnetic field data on both sides of the symmetrical measurement array have temporal consistency, thereby providing a reliable data foundation for subsequent attenuation characteristic analysis based on symmetry.

[0056] Step S22: Extract the component consistent with the normal direction from the magnetic field vector data based on the normal direction of the oil tank surface as the normal magnetic field component of the measurement point;

[0057] In one embodiment, after obtaining the magnetic field vector data corresponding to each measurement point, the component that is consistent with the normal direction of the oil tank surface is extracted from the three orthogonal magnetic field components output by each triaxial magnetic field sensor according to the known normal direction of the outer surface of the oil tank, and is used as the normal magnetic field component of the measurement point. This extraction process can be directly selected through the coordinate axis correspondence or completed through coordinate transformation to ensure that the normal magnetic field components of different measurement points have a unified direction definition.

[0058] For example, when a certain axis of the sensor is set to be aligned with the normal of the oil tank surface during installation and calibration, the magnetic field component corresponding to that axis can be directly read as the normal magnetic field component.

[0059] Step S23: Record the normal magnetic field components of the three measurement points on the first measurement baseline as the first set of normal components, and record the normal magnetic field components of the three measurement points on the second measurement baseline as the second set of normal components;

[0060] In one embodiment, the normal magnetic field components of the six extracted measurement points are grouped according to their spatial relationship: the normal magnetic field components corresponding to the three measurement points located on the first measurement baseline are uniformly recorded as the first group of normal components, and the normal magnetic field components corresponding to the three measurement points located on the second measurement baseline are uniformly recorded as the second group of normal components; through this grouping method, the two groups of normal components correspond to the leakage magnetic attenuation data on both sides of the symmetrical measurement array.

[0061] For example, the data can be labeled according to the measurement baseline number, with the measurement point data on the first measurement baseline labeled as group one and the measurement point data on the second measurement baseline labeled as group two, so that they can be called respectively in subsequent fitting processing.

[0062] Step S24: Perform function fitting on the first group of normal components and the second group of normal components respectively to obtain the leakage magnetic flux attenuation characteristic parameters.

[0063] In one embodiment, the first set of normal components and the second set of normal components are used as input data, and combined with the spatial distance information from each measurement point to the surface of the oil tank, the two sets of normal components are subjected to function fitting processing to characterize the attenuation law of leakage magnetic intensity as a function of spatial distance, and the corresponding leakage magnetic attenuation characteristic parameters are extracted accordingly. The leakage magnetic attenuation characteristic parameters are used to characterize the leakage magnetic amplitude level and attenuation trend on each side of the measurement baseline, providing a basic input for the subsequent calculation of residual magnetic characteristic parameters.

[0064] For example, the data from the three measurement points corresponding to the first set of normal components can be fitted to obtain a set of parameters describing the leakage magnetic flux attenuation trend. Then, the second set of normal components can be fitted independently in the same way to form a pair of leakage magnetic flux attenuation characteristic parameters.

[0065] Furthermore, the leakage magnetic flux attenuation characteristic parameters include a first amplitude parameter, a second amplitude parameter, a first attenuation index, a second attenuation index, a first fitting residual, and a second fitting residual. Step S24 includes the following steps:

[0066] Step S241: Extract the distance from each measurement point on the first measurement baseline to the surface of the fuel tank, and calculate the common logarithm of the distance to obtain the logarithmic value of the distance;

[0067] In one embodiment, after the spatial arrangement of three measurement points on the first measurement baseline is completed, the vertical distance value of each measurement point relative to the outer surface of the oil tank is read, and the distance value is used as an independent variable to describe the spatial position of the measurement point; then, the common logarithm of each distance value is calculated to convert the original distance data into a distance logarithmic value, which is used for subsequent linear fitting processing between the distance logarithmic value and the magnetic field logarithmic value.

[0068] For example, if the distances from three measurement points on the first measurement baseline to the surface of the fuel tank are 10mm, 20mm and 40mm respectively, then their common logarithmic values ​​can be calculated to form a set of three logarithmic distance data points.

[0069] It should be noted that the common method of taking the logarithm of the distance is to transform the power-law decay relationship of leakage magnetic flux with distance into a linear form, which facilitates subsequent stable fitting using the linear least squares method.

