Transformer magnetic flux leakage detection method and device and nonvolatile storage medium

Abstract

The invention discloses a transformer magnetic flux leakage detection method and device and a nonvolatile storage medium. The method comprises the steps that the position of a target measuring point on the oil tank wall of a target transformer is determined; at the position of the target measuring point, an original magnetic field signal is obtained through an atomic magnetometer, and the original magnetic field signal is a time sequence signal; based on the original magnetic field signal, a space-time magnetic graph cube is constructed, and the space-time magnetic graph cube represents space distribution information and time change information of a leakage magnetic field of the target transformer; and based on the space-time magnetic graph cube, determining a magnetic flux leakage detection result of the target transformer, the magnetic flux leakage detection result including a magnetic flux leakage fault reason and a magnetic flux leakage fault degree. The technical problem that early-stage accurate capturing and positioning of weak fault signals are difficult to achieve through a traditional transformer magnetic flux leakage detection technology is solved.

Description

Technical Field

[0001] This invention relates to the field of power equipment testing technology, and more specifically, to a method, apparatus, and non-volatile storage medium for detecting transformer leakage flux. Background Technology

[0002] As a critical component of the power system, monitoring the operating status of transformers is essential for the safety and stability of the power system. Under normal operating conditions, the magnetic field in the transformer core should be completely confined. However, when faults such as winding displacement, inter-turn short circuits, multi-point grounding of the core, or loosening of clamping components occur, the distribution of the magnetic field is affected, causing some of the magnetic field to leak to the outside of the transformer, forming what is known as "leakage magnetic field." Changes in the characteristics of the leakage magnetic field can directly reflect changes in the internal structure of the transformer; therefore, its detection has become an important means of assessing the health status of transformers.

[0003] However, existing magnetic flux leakage detection technologies mainly rely on Hall sensors or induction coils, which have significant limitations. Hall sensors typically have sensitivity only on the order of nT, insufficient for capturing early, subtle magnetic field changes caused by faults, making early warning difficult. Induction coils, limited by their insensitivity to DC and low-frequency signals, cannot acquire complete leakage magnetic field spectrum information, reducing the comprehensiveness of fault diagnosis. Furthermore, traditional electromagnetic sensor deployment methods are usually single-point or sparsely distributed, making it difficult to construct effective magnetic maps and limiting the ability to accurately locate faults. In complex electromagnetic environments, these sensors are also susceptible to interference, resulting in low signal-to-noise ratios and affecting detection accuracy.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This invention provides a method, apparatus, and non-volatile storage medium for detecting transformer leakage flux, thereby at least solving the technical problem that traditional transformer leakage flux detection technology struggles to achieve early and accurate capture and location of weak fault signals.

[0006] According to one aspect of the present invention, a method for detecting leakage magnetic flux in a transformer is provided, comprising: determining the location of a target measuring point on the tank wall of a target transformer; acquiring an original magnetic field signal at the target measuring point location using an atomic magnetometer, wherein the original magnetic field signal is a time-series signal; constructing a spatiotemporal magnetic map cube based on the original magnetic field signal, wherein the spatiotemporal magnetic map cube characterizes the spatial distribution and temporal variation information of the leakage magnetic field of the target transformer; and determining the leakage magnetic flux detection result of the target transformer based on the spatiotemporal magnetic map cube, wherein the leakage magnetic flux detection result includes the cause of the leakage magnetic flux fault and the degree of the leakage magnetic flux fault.

[0007] Optionally, determining the target measurement point location on the tank wall of the target transformer includes: constructing a three-dimensional rectangular coordinate system outside the tank wall, around the projection areas corresponding to the high-voltage winding, low-voltage winding, and core of the target transformer. The first axis of the three-dimensional rectangular coordinate system is parallel to the winding axis of the target transformer, the second axis is parallel to the winding radial direction of the target transformer, and the third axis is perpendicular to the surface of the tank wall. An electromagnetic simulation model of the target transformer is established in the three-dimensional rectangular coordinate system. An initial measurement point set is constructed, which is an empty set. Based on the electromagnetic simulation model, a target location is selected from all available locations and added to the initial measurement point set, where the target location is the one that minimizes the posterior uncertainty of the electromagnetic simulation model in its current state. Adding locations stops when the number of locations in the initial measurement point set exceeds a preset threshold, resulting in the final target measurement point set, which serves as the target measurement point location.

[0008] Optionally, the probe of the atomic magnetometer retains an opening only in the direction of the sensitive axis. The probe of the atomic magnetometer is connected to the central processing unit via optical fiber. The central processing unit is used to analyze and process the acquired raw magnetic field signals.

[0009] Optionally, based on the original magnetic field signal, a spatiotemporal magnetic map cube is constructed, including: acquiring electrical data of the target transformer, wherein the electrical data includes voltage data of the primary side of the target transformer, current data of the primary side of the target transformer, voltage data of the secondary side of the target transformer, and current data of the secondary side of the target transformer; denoising the original magnetic field signal, and synchronizing and aligning the original magnetic field signal based on the electrical data to obtain the processed initial magnetic field signal; generating a three-dimensional leakage magnetic field distribution cloud map corresponding to any one moment in the initial magnetic field signal, obtaining multiple three-dimensional leakage magnetic field distribution cloud maps corresponding to each moment; stacking the multiple three-dimensional leakage magnetic field distribution cloud maps corresponding to each moment in chronological order to form a spatiotemporal magnetic map cube.