[0070] Step S242: Extract the magnetic field components of each measurement point from the first group of normal components, calculate the common logarithm of the absolute value of the magnetic field components, and obtain the magnetic field logarithm value.

[0071] In one embodiment, the normal magnetic field component values ​​corresponding to the three measurement points on the first measurement baseline are extracted from the first group of normal components, and the absolute value of each magnetic field component is taken to eliminate the influence of magnetic field direction difference on amplitude analysis; then, the common logarithm is calculated for the obtained absolute values ​​of magnetic field components to obtain magnetic field logarithmic data that correspond one-to-one with the distance logarithmic values ​​in step S241.

[0072] For example, if the normal magnetic field components at the three measurement points are -5μT, -2μT and -0.8μT respectively, their absolute values ​​are taken first, and then the corresponding common logarithms are calculated to form three sets of magnetic field logarithmic data.

[0073] Step S243: Use the least squares method to perform linear fitting on the logarithmic values ​​of distance and magnetic field, and solve for the slope and intercept of the fitted line; take the opposite of the slope to obtain the first attenuation exponent, and perform exponential operation with the intercept to the base of 10 to obtain the first amplitude parameter.

[0074] In one embodiment, the logarithmic value of the distance is used as the independent variable and the logarithmic value of the magnetic field is used as the dependent variable. The least squares method is used to fit a straight line to obtain the slope and intercept of the fitted line describing the linear relationship between the two. After obtaining the fitting result, the opposite of the slope of the fitted line is taken as the first attenuation exponent to characterize the attenuation rate of the leakage magnetic field with the distance. At the same time, the intercept of the fitted line is subjected to an exponential operation with the base 10 to obtain the first amplitude parameter to characterize the amplitude level of the leakage magnetic field at the reference distance.

[0075] For example, after completing the linear fitting of three sets of data points, if the obtained fitting slope is -2.3, take its opposite to obtain the first decay exponent of 2.3, and perform an exponential operation on the fitting intercept to obtain the first amplitude parameter.

[0076] Furthermore, step S24 also includes the following steps:

[0077] Step S244: Divide the first amplitude parameter by the first attenuation exponent power of the distance from each measurement point to the surface of the oil tank to obtain the theoretical magnetic field value of each measurement point;

[0078] In one embodiment, after obtaining the first amplitude parameter and the first attenuation index, the first amplitude parameter is divided by the first attenuation index power of the distance from each measurement point to the surface of the oil tank according to a preset power function relationship, and the theoretical magnetic field value corresponding to each measurement point is calculated; the theoretical magnetic field value is used to describe the leakage magnetic field intensity level that should be presented at each measurement point under the ideal attenuation model.

[0079] For example, the first amplitude parameter can be substituted into the formula to calculate the theoretical magnetic field values ​​corresponding to the three measurement distances, forming a set of theoretical data that corresponds one-to-one with the measured magnetic field components.

[0080] Step S245: Take the normal magnetic field component in the first group of normal components as the measured value, calculate the deviation between the measured value and the theoretical magnetic field value at three measurement points, and calculate the standard deviation of the three deviation values ​​as the first fitting residual.

[0081] In one embodiment, the normal magnetic field components of three measurement points in the first group of normal components are taken as measured values ​​and compared with the corresponding theoretical magnetic field values ​​respectively. The deviation between the measured value and the theoretical value of each measurement point is calculated. Then, the standard deviation of the three deviation values ​​is calculated and the standard deviation is used as the first fitting residual to characterize the degree of consistency between the leakage magnetic field attenuation model of the first measurement baseline and the measured data.

[0082] For example, if the deviation values ​​of the three measurement points are relatively concentrated, the calculated standard deviation is small, and the corresponding first fitting residual is low, indicating that the fitting result is relatively reliable.

[0083] Step S246: Perform the calculations from steps S241 to S245 on the second group of normal components to obtain the second amplitude parameter, the second attenuation index, and the second fitting residual.

[0084] In one embodiment, the same processing flow as described above is used to perform distance logarithmic calculation, magnetic field logarithmic calculation, linear least squares fitting, parameter inversion, and fitting residual calculation on the second set of normal components, thereby obtaining the second amplitude parameter, the second attenuation index, and the second fitting residual corresponding to the second measurement baseline; the parameters are consistent with the first set of parameters in terms of physical meaning and calculation method, and are used for subsequent symmetry verification and comprehensive calculation of remanent magnetization characteristic parameters.

[0085] For example, the data from three measurement points on the second measurement baseline can be used as independent inputs, and the above fitting and residual calculation process can be repeated to form parameter pairs that correspond one-to-one with the first measurement baseline.

[0086] It should be noted that calculating the leakage magnetic flux attenuation characteristic parameters independently for the two sets of normal components is a prerequisite for subsequent mirror symmetry verification and an important basis for distinguishing environmental interference from the actual remanent magnetization effect.

[0087] Furthermore, in step S3, the calculation of the residual magnetism characteristic parameters based on the leakage magnetic flux attenuation characteristic parameters corresponding to each corner projection point includes:

[0088] The correction amplitude is obtained by multiplying the absolute values ​​of the first amplitude parameter and the second amplitude parameter and taking the square root.

[0089] In one embodiment, after calculating the leakage magnetic flux attenuation characteristic parameters corresponding to the first and second measurement baselines, the first amplitude parameter and the second amplitude parameter are read respectively, and their absolute values ​​are taken to eliminate the sign effect caused by the difference in magnetic field direction; then, the absolute value of the first amplitude parameter is multiplied by the absolute value of the second amplitude parameter, and the square root operation is performed on the product to obtain the corrected amplitude, which is used to characterize the comprehensive level of leakage magnetic flux amplitude information on both sides of the corner projection point.

[0090] For example, if the first amplitude parameter and the second amplitude parameter are respectively - and- First, take its absolute value, then calculate. This serves as the correction amplitude corresponding to the projection point at that corner.

[0091] The average decay index is obtained by calculating the arithmetic mean of the first decay index and the second decay index.

[0092] In one embodiment, after obtaining the first attenuation index and the second attenuation index, the two are arithmetically averaged to obtain the average attenuation index, which reflects the overall attenuation characteristics of leakage magnetic flux on both sides of the corner projection point as a function of distance; the average attenuation index is a key index term in the subsequent calculation of residual magnetism characteristic parameters, used to describe the influence of leakage magnetic flux attenuation rate on the magnitude of residual magnetism.

[0093] For example, if the first decay exponent is The second decay index is Then calculate This serves as the average attenuation index for the current corner projection point.

[0094] It should be noted that using the arithmetic mean instead of selecting a single-sided attenuation index can effectively reduce the bias of local interference or installation errors in the judgment of attenuation characteristics.

[0095] The preset empirical coefficient is 1.3-1.7. The empirical coefficient of the average attenuation index is calculated to the power of the empirical coefficient. The correction amplitude is divided by the empirical coefficient of the average attenuation index to obtain the remanent magnetic characteristic parameters of the current corner projection point.

[0096] In one embodiment, after obtaining the correction amplitude and the average attenuation index, an empirical coefficient is preset, with its value range limited to between 1.3 and 1.7; subsequently, the empirical coefficient is raised to a power of the average attenuation index, that is, the empirical coefficient is used as the base and the average attenuation index is used as the exponent for power operation, and the correction amplitude is divided by the power operation result to obtain the remanent magnetization characteristic parameter corresponding to the current corner projection point; the remanent magnetization characteristic parameter is used to characterize the relative level of the remanent magnetization intensity of the iron core in the corner region.

[0097] For example, when the empirical coefficient is chosen to be 1.5, the average decay exponent is... When, can be calculated Then, the correction amplitude is divided by this value to obtain the corresponding remanence characteristic parameters.

[0098] It should be noted that the introduction of empirical coefficients is used to adjust the influence of the decay index on the remanence inversion results. The range of values ​​is determined through experiments and simulations to ensure that the remanence characteristic parameters of different corner projection points are comparable without changing the way the subsequent calibration relationship is constructed.