[0010] Optionally, based on the spatiotemporal magnetic map cube, the leakage flux detection result of the target transformer is determined, including: extracting multiple leakage flux features from the spatiotemporal magnetic map cube; inputting the multiple leakage flux features into a preset deep learning model to obtain the leakage flux detection result, wherein the causes of leakage flux faults include normal, slight winding deformation, severe winding deformation or multi-point grounding of the iron core, and the degree of leakage flux faults includes mild, moderate or severe and the corresponding confidence probability.

[0011] Optionally, multiple leakage magnetic characteristics include spatial gradient, centroid, moment of inertia, power frequency fundamental frequency, harmonic amplitude, harmonic phase, and high-frequency noise energy.

[0012] According to another aspect of the present invention, a transformer leakage flux detection device is also provided, comprising: a first determining module for determining the location of a target measuring point on the tank wall of a target transformer; an acquiring module for acquiring an original magnetic field signal at the target measuring point location using an atomic magnetometer, wherein the original magnetic field signal is a time-series signal; a constructing module for constructing a spatiotemporal magnetic map cube based on the original magnetic field signal, wherein the spatiotemporal magnetic map cube characterizes the spatial distribution information and temporal variation information of the leakage magnetic field of the target transformer; and a second determining module for determining the leakage flux detection result of the target transformer based on the spatiotemporal magnetic map cube, wherein the leakage flux detection result includes the cause of the leakage flux fault and the degree of the leakage flux fault.

[0013] According to another aspect of the present invention, a non-volatile storage medium is also provided, the non-volatile storage medium including a stored program, wherein, when the program is running, it controls the device where the non-volatile storage medium is located to execute any of the above-described transformer leakage flux detection methods.

[0014] According to another aspect of the present invention, a computer device is also provided, the computer device including a processor, the processor being configured to run a program, wherein the program executes any of the above-described transformer leakage flux detection methods during execution.

[0015] According to another aspect of the present invention, a computer program product is also provided, including a computer program that, when executed by a processor, implements any of the above-described transformer leakage flux detection methods.

[0016] In this embodiment of the invention, a transformer leakage magnetic flux detection method is adopted. The target measurement point location on the tank wall of the target transformer is determined. At the target measurement point location, an atomic magnetometer is used to acquire the original magnetic field signal, which is a time-series signal. Based on the original magnetic field signal, a spatiotemporal magnetic map cube is constructed, which represents the spatial distribution and temporal variation information of the leakage magnetic field of the target transformer. Based on the spatiotemporal magnetic map cube, the leakage magnetic flux detection result of the target transformer is determined. The leakage magnetic flux detection result includes the cause and degree of leakage magnetic flux fault, achieving the purpose of intelligent detection and early warning of transformer leakage magnetic flux. This realizes the technical effect of precise monitoring of the internal state of the transformer and early fault warning, thereby solving the technical problem that traditional transformer leakage magnetic flux detection technology is difficult to achieve early and accurate capture and location of weak fault signals. Attached Figure Description

[0017] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0018] Figure 1 A hardware block diagram of a computer terminal for implementing a transformer leakage flux detection method is shown.

[0019] Figure 2 This is a flowchart illustrating the transformer leakage flux detection method provided in an embodiment of the present invention;

[0020] Figure 3 This is a diagram of a transformer leakage flux detection system architecture provided by an optional embodiment of the present invention;

[0021] Figure 4 This is a flowchart of transformer leakage flux detection implementation provided by an optional embodiment of the present invention;

[0022] Figure 5 This is a structural block diagram of a transformer leakage flux detection device provided according to an embodiment of the present invention. Detailed Implementation

[0023] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0024] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0025] According to an embodiment of the present invention, a method for detecting transformer leakage flux is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0026] The method embodiment provided in Embodiment 1 of this application can be executed on a mobile terminal, computer terminal, or similar computing device. Figure 1 A hardware block diagram of a computer terminal for implementing a transformer leakage flux detection method is shown. Figure 1 As shown, the computer terminal 10 may include one or more processors (shown as 102a, 102b, ..., 102n in the figure) (the processor may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0027] It should be noted that the aforementioned one or more processors and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10. As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0028] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the transformer leakage flux detection method in this embodiment of the invention. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby implementing the transformer leakage flux detection method of the aforementioned application program. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0029] The display can be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 10.

[0030] Figure 2This is a flowchart illustrating the transformer leakage flux detection method provided in an embodiment of the present invention, as shown below. Figure 2 As shown, the method includes the following steps:

[0031] Step S201: Determine the location of the target measuring point on the tank wall of the target transformer.

[0032] In this step, determining the target measurement point location on the tank wall of the target transformer refers to the first step in selecting and determining the specific location for placing the atomic magnetometer when implementing the transformer leakage magnetic field detection method. This series of selected locations constitutes a measurement point layout, which aims to optimize the detection effect of the transformer leakage magnetic field.

[0033] Specifically, a three-dimensional Cartesian coordinate system is constructed, with its X, Y, and Z axes corresponding to the transformer's major axis (winding axial direction), minor axis (winding radial direction), and the normal to the tank surface, respectively. Then, by arranging atomic magnetometer probes on each side of the tank's outer wall according to a pre-defined two-dimensional grid, a measurement point layout effectively covers the transformer's critical components. To determine the optimal target measurement point locations, this embodiment employs a mathematical optimization method. Starting from an initial empty set using a Bayesian framework, measurement points are iteratively added, each time selecting the location that minimizes posterior uncertainty. When the number of measurement points reaches a preset threshold (a given number of probes), the iteration stops, and the resulting set is the target measurement point set, which can be used to determine the target measurement point locations.

[0034] Step S202: At the target measurement point, the original magnetic field signal is acquired by an atomic magnetometer, wherein the original magnetic field signal is a time series signal.