[0099] The physical basis for calculating the geometric mean of the two amplitude parameters as the correction amplitude is as follows:

[0100] The first amplitude parameter reflects the leakage magnetic field strength at the first measurement baseline position, and the second amplitude parameter reflects the leakage magnetic field strength at the second measurement baseline position. Since the two measurement baselines are mirror-symmetric about the edge of the iron core, theoretically, the leakage magnetic field generated by the remanent magnetization of the iron core should satisfy the mirror inversion relationship where the absolute values ​​of the two amplitude parameters are equal but opposite in sign. The uniform interference magnetic field in the environment is approximately uniformly distributed within the measurement scale because the distance from the source is much greater than the distance between the sensors, and it does not satisfy the mirror inversion relationship, which will cause the two amplitude parameters to deviate from the theoretical relationship. The geometric mean is used to calculate the square root of the product of the absolute values ​​of the two amplitude parameters, rather than a simple arithmetic mean. This utilizes the mathematical property that the product operation has the effect of mutual cancellation of environmental interference components that do not satisfy mirror symmetry, thereby achieving physical suppression of environmental interference at the computational level and extracting the correction amplitude mainly contributed by the remanent magnetization.

[0101] This also includes obtaining remanence characteristic parameters by dividing the correction amplitude by a power of the average decay exponent, including:

[0102] The arithmetic mean of the first and second attenuation indices is calculated to obtain the average attenuation index, which reflects the rate at which the leakage magnetic field decays with distance. When the average attenuation index is close to two, it indicates that the spatial distribution of the magnetic field source is close to the surface magnetic charge characteristic. When the average attenuation index is close to three, it indicates that the magnetic field source is close to the magnetic dipole characteristic. An empirical coefficient between 1.3 and 1.7 is selected, and the empirical coefficient of the average attenuation index is raised to the power of the empirical coefficient. The correction amplitude is divided by the power of the result to obtain the remanent magnetization characteristic parameter. This operation realizes the normalization processing for different measurement distances and different spatial distribution characteristics of magnetic field sources, so that the remanent magnetization characteristic parameter mainly reflects the magnitude of the remanence of the iron core, and is not sensitive to the specific selection of the measurement position and the spatial difference of the tank shielding effect.

[0103] Furthermore, step S3, which determines the validity of the remanent magnetization characteristic parameters based on the fitting consistency of the leakage flux attenuation characteristic parameters, includes:

[0104] The ratio of the second decay index to the first decay index is used as a symmetry verification parameter.

[0105] In one embodiment, after calculating the leakage magnetic flux attenuation characteristic parameters corresponding to the first and second measurement baselines, the first attenuation index and the second attenuation index are read respectively, and the ratio of the second attenuation index to the first attenuation index is calculated to obtain the symmetry verification parameter. The symmetry verification parameter is used to quantitatively describe the consistency between the leakage magnetic flux attenuation characteristics on both sides of the corner projection point, thereby providing a quantitative basis for subsequent validity determination.

[0106] For example, if the first decay exponent is The second decay index is Then calculate This serves as a symmetry verification parameter for the projection point at the corner.

[0107] It should be noted that using the ratio of the attenuation index rather than the difference for symmetry evaluation can avoid the impact of the absolute magnitude difference of the attenuation index at different corner projection points on the consistency judgment.

[0108] When the symmetry verification parameters fall within the preset range, it indicates that the magnetic field distribution of the corresponding corner projection point satisfies the mirror symmetry and the environmental interference is small, and the residual magnetism characteristic parameters of the corner projection point are determined to be valid.

[0109] In one embodiment, the calculated symmetry verification parameters are compared with a pre-set validity judgment range. When the symmetry verification parameters fall within the preset range, it is determined that the leakage magnetic flux attenuation characteristic parameters corresponding to the measurement baselines on both sides of the corner projection point have good consistency in the attenuation trend, indicating that the magnetic field distribution at this location satisfies the mirror symmetry assumption, and the external environmental magnetic interference or local anomalies have little impact on the measurement results. In this case, the residual magnetic flux characteristic parameters calculated based on the corner projection point are marked as valid and used for subsequent residual magnetic flux estimation and weighted fusion.

[0110] For example, when the symmetry verification parameter is close to 1 and within the preset allowable fluctuation range, the attenuation indexes on both sides can be considered to be basically the same, thus determining that the remanent magnetic characteristic parameter of the corner projection point is valid.