[0035] In this step, atomic magnetometer probes, as the core sensors, are precisely deployed at the target measurement points. These probes can measure the absolute strength of the magnetic field, with sensitivity reaching the fT / √Hz level, thus capturing extremely weak magnetic field changes. At each measurement point, a triaxially orthogonal group of atomic magnetometer probes is used to synchronously measure the spatial magnetic induction intensity vector B(x,y,z,t)=(Bx,By,Bz). Vector measurement is crucial for subsequent magnetic field source inversion and fault mode identification. Simultaneously, one or more reference probes are set up in a region far from the transformer body, assumed to be the background magnetic field, to monitor and subtract environmental magnetic noise (such as geomagnetic field fluctuations and distant interference sources). When the system starts, the atomic magnetometers begin acquiring magnetic field signals at each measurement point. These signals change continuously over time, forming raw magnetic field data in time-series format. The data acquisition system is uniformly triggered by a highly stable clock, ensuring strict time synchronization of data from all measurement points, with a synchronization accuracy better than 1. In addition, the system also synchronously collects electrical parameters of the transformer body, including voltage and current waveforms of the primary and secondary sides. These data are also presented in time series form for subsequent data analysis and fault location.

[0036] Step S203: Based on the original magnetic field signal, construct a spatiotemporal magnetic map cube, wherein the spatiotemporal magnetic map cube represents the spatial distribution and temporal variation information of the leakage magnetic field of the target transformer.

[0037] In this step, the system first receives and preprocesses the raw magnetic field signals from the atomic magnetometer at the target measurement point. These signals record the instantaneous intensity of the leakage magnetic field in the form of a time series. The preprocessing includes noise reduction, employing an adaptive noise cancellation technique based on the reference measurement point to effectively filter out power frequency and harmonic interference in the environment, ensuring that the acquired magnetic field signal is pure and represents the actual state of the transformer. Subsequently, based on synchronously acquired electrical parameters (such as voltage and current waveforms), the phase-synchronized magnetic field signal is aligned to obtain the processed magnetic field signal. At each time tk, a three-dimensional leakage magnetic field distribution cloud map is constructed based on the spatial coordinates of all measurement points in the three-dimensional rectangular coordinate system. The leakage magnetic field distribution cloud maps of the continuous time series are stacked to form a spatiotemporal data cube B(x,y,tk), i.e., a spatiotemporal magnetic map cube, thus obtaining data indicators that simultaneously contain information on the spatial distribution and temporal evolution of the leakage magnetic field.

[0038] Step S204: Based on the spatiotemporal magnetic map cube, determine the leakage magnetic flux detection result of the target transformer, wherein the leakage magnetic flux detection result includes the cause of leakage magnetic flux fault and the degree of leakage magnetic flux fault.

[0039] In this step, diverse spatiotemporal spectral features are first extracted from the spatiotemporal magnetic map cube B(x,y,tk), including spatial gradient, centroid, moment of inertia, fundamental frequency, harmonic amplitude, harmonic phase, and high-frequency noise energy. These feature vectors can comprehensively reflect the state changes of the transformer's leakage magnetic field, especially abnormal magnetic field patterns caused by internal faults. Feature extraction aims to quantify the heterogeneity, asymmetry, and time-varying nature of the magnetic field distribution, as well as its phase relationship with electrical quantities. Subsequently, the above feature data are input into a pre-trained deep learning model. The model learns from a large amount of transformer leakage magnetic field data under known conditions, establishing a complex mapping from the feature space to the healthy state. The output not only includes the causes of leakage magnetic field faults, such as "minor winding deformation," "severe winding deformation," and "multiple core grounding," but also provides a quantitative assessment of the severity of the leakage magnetic field fault, for example, expressed as "mild," "moderate," and "severe" and their corresponding confidence probabilities.

[0040] Through the above steps, the purpose of intelligent detection and early warning of transformer leakage flux is achieved, thereby realizing the technical effect of precise monitoring of the internal state of the transformer and early fault warning, and thus solving the technical problem that traditional transformer leakage flux detection technology is difficult to achieve early and accurate capture and location of weak fault signals.

[0041] As an optional embodiment, determining the target measurement point location on the tank wall of the target transformer includes: constructing a three-dimensional rectangular coordinate system outside the tank wall, around the projection areas corresponding to the high-voltage winding, low-voltage winding, and core of the target transformer. The first axis of the three-dimensional rectangular coordinate system is parallel to the winding axis of the target transformer, the second axis is parallel to the winding radial direction of the target transformer, and the third axis is perpendicular to the surface of the tank wall. An electromagnetic simulation model of the target transformer is established in the three-dimensional rectangular coordinate system. An initial measurement point set is constructed, which is an empty set. Based on the electromagnetic simulation model, a target location is selected from all available locations and added to the initial measurement point set, where the target location is the one that minimizes the posterior uncertainty of the electromagnetic simulation model in its current state. If the number of locations in the initial measurement point set exceeds a preset threshold, the addition is stopped, resulting in the final target measurement point set, which serves as the target measurement point location.