[0111] When the symmetry verification parameters are not within the preset range, the residual magnetic characteristic parameters of the corresponding corner projection point are determined to be invalid.

[0112] In one embodiment, when the symmetry verification parameter exceeds the preset validity judgment range, it is determined that there is a significant inconsistency between the leakage magnetic flux attenuation characteristic parameters on both sides of the corner projection point, indicating that the magnetic field distribution at this location fails to meet the mirror symmetry assumption and may be affected by factors such as environmental magnetic field interference, local structural anomalies, or unstable measurement conditions. In this case, the residual magnetic characteristic parameter corresponding to the corner projection point is marked as invalid and will not participate in the fusion calculation of the residual magnetic flux estimation results in subsequent processing.

[0113] For example, if the symmetry verification parameter deviates significantly from 1 and exceeds the upper or lower limit of the preset range, the data of the corner projection point will be directly determined as an invalid data point.

[0114] Of particular importance is that the preset range of the symmetry verification parameter in step S3 is set to 0.85-1.15, and the specific judgment rule is as follows:

[0115] When the symmetry verification parameter is greater than or equal to 0.85 and less than or equal to 1.15, the relative deviation between the second decay index and the first decay index is no more than 15. The residual magnetic characteristic parameters of the corner projection point are determined to be valid.

[0116] When the symmetry verification parameter is less than 0.85, the decay rate of the leakage magnetic field of the second measurement baseline is significantly faster than that of the first measurement baseline. It is determined that there is a local shielding asymmetry or an interference source near the second measurement baseline. The residual magnetic characteristic parameter of the corner projection point is invalid.

[0117] When the symmetry verification parameter is greater than 1.15, the decay rate of the leakage magnetic field of the first measurement baseline is significantly faster than that of the second measurement baseline. It is determined that there is a local shielding asymmetry or an interference source near the first measurement baseline, and the residual magnetic characteristic parameter of the corner projection point is invalid.

[0118] Furthermore, step S4 includes the following steps:

[0119] Step S41: Obtain the core dimensions and material parameters according to the transformer model, and establish a finite element model;

[0120] In one embodiment, based on the model information of the transformer to be tested, the geometric dimensions and material magnetic parameters of the core corresponding to the model are obtained. The geometric dimensions include the cross-sectional dimensions of the core column, the dimensions of the yoke, and the overall layout relationship. The material parameters include the magnetic permeability characteristics and saturation magnetization characteristics of the core material. Based on this, a three-dimensional finite element model corresponding to the actual transformer structure is established according to the obtained parameters for subsequent static magnetic field simulation analysis.

[0121] For example, the cross-sectional dimensions and material type of the iron core can be input based on the transformer's factory technical data or design drawings, and the corresponding iron core geometric model can be constructed in finite element simulation software.

[0122] Step S42: Set a uniform remanence of 0.1-1 Tesla in the iron core to perform static magnetic field simulation on the finite element model, extract the normal magnetic field distribution along the measurement baseline in the simulation results, and calculate the simulation remanence characteristic parameters.

[0123] In one embodiment, in the established finite element model, a uniformly distributed remanent magnetization condition is set for the entire iron core. The remanent magnetization intensity takes multiple discrete values ​​in the range of 0.1 Tesla to 1 Tesla, and static magnetic field simulation is performed under each remanent magnetization intensity condition. After the simulation is completed, the normal magnetic field distribution data along the position corresponding to the actual measurement baseline is extracted from the simulation results, and the corresponding simulated remanent magnetization characteristic parameters are calculated according to the processing flow that is completely consistent with the measured data.

[0124] For example, uniform remanence conditions of 0.2 Tesla, 0.5 Tesla and 0.8 Tesla can be set sequentially, static magnetic field simulations can be run respectively, and the normal magnetic field data in the measurement baseline direction can be extracted from the simulation results to calculate characteristic parameters.

[0125] Step S43: Establish data pairs between the true value of remanence and the simulated remanence characteristic parameters, and fit them to obtain the remanence calibration relationship;

[0126] In one embodiment, a set remanence intensity is used as the true remanence value, and the corresponding remanence characteristic parameters obtained from simulation calculation are used as feature inputs to construct a data pair between the true remanence value and the simulated remanence characteristic parameters. Subsequently, the data pair is fitted to establish a remanence calibration relationship that describes the mapping relationship between the true remanence value and the remanence characteristic parameters, which is used for the inversion calculation of subsequent measured data.