[0042] Optionally, a three-dimensional rectangular coordinate system is constructed on the outer wall of the transformer tank, around the projected areas corresponding to the high-voltage and low-voltage windings and the core. The first axis (X-axis) is parallel to the transformer's long axis (i.e., the winding axial direction), the second axis (Y-axis) is parallel to the transformer's short axis (i.e., the winding radial direction), and the third axis (Z-axis) is perpendicular to the tank surface (i.e., the normal direction). An electromagnetic simulation model is established to quantify the impact of different point placement schemes on the uncertainty of fault diagnosis results, actively seeking the optimal measurement point locations that can most effectively reduce diagnostic uncertainty. Specifically, establishing the electromagnetic simulation model first requires constructing a parameterized transformer electromagnetic field finite element model. This model takes the transformer's internal state θ as input parameters, which can include the winding radial displacement, axial displacement, inter-turn short circuit location and degree, core fault location, etc., forming a high-dimensional parameter space. The output is the predicted magnetic field distribution B on the tank surface, which contains the three-dimensional components of all possible measurement points. The calculation formula is as follows:

[0043]

[0044] Here, ε represents the measurement noise, and G represents the mapping from the internal state to the external magnetic field. Next, a priori fault library F is defined, containing M possible fault states. and a normal state Initially, the uncertainty regarding the true state of the transformer is represented by a prior probability distribution P(θ). The goal is to find the optimal set of probe locations S for a given number of probes N, minimizing the uncertainty in state identification. Using a Bayesian framework, after obtaining a set of measurement data d, the posterior probability distribution of state θ is:

[0045]

[0046] Where L(d|θ) represents the probability of observing data d given state θ.

[0047]

[0048] The goal is to find the point layout scheme S that minimizes U(S):

[0049]

[0050] After the simulation model is established, the target measurement point set is solved. First, an initial measurement point set S_0 is established, which is an empty set. Second, the target position s is gradually added from all remaining candidate positions. In the initial set of measurement points, the target position is the position that minimizes the posterior uncertainty of the electromagnetic simulation model in the current state. The calculation formula is:

[0051]

[0052] Finally, the iteration stops when the number of elements in the set reaches a preset value, i.e., the number of probes N. The final set S_N is the target measurement point set, i.e., the optimal placement scheme. Based on the target measurement point set, the target measurement point locations for probe placement can be obtained. After deploying the system and collecting real data, the measured data can be compared with the model predictions. G(θ) is calibrated using the measured data to correct model errors. Based on the calibrated new model and the new posterior uncertainty, the above iterative process can be repeated to dynamically adjust the placement, achieving a closed loop of "perception-learning-optimization".

[0053] As an optional embodiment, the atomic magnetometer probe retains an opening only in the direction of the sensitive axis. The atomic magnetometer probe is connected to the central processing unit via optical fiber, and the central processing unit is used to analyze and process the acquired raw magnetic field signal.

[0054] Optionally, the atomic magnetometer probe is placed inside a magnetically shielded cylinder, with only an opening along the sensitive axis, to suppress lateral interference. The probe is connected to the central processing unit located in the control room via multimode fiber to transmit pump and probe light. This fiber optic link fundamentally eliminates electromagnetic interference introduced by long cables and achieves electrical isolation between the probe and the acquisition circuitry. After acquiring the raw magnetic field signal, the probe transmits it to the central processing unit via fiber optic cable, where the central processing unit analyzes and processes the acquired raw magnetic field signal.

[0055] As an optional embodiment, a spatiotemporal magnetic map cube is constructed based on the original magnetic field signal, including: acquiring electrical data of the target transformer, wherein the electrical data includes voltage data of the primary side of the target transformer, current data of the primary side of the target transformer, voltage data of the secondary side of the target transformer, and current data of the secondary side of the target transformer; denoising the original magnetic field signal, and synchronizing and aligning the original magnetic field signal based on the electrical data to obtain the processed initial magnetic field signal; generating a three-dimensional leakage magnetic field distribution cloud map corresponding to any one moment in the initial magnetic field signal, obtaining multiple three-dimensional leakage magnetic field distribution cloud maps corresponding to each moment; stacking the multiple three-dimensional leakage magnetic field distribution cloud maps corresponding to each moment in chronological order to form a spatiotemporal magnetic map cube.

[0056] Optionally, firstly, electrical data of the target transformer is synchronously acquired, using high-precision voltage and current transformers to obtain the voltage and current waveforms of the primary and secondary sides. This electrical data serves as a reference signal for cross-correlation analysis and phase synchronization alignment with the magnetic field data, distinguishing between changes in leakage flux caused by internal transformer state variations and changes caused by grid fluctuations. Secondly, the original magnetic field signal B(t) is preprocessed using adaptive noise cancellation technology based on reference probes to eliminate power frequency and harmonic interference, random white noise, etc., in the environment. Using the reference voltage and current signals, the magnetic field signal is synchronized and phase-aligned to ensure data consistency, resulting in the processed initial magnetic field signal. For any time tk in the initial magnetic field signal, the magnetic field data from all measuring points are used to generate a three-dimensional leakage flux distribution cloud map based on their spatial coordinates. The leakage flux distribution cloud maps of the continuous time series are stacked to form a spatiotemporal data cube B(x,y,tk). This cube not only contains the spatial distribution information of the leakage magnetic field but also, through time series integration, demonstrates the trend and periodicity of magnetic field changes.

[0057] As an optional embodiment, the leakage flux detection result of the target transformer is determined based on the spatiotemporal magnetic map cube, including: extracting multiple leakage flux features from the spatiotemporal magnetic map cube; inputting the multiple leakage flux features into a preset deep learning model to obtain the leakage flux detection result, wherein the leakage flux fault causes include normal, slight winding deformation, severe winding deformation or multi-point grounding of the iron core, and the leakage flux fault degree includes mild, moderate or severe and the corresponding confidence probability.

[0058] Optionally, multiple magnetic leakage features are first extracted from the spatiotemporal magnetogram cube. The spatial gradient of the magnetogram is then calculated. B(x,y), a high-gradient region, typically corresponds to the center of local magnetic flux distortion caused by a fault. It is also necessary to calculate the centroid and moment of inertia of the magnetic field map to quantify the overall symmetry of the magnetic field distribution. Spectral analysis is performed on the magnetic field signal at each measuring point or specific region to extract the amplitude and phase of the fundamental frequency, each harmonic, and noise energy in specific high-frequency bands (>1kHz), serving as characteristics of winding loosening or partial discharge. The magnetic field map is matched with a fault template library (such as winding deformation templates and core fault templates) established through simulation or historical data to initially locate abnormal areas. By analyzing the time-varying patterns of spatial characteristics and their phase relationship with electrical quantities, the fault type can be preliminarily determined, and its approximate physical location can be determined.

[0059] Secondly, the extracted diverse features (spatial gradient, moment features, harmonic amplitude, high-frequency energy, etc.) are fused into a high-dimensional feature vector, which is then input into a pre-defined deep learning model. This model is trained using a large amount of transformer leakage flux feature vector data under known states (normal, various faults). By learning from a large amount of transformer leakage flux data under known states, the deep learning model can establish a complex nonlinear mapping from the "feature space" to the "health state," outputting the transformer leakage flux detection results. These results include the cause and severity of the leakage flux fault. Causes include normal operation, slight winding deformation, severe winding deformation, or multiple grounding points in the core. Severity includes mild, moderate, or severe, along with the corresponding confidence probability.

[0060] As an optional embodiment, multiple leakage magnetic characteristics include spatial gradient, centroid, moment of inertia, power frequency fundamental frequency, harmonic amplitude, harmonic phase, and high-frequency noise energy.

[0061] Optionally, spatial gradient refers to the rate at which the magnetic field strength changes with spatial location in a three-dimensional leakage magnetic field distribution cloud map. High gradient regions typically indicate local distortions in the magnetic field distribution, potentially corresponding to fault sources within the transformer, such as winding deformation or core anomalies. In geometry, the centroid refers to the center point of the average mass or density distribution of the shape. Applied to magnetic field distribution analysis, changes in the centroid's position reflect the shift in the overall centroid of the leakage magnetic field distribution, closely related to changes in the transformer's internal state, especially the movement of physical locations associated with faults. Moment of inertia is a physical quantity describing the magnitude of an object's inertia during rotation, used to quantify an object's ability to resist rotation. In this context, changes in the moment of inertia of the magnetic field distribution reveal the symmetry and stability of the leakage magnetic field distribution, helping to identify non-uniform magnetic field effects caused by winding slack or structural changes. The power frequency fundamental frequency refers to the first-order sinusoidal component corresponding to the standard frequency of the power system (e.g., 50Hz or 60Hz). It is the most dominant frequency component in the leakage magnetic field signal, and its amplitude and phase variations can indicate significant changes in the transformer's electrical state. Harmonics are frequency components other than the power frequency fundamental frequency, typically appearing as integer multiples of the fundamental frequency. The increase or decrease in harmonic amplitude reflects non-ideal operating conditions within the transformer, such as inter-winding short circuits, inter-turn short circuits, or core saturation. Harmonic phase refers to the phase difference between the harmonic signal and the fundamental frequency or other specific reference signal. Changes in harmonic phase can reveal the dynamic relationship between the leakage magnetic field and the internal structure of the transformer, thus aiding in fault identification and location. High-frequency noise energy refers to the total energy of noise signals in frequency ranges higher than the conventional power frequency range (e.g., >1kHz). An increase in high-frequency noise energy often indicates rapidly changing abnormal activities within the transformer, such as partial discharge, vibration, or poor electrical contact, and is another important indicator for assessing transformer health. These characteristics collectively constitute a diverse set of information describing the leakage magnetic field state of the target transformer. Through intelligent analysis using deep learning models, accurate identification of the causes and extent of transformer faults can be achieved.

[0062] As an optional embodiment, Figure 3 This is an architecture diagram of a transformer leakage flux detection system provided according to an optional embodiment of the present invention, such as... Figure 3As shown, the system comprises a power transformer, an atomic magnetometer probe array, a signal acquisition layer, and a data processing and diagnostic layer. The power transformer is the monitored device, containing core components such as the core and windings. The atomic magnetometer probe array is a quantum magnetic field sensor network arranged in a grid pattern on the outer wall of the transformer tank. Each measurement point contains a triaxially orthogonal atomic magnetometer probe for measuring three-dimensional magnetic field components. An optical fiber network connects the probe array to the host computer, enabling optical signal transmission and electrical isolation, effectively resisting electromagnetic interference. The reference electrical quantity acquisition unit synchronously acquires the voltage U(t) and current I(t) signals of the transformer during operation via voltage transformers (PTs) and current transformers (CTs). The synchronous acquisition host contains a laser driver and photoelectric conversion module, a high-precision analog-to-digital converter, a high-stability clock source, and a channel synchronous data acquisition card, connected to the optical fiber network, for processing the acquired electrical data. The data analysis and diagnostic terminal is a high-performance computing workstation that runs signal processing, magnetograph reconstruction, feature extraction, and AI diagnostic algorithms, providing a visual interface and early warning information.