[0127] For example, data pairs obtained under multiple conditions of different remanence intensity can be used as samples to obtain the remanence calibration function through curve fitting.

[0128] Step S44: Substitute the effective remanence characteristic parameters into the remanence calibration relationship to calculate the corresponding estimated remanence value.

[0129] In one embodiment, for each corner projection point that is determined to be valid, the corresponding measured residual magnetism characteristic parameters are read, and the residual magnetism characteristic parameters are substituted into the established residual magnetism calibration relationship for calculation, thereby obtaining the residual magnetism estimate corresponding to the corner projection point; the residual magnetism estimate is used as an independent estimation result of the residual magnetism level of the iron core in the corner area.

[0130] For example, substitution calculations can be performed individually for each valid corner projection point to obtain multiple independent estimates of remanence.

[0131] See Figure 3 The horizontal axis represents the residual magnetism characteristic parameters obtained through simulation and actual measurement processes, and the vertical axis represents the set or known residual magnetism intensity value inside the iron core. In this embodiment, a finite element model of the iron core is first established based on a specific transformer model, and uniform residual magnetism conditions of different amplitudes are applied in the iron core to obtain a set of corresponding data points for "true residual magnetism value - simulated residual magnetism characteristic parameters," which are represented as discrete points in the figure. Subsequently, the above data points are fitted with a function to obtain a quantitative mapping relationship between the residual magnetism characteristic parameters and the actual residual magnetism. The solid line in the figure represents the calibration curve obtained by fitting, and the corresponding fitting expression and correlation coefficient are given to characterize the stability and reliability of the calibration relationship. In the actual measurement process, after calculating the measured residual magnetism characteristic parameters of a certain corner projection point based on the leakage magnetic field data of the outer surface of the oil tank, the characteristic parameters can be substituted into the calibration relationship curve along the inversion path shown in the figure to deduce the corresponding estimated value of the iron core residual magnetism. It should be noted that this calibration relationship is not a simple empirical curve, but is established by combining specific core material parameters, structural dimensions and measurement baseline arrangement. Therefore, it can effectively reduce the impact of structural differences between different transformer models on the residual magnetism estimation results and improve the consistency and repeatability of the method in engineering applications.

[0132] Furthermore, step S4 also includes the following steps:

[0133] Step S45: For the corner projection points corresponding to each effective remanent magnetization characteristic parameter, take the average of the first fitting residual and the second fitting residual as the comprehensive fitting residual at that position.

[0134] In one embodiment, for each corner projection point corresponding to a valid residual magnetic characteristic parameter, its first fitting residual and second fitting residual are read, and the two are arithmetically averaged to obtain the comprehensive fitting residual corresponding to the corner projection point; the comprehensive fitting residual is used to comprehensively characterize the overall consistency level of the leakage magnetic flux fitting results of the measurement baselines on both sides of the position.

[0135] For example, if the first and second fitting residuals of a certain corner projection point are close in value, then the overall fitting residual is relatively small.

[0136] Step S46: Set the weight coefficient of the corner projection point of each effective remanent magnetization characteristic parameter to the reciprocal of the square of its comprehensive fitting residual;

[0137] In one embodiment, a corresponding weight coefficient is calculated for each valid corner projection point based on the comprehensive fitting residual. Specifically, the reciprocal of the square of the comprehensive fitting residual is used as the weight coefficient of the corner projection point. Through this weight setting method, corner projection points with better fitting consistency occupy a higher weight in the final fusion.

[0138] For example, when the comprehensive fitting residual of a certain corner projection point is small, its corresponding weight coefficient is large, and it contributes more to the fusion calculation.

[0139] Step S47: Multiply all valid residual magnetism estimates by their corresponding weighting coefficients, sum them up, and then divide by the sum of all weighting coefficients to obtain the weighted and merged transformer core residual magnetism estimate.