[0063] As an optional embodiment, Figure 4 This is a flowchart of the transformer leakage flux detection implementation provided by an optional embodiment of the present invention, such as... Figure 4 As shown, the process includes system deployment, data acquisition and magnetograph generation, feature extraction and localization, AI diagnosis, early warning, and verification. First, an array of atomic magnetometers is deployed on the transformer tank wall. The probes are connected to the acquisition system located in the main control room via optical fiber, simultaneously acquiring three-phase voltage and current. The system continuously acquires data and, under a certain load, reconstructs the magnetograph of the normal component Bz. A localized area of ​​enhanced magnetic flux density, in phase with the fundamental magnetic field, is discovered. Second, the spatial gradient is calculated to determine the center coordinates of this abnormal area. Spectral analysis compares the amplitude of the power frequency (50Hz) magnetic flux density in this area with the normal baseline. Finally, feature vectors containing information such as "peak spatial gradient in a certain area," "power frequency amplitude increment," and "harmonic increment" are input into a pre-trained hybrid model. After model analysis, the diagnostic results and confidence levels are output, a diagnostic report is generated, and an early warning is issued.

[0064] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0065] Through the above description of the embodiments, those skilled in the art can clearly understand that the transformer leakage flux detection method according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0066] According to embodiments of the present invention, an apparatus for implementing the above-described transformer leakage flux detection method is also provided. Figure 5 This is a structural block diagram of a transformer leakage flux detection device provided according to an embodiment of the present invention, such as... Figure 5 As shown, the device includes: a first determining module 51, an acquiring module 52, a constructing module 53, and a second determining module 54. The device will be described below.

[0067] The first determining module 51 is used to determine the location of the target measuring point on the tank wall of the target transformer.

[0068] The acquisition module 52 is connected to the first determination module 51 and is used to acquire the original magnetic field signal at the target measurement point location using an atomic magnetometer, wherein the original magnetic field signal is a time series signal.

[0069] The construction module 53, connected to the acquisition module 52, is used to construct a spatiotemporal magnetic map cube based on the original magnetic field signal. The spatiotemporal magnetic map cube represents the spatial distribution and temporal variation information of the leakage magnetic field of the target transformer.

[0070] The second determining module 54, connected to the construction module 53, is used to determine the leakage magnetic field detection result of the target transformer based on the spatiotemporal magnetic map cube. The leakage magnetic field detection result includes the cause of the leakage magnetic field fault and the degree of the leakage magnetic field fault.

[0071] It should be noted that the first determining module 51, the obtaining module 52, the constructing module 53, and the second determining module 54 mentioned above correspond to steps S201 to S204 in the embodiments. Multiple modules implement the same instances and application scenarios as their corresponding steps, but are not limited to the content disclosed in the above embodiments. It should also be noted that the above modules, as part of the device, can run on the computer terminal 10 provided in the embodiments.

[0072] Embodiments of the present invention may provide a computer device. Optionally, in this embodiment, the computer device may be located in at least one of a plurality of network devices in a computer network. The computer device includes a memory and a processor.

[0073] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the transformer leakage flux detection method and device in this embodiment of the invention. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby realizing the aforementioned transformer leakage flux detection method. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to a computer terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0074] The processor can call the information and application program stored in the memory through the transmission device to perform the following steps: determine the location of the target measuring point on the tank wall of the target transformer; at the target measuring point location, acquire the original magnetic field signal through an atomic magnetometer, wherein the original magnetic field signal is a time series signal; based on the original magnetic field signal, construct a spatiotemporal magnetic map cube, wherein the spatiotemporal magnetic map cube represents the spatial distribution information and temporal variation information of the leakage magnetic field of the target transformer; based on the spatiotemporal magnetic map cube, determine the leakage magnetic field detection result of the target transformer, wherein the leakage magnetic field detection result includes the cause of leakage magnetic field fault and the degree of leakage magnetic field fault.

[0075] Optionally, the processor may also execute program code for the following steps: determining the target measurement point location on the tank wall of the target transformer, including: constructing a three-dimensional rectangular coordinate system outside the tank wall, around the projection areas corresponding to the high-voltage winding, low-voltage winding, and core of the target transformer, wherein the first axis of the three-dimensional rectangular coordinate system is parallel to the winding axis of the target transformer, the second axis of the three-dimensional rectangular coordinate system is parallel to the winding radial direction of the target transformer, and the third axis of the three-dimensional rectangular coordinate system is perpendicular to the surface of the tank wall; establishing an electromagnetic simulation model of the target transformer in the three-dimensional rectangular coordinate system; constructing an initial measurement point set, wherein the initial measurement point set is an empty set; based on the electromagnetic simulation model, selecting one target location from all available locations and adding it to the initial measurement point set, wherein the target location is the location that reduces the posterior uncertainty of the electromagnetic simulation model in the current state to the greatest extent; stopping the addition when the number of locations in the initial measurement point set exceeds a preset threshold, and obtaining the final target measurement point set as the target measurement point location.

[0076] Optionally, the processor can also execute program code for the following steps: the probe of the atomic magnetometer retains an opening only in the direction of the sensitive axis; the probe of the atomic magnetometer is connected to the central processing host via an optical fiber; and the central processing host is used to analyze and process the acquired raw magnetic field signal.

[0077] Optionally, the processor may also execute program code for the following steps: constructing a spatiotemporal magnetic map cube based on the original magnetic field signal, including: acquiring electrical data of the target transformer, wherein the electrical data includes voltage data of the primary side of the target transformer, current data of the primary side of the target transformer, voltage data of the secondary side of the target transformer, and current data of the secondary side of the target transformer; performing noise reduction processing on the original magnetic field signal, and synchronizing and aligning the original magnetic field signal with phase based on the electrical data to obtain the processed initial magnetic field signal; generating a three-dimensional leakage magnetic field distribution cloud map corresponding to any one moment in the initial magnetic field signal for the magnetic field data at any moment, obtaining multiple three-dimensional leakage magnetic field distribution cloud maps corresponding to each moment; stacking the multiple three-dimensional leakage magnetic field distribution cloud maps corresponding to each moment in chronological order to form a spatiotemporal magnetic map cube.

[0078] Optionally, the processor may also execute program code for the following steps: determining the leakage flux detection result of the target transformer based on the spatiotemporal magnetic map cube, including: extracting multiple leakage flux features from the spatiotemporal magnetic map cube; inputting the multiple leakage flux features into a preset deep learning model to obtain the leakage flux detection result, wherein the leakage flux fault causes include normal, slight winding deformation, severe winding deformation or multi-point grounding of the iron core, and the leakage flux fault degree includes mild, moderate or severe and the corresponding confidence probability.

[0079] Optionally, the processor may also execute program code for the following steps: multiple leakage magnetic characteristics including spatial gradient, centroid, moment of inertia, power frequency fundamental wave, harmonic amplitude, harmonic phase, and high-frequency noise energy.

[0080] This invention provides a scheme for detecting transformer leakage flux. The scheme involves determining the location of a target measurement point on the tank wall of the target transformer; acquiring the original magnetic field signal at the target measurement point using an atomic magnetometer, wherein the original magnetic field signal is a time-series signal; constructing a spatiotemporal magnetic map cube based on the original magnetic field signal, wherein the spatiotemporal magnetic map cube represents the spatial distribution and temporal variation information of the leakage magnetic field of the target transformer; and determining the leakage flux detection result of the target transformer based on the spatiotemporal magnetic map cube, wherein the leakage flux detection result includes the cause and degree of leakage flux fault, thereby achieving the purpose of intelligent detection and early warning of transformer leakage flux, and solving the technical problem in related technologies where traditional transformer leakage flux detection technology struggles to achieve early and accurate capture and location of weak fault signals.

[0081] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a non-volatile storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0082] Embodiments of the present invention also provide a non-volatile storage medium. Optionally, in this embodiment, the aforementioned non-volatile storage medium can be used to store the program code executed by the transformer leakage flux detection method provided in the above embodiments.

[0083] Optionally, in this embodiment, the non-volatile storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals.

[0084] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: determining the location of a target measurement point on the tank wall of the target transformer; acquiring the original magnetic field signal at the target measurement point location using an atomic magnetometer, wherein the original magnetic field signal is a time-series signal; constructing a spatiotemporal magnetic map cube based on the original magnetic field signal, wherein the spatiotemporal magnetic map cube characterizes the spatial distribution and temporal variation information of the leakage magnetic field of the target transformer; and determining the leakage magnetic field detection result of the target transformer based on the spatiotemporal magnetic map cube, wherein the leakage magnetic field detection result includes the cause and degree of leakage magnetic field fault.

[0085] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: determining the target measurement point location on the tank wall of the target transformer, including: constructing a three-dimensional rectangular coordinate system outside the tank wall, around the projection areas corresponding to the high-voltage winding, low-voltage winding, and core of the target transformer, wherein the first axis of the three-dimensional rectangular coordinate system is parallel to the winding axis of the target transformer, the second axis of the three-dimensional rectangular coordinate system is parallel to the winding radial direction of the target transformer, and the third axis of the three-dimensional rectangular coordinate system is perpendicular to the surface of the tank wall; establishing an electromagnetic simulation model of the target transformer in the three-dimensional rectangular coordinate system; constructing an initial measurement point set, wherein the initial measurement point set is an empty set; based on the electromagnetic simulation model, selecting one target location from all selectable locations and adding it to the initial measurement point set, wherein the target location is the location that reduces the posterior uncertainty of the electromagnetic simulation model in the current state the most; stopping the addition when the number of locations in the initial measurement point set exceeds a preset threshold, and obtaining the final target measurement point set as the target measurement point location.

[0086] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: the probe of the atomic magnetometer retains an opening only in the direction of the sensitive axis, the probe of the atomic magnetometer is connected to the central processing unit via an optical fiber, and the central processing unit is used to analyze and process the acquired raw magnetic field signal.

[0087] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: constructing a spatiotemporal magnetic map cube based on the original magnetic field signal, including: acquiring electrical data of the target transformer, wherein the electrical data includes voltage data of the primary side of the target transformer, current data of the primary side of the target transformer, voltage data of the secondary side of the target transformer, and current data of the secondary side of the target transformer; performing noise reduction processing on the original magnetic field signal, and synchronizing and aligning the original magnetic field signal based on the electrical data to obtain the processed initial magnetic field signal; generating a three-dimensional leakage magnetic field distribution cloud map corresponding to any one moment in the initial magnetic field signal for the magnetic field data at any moment, obtaining multiple three-dimensional leakage magnetic field distribution cloud maps corresponding to each moment; stacking the multiple three-dimensional leakage magnetic field distribution cloud maps corresponding to each moment in chronological order to form a spatiotemporal magnetic map cube.

[0088] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: determining the leakage flux detection result of the target transformer based on the spatiotemporal magnetic map cube, including: extracting multiple leakage flux features from the spatiotemporal magnetic map cube; inputting the multiple leakage flux features into a preset deep learning model to obtain the leakage flux detection result, wherein the leakage flux fault causes include normal, slight winding deformation, severe winding deformation or multi-point grounding of the iron core, and the leakage flux fault degree includes mild, moderate or severe and the corresponding confidence probability.

[0089] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: multiple leakage magnetic features include spatial gradient, centroid, moment of inertia, power frequency fundamental wave, harmonic amplitude, harmonic phase, and high-frequency noise energy.