[0140] In one embodiment, the residual magnetism estimation values ​​corresponding to all effective corner projection points are multiplied by their corresponding weight coefficients, and the weighted residual magnetism estimation values ​​are summed. Then, the summation result is divided by the sum of the weight coefficients of all effective corner projection points to obtain the weighted and fused residual magnetism estimation result of the transformer core. This result is used as the final output residual magnetism estimation value of the overall core.

[0141] For example, when multiple corner projection points pass the validity determination, a single global remanence estimation result can be obtained through the weighted average method described above.

[0142] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0143] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for estimating the residual magnetism of a transformer core based on leakage flux detection, characterized in that, Includes the following steps: Step S1: Perform a magnetic field scan on the outer surface of the transformer tank to determine the projection points of the iron core edge and corners, and arrange measurement points at each corner projection point to form a symmetrical measurement array; Step S2: Collect magnetic field vector data at each measurement point in the symmetrical measurement array, extract the normal magnetic field component, and perform fitting processing on the normal magnetic field component to obtain leakage magnetic field attenuation characteristic parameters. Step S3: Calculate the residual magnetism characteristic parameters based on the leakage magnetic flux attenuation characteristic parameters corresponding to each corner projection point, and determine the validity of the residual magnetism characteristic parameters based on the fitting consistency of the leakage magnetic flux attenuation characteristic parameters. Step S4: Substitute the valid residual magnetism characteristic parameters into the pre-established residual magnetism calibration relationship to determine the residual magnetism estimation value of each corner region, and perform weighted fusion of each residual magnetism estimation value to obtain the residual magnetism estimation result of the transformer core.

2. The method for estimating the residual magnetism of a transformer core based on leakage flux detection according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Use a single-axis magnetic field sensor to move along the surface of the oil tank and record the normal magnetic field components and vector magnitude at each position; Step S12: When the sign of the normal magnetic field component of adjacent measurement points is reversed and the vector magnitude value has a local maximum, mark the position as the positioning point, and connect multiple positioning points to form the iron core edge projection line; Step S13: Determine the corner projection point at the intersection of the projection line of the iron core edge with the outline of the iron core column and the outline of the iron yoke; Step S14: With the corner projection point as the center, establish a first measurement baseline along the normal direction perpendicular to the surface of the oil tank and pointing into the iron core. With the iron core edge projection line as the symmetry reference, establish a second measurement baseline parallel to the first measurement baseline at the mirror symmetric position of the corner projection point to form a symmetrical measurement array.

3. The method for estimating the residual magnetism of a transformer core based on leakage flux detection according to claim 2, characterized in that, Step S14 includes the following: With the corner projection point as the center, establish the first measurement baseline along the normal direction perpendicular to the surface of the oil tank and pointing into the iron core. On the first measurement baseline, set three measurement points from the surface of the oil tank outward at geometric intervals. Determine the center line of the projection line of the iron core edge, and establish a second measurement baseline at the mirror symmetrical position of the corner projection point with the center line as the axis of symmetry. Set three measurement points on the second measurement baseline with the same spacing as the first measurement baseline. A symmetrical measurement array is formed by placing axially aligned triaxial magnetic field sensors at six measurement points.

4. The method for estimating the residual magnetism of a transformer core based on leakage flux detection according to claim 3, characterized in that, Step S2 includes the following steps: Step S21: Synchronously trigger the six triaxial magnetic field sensors in the symmetrical measurement array to acquire magnetic field vector data, wherein the magnetic field vector data includes magnetic field components in three orthogonal directions output by each sensor; Step S22: Extract the component consistent with the normal direction from the magnetic field vector data based on the normal direction of the oil tank surface as the normal magnetic field component of the measurement point; Step S23: Record the normal magnetic field components of the three measurement points on the first measurement baseline as the first set of normal components, and record the normal magnetic field components of the three measurement points on the second measurement baseline as the second set of normal components; Step S24: Perform function fitting on the first group of normal components and the second group of normal components respectively to obtain the leakage magnetic flux attenuation characteristic parameters.