[0090] Embodiments of the present invention also provide a computer program product, including a computer program. Optionally, in this embodiment, when the computer program is executed by a processor, it can: determine the location of a target measuring point on the tank wall of a target transformer; acquire the original magnetic field signal at the target measuring point location using an atomic magnetometer, wherein the original magnetic field signal is a time-series signal; construct a spatiotemporal magnetic map cube based on the original magnetic field signal, wherein the spatiotemporal magnetic map cube characterizes the spatial distribution and temporal variation information of the leakage magnetic field of the target transformer; and determine the leakage magnetic field detection result of the target transformer based on the spatiotemporal magnetic map cube, wherein the leakage magnetic field detection result includes the cause and degree of leakage magnetic field fault.

[0091] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0092] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0093] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0094] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0095] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0096] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a non-volatile storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0097] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for detecting transformer leakage flux, characterized in that, include: Determine the location of the target measuring point on the tank wall of the target transformer; At the target measurement point, the original magnetic field signal is acquired by an atomic magnetometer, wherein the original magnetic field signal is a time series signal; Based on the original magnetic field signal, a spatiotemporal magnetic map cube is constructed, wherein the spatiotemporal magnetic map cube represents the spatial distribution and temporal variation information of the leakage magnetic field of the target transformer; Based on the spatiotemporal magnetic map cube, the leakage flux detection result of the target transformer is determined, wherein the leakage flux detection result includes the cause of the leakage flux fault and the degree of the leakage flux fault.

2. The method according to claim 1, characterized in that, Determining the location of the target measuring point on the tank wall of the target transformer includes: Outside the tank wall, a three-dimensional rectangular coordinate system is constructed around the projection areas corresponding to the high-voltage winding, low-voltage winding, and core of the target transformer. The first axis of the three-dimensional rectangular coordinate system is parallel to the winding axis of the target transformer, the second axis is parallel to the winding radial direction of the target transformer, and the third axis is perpendicular to the surface of the tank wall. An electromagnetic simulation model of the target transformer is established in the three-dimensional rectangular coordinate system. Construct an initial set of measurement points, wherein the initial set of measurement points is an empty set; Based on the electromagnetic simulation model, a target location is selected from all available locations and added to the initial measurement point set. The target location is the location that reduces the posterior uncertainty of the electromagnetic simulation model in the current state the most. If the number of locations in the initial set of measuring points exceeds a preset threshold, the addition is stopped, and the final set of target measuring points is obtained, which is used as the target measuring point location.

3. The method according to claim 1, characterized in that, The probe of the atomic magnetometer has an opening only in the direction of the sensitive axis. The probe of the atomic magnetometer is connected to the central processing unit via an optical fiber. The central processing unit is used to analyze and process the acquired raw magnetic field signal.

4. The method according to claim 1, characterized in that, The construction of the spatiotemporal magnetograph cube based on the original magnetic field signal includes: Acquire the electrical data of the target transformer, wherein the electrical data includes the voltage data of the primary side of the target transformer, the current data of the primary side of the target transformer, the voltage data of the secondary side of the target transformer, and the current data of the secondary side of the target transformer; The original magnetic field signal is denoised, and the original magnetic field signal is synchronized and phase-aligned based on the electrical data to obtain the processed initial magnetic field signal. For the magnetic field data at any time in the initial magnetic field signal, generate a three-dimensional magnetic flux leakage distribution cloud map at any time, and obtain three-dimensional magnetic flux leakage distribution cloud maps at multiple times. The three-dimensional magnetic flux leakage distribution cloud maps corresponding to the multiple time points are stacked in chronological order to form the spatiotemporal magnetic map cube.

5. The method according to claim 1, characterized in that, The determination of the leakage flux detection result of the target transformer based on the spatiotemporal magnetic map cube includes: Multiple magnetic leakage features were extracted from the spatiotemporal magnetic map cube; The multiple leakage magnetic features are input into a preset deep learning model to obtain the leakage magnetic detection result. The causes of the leakage magnetic fault include normal, slight winding deformation, severe winding deformation or multi-point grounding of the iron core. The degree of leakage magnetic fault includes mild, moderate or severe and the corresponding confidence probability.

6. The method according to claim 5, characterized in that, The multiple leakage magnetic characteristics include spatial gradient, centroid, moment of inertia, power frequency fundamental frequency, harmonic amplitude, harmonic phase, and high-frequency noise energy.

7. A transformer leakage flux detection device, characterized in that, include: The first determining module is used to determine the location of the target measuring point on the tank wall of the target transformer; The acquisition module is used to acquire the original magnetic field signal at the target measurement point location using an atomic magnetometer, wherein the original magnetic field signal is a time series signal; A construction module is used to construct a spatiotemporal magnetic map cube based on the original magnetic field signal, wherein the spatiotemporal magnetic map cube represents the spatial distribution and temporal variation information of the leakage magnetic field of the target transformer; The second determining module is used to determine the leakage magnetic field detection result of the target transformer based on the spatiotemporal magnetic map cube, wherein the leakage magnetic field detection result includes the cause of the leakage magnetic field fault and the degree of the leakage magnetic field fault.

8. A non-volatile storage medium, characterized in that, The non-volatile storage medium includes a stored program, wherein, when the program is executed, it controls the device containing the non-volatile storage medium to perform the transformer leakage flux detection method according to any one of claims 1 to 6.

9. A computer device, characterized in that, include: Memory and processor The memory stores computer programs; The processor is configured to execute a computer program stored in the memory, wherein when the computer program is executed, the processor performs the transformer leakage flux detection method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the transformer leakage flux detection method according to any one of claims 1 to 6.

Images