5. The method for estimating the residual magnetism of a transformer core based on leakage flux detection according to claim 4, wherein the leakage flux attenuation characteristic parameters include a first amplitude parameter, a second amplitude parameter, a first attenuation index, a second attenuation index, a first fitting residual, and a second fitting residual, characterized in that, Step S24 includes the following steps: Step S241: Extract the distance from each measurement point on the first measurement baseline to the surface of the fuel tank, and calculate the common logarithm of the distance to obtain the logarithmic value of the distance; Step S242: Extract the normal magnetic field components of each measurement point from the first group of normal components, calculate the common logarithm of the absolute value of the normal magnetic field components, and obtain the magnetic field logarithm value. Step S243: Use the least squares method to perform linear fitting on the logarithmic values ​​of distance and magnetic field, and solve for the slope and intercept of the fitted line; take the opposite of the slope to obtain the first attenuation exponent, and perform exponential operation with the intercept to the base of 10 to obtain the first amplitude parameter.

6. The method for estimating the residual magnetism of a transformer core based on leakage flux detection according to claim 5, characterized in that, Step S24 also includes the following steps: Step S244: Divide the first amplitude parameter by the first attenuation exponent power of the distance from each measurement point to the surface of the oil tank to obtain the theoretical magnetic field value of each measurement point; Step S245: Take the normal magnetic field component in the first group of normal components as the measured value, calculate the deviation between the measured value and the theoretical magnetic field value at three measurement points, and calculate the standard deviation of the three deviation values ​​as the first fitting residual. Step S246: Perform the calculations from steps S241 to S245 on the second group of normal components to obtain the second amplitude parameter, the second attenuation index, and the second fitting residual.

7. The method for estimating the residual magnetism of a transformer core based on leakage flux detection according to claim 6, characterized in that, Step S3, which calculates the remanent magnetization characteristic parameters based on the leakage magnetic flux attenuation characteristic parameters corresponding to each corner projection point, includes: The correction amplitude is obtained by multiplying the absolute values ​​of the first amplitude parameter and the second amplitude parameter and taking the square root. The average decay index is obtained by calculating the arithmetic mean of the first decay index and the second decay index. The preset empirical coefficient is 1.3-1.

7. The empirical coefficient of the average attenuation index is calculated to the power of the empirical coefficient. The correction amplitude is divided by the empirical coefficient of the average attenuation index to obtain the remanent magnetic characteristic parameters of the current corner projection point.

8. The method for estimating the residual magnetism of a transformer core based on leakage flux detection according to claim 7, characterized in that, Step S3, determining the validity of the remanent magnetization characteristic parameters based on the fitting consistency of the leakage flux attenuation characteristic parameters, includes: The ratio of the second decay index to the first decay index is used as a symmetry verification parameter. When the symmetry verification parameters fall within the preset range, it indicates that the magnetic field distribution of the corresponding corner projection point satisfies the mirror symmetry and the environmental interference is small, and the residual magnetism characteristic parameters of the corner projection point are determined to be valid. When the symmetry verification parameters are not within the preset range, the residual magnetic characteristic parameters of the corresponding corner projection point are determined to be invalid.

9. The method for estimating the residual magnetism of a transformer core based on leakage flux detection according to claim 8, characterized in that, Step S4 includes the following steps: Step S41: Obtain the core dimensions and material parameters according to the transformer model, and establish a finite element model; Step S42: Set a uniform remanence of 0.1-1 Tesla in the iron core to perform static magnetic field simulation on the finite element model, extract the normal magnetic field distribution along the measurement baseline in the simulation results, and calculate the simulation remanence characteristic parameters. Step S43: Establish data pairs between the true value of remanence and the simulated remanence characteristic parameters, and fit them to obtain the remanence calibration relationship; Step S44: Substitute the effective remanence characteristic parameters into the remanence calibration relationship to calculate the corresponding estimated remanence value.

10. The method for estimating the residual magnetism of a transformer core based on leakage flux detection according to claim 9, characterized in that, Step S4 also includes the following steps: Step S45: For the corner projection points corresponding to each effective remanent magnetization characteristic parameter, take the average of the first fitting residual and the second fitting residual as the comprehensive fitting residual at that position. Step S46: Set the weight coefficient of the corner projection point of each effective remanent magnetization characteristic parameter to the reciprocal of the square of its comprehensive fitting residual; Step S47: Multiply all valid residual magnetism estimates by their corresponding weighting coefficients, sum them up, and then divide by the sum of all weighting coefficients to obtain the weighted and merged transformer core residual magnetism estimate.

Images