Strain monitoring system and method for transformer winding

Abstract

The invention relates to a strain monitoring system and method for a transformer winding. The strain monitoring system comprises the transformer winding and a strain monitoring device. The strain monitoring device comprises a fiber bragg grating array which is arranged on a transformer winding and is used for capturing strain signals of the transformer winding; the signal processor is connected with the fiber grating array and is used for receiving and processing the strain signal to generate an electric signal; the data analysis unit is connected with the signal processor, and the data analysis unit is used for receiving and analyzing the electric signal to obtain strain information of the transformer winding; and the wireless transmission unit is connected with the data analysis unit and is used for wirelessly transmitting the strain information to a monitoring center. According to the invention, the fiber grating array is arranged on the transformer winding, and the fiber grating array transmits the strain information through the optical signal, so that the electromagnetic noise is suppressed and the monitoring accuracy is improved by using the characteristic that the fiber is not interfered by the strong electromagnetic environment of the transformer.

Description

Technical Field

[0001] The present application relates to the technical field of transformer monitoring, and in particular to a strain monitoring system and method for transformer windings. Background Art

[0002] With the large-scale deployment of renewable energy, the large power transformers supporting wind turbines are shouldering a critical role in power conversion. As a core component of the transformer, the structural integrity of the windings directly determines the equipment lifespan and system stability. During operation, the windings are subjected to complex alternating electromagnetic forces, mechanical vibration, and thermal stresses, leading to cumulative strain damage (such as insulation paper creep and conductor deformation), potentially causing inter-turn short circuits, localized overheating, and even sudden burnout.

[0003] Currently, transformer winding strain monitoring mainly relies on at least the following two types of technologies: resistance strain gauge solutions and vibration sensor solutions. However, these solutions have at least the following problems: single-point sensors cannot achieve both high resolution and large-scale monitoring, and electromagnetic noise and harsh operating conditions lead to a high risk of sensor failure. Therefore, there is an urgent need for a strain monitoring method that can provide high precision and high reliability to adapt to complex and changing environmental conditions and meet the needs of new power systems. Summary of the Invention

[0004] Based on this, it is necessary to provide a strain monitoring system and method for transformer windings to address the problems of high failure risk, low accuracy and low reliability of existing transformer winding strain monitoring technology.

[0005] In a first aspect, an embodiment of the present application provides a strain monitoring system for a transformer winding, the strain monitoring system comprising a transformer winding and a strain monitoring device;

[0006] Wherein, the strain monitoring device includes:

[0007] A fiber Bragg grating array is provided on the transformer winding, and the fiber Bragg grating array is used to capture the strain signal of the transformer winding;

[0008] a signal processor connected to the fiber grating array, the signal processor being configured to receive and process the strain signal to generate an electrical signal;

[0009] a data analysis unit connected to the signal processor, the data analysis unit being configured to receive and analyze the electrical signal to obtain strain information of the transformer winding;

[0010] A wireless transmission unit is connected to the data analysis unit, and is used to wirelessly transmit the strain information to a monitoring center.

[0011] In one embodiment, the fiber grating array is provided on the surface of the transformer winding;

[0012] And / or, the transformer winding includes an insulation structure, the insulation structure includes an insulation spacer and the strain monitoring device embedded in the insulation spacer.

[0013] In one embodiment, the signal processor includes a wavelength demodulation module, and the wavelength demodulation module is used to convert the wavelength change of each fiber Bragg grating center in the fiber Bragg grating array into the electrical signal;

[0014] And / or, the data analysis unit includes a strain calculation module, and the strain calculation module is used to calculate the strain magnitude of the transformer winding according to the electrical signal.

[0015] In one embodiment, the transformer winding has multiple monitoring areas, each monitoring area corresponds to a different stress; each monitoring area is provided with the strain monitoring device, and the arrangement density of each fiber grating in the fiber grating array in each monitoring area is different.

[0016] In one embodiment, the transformer winding has a first stress monitoring area and a second stress monitoring area, and the arrangement density of the fiber gratings in the first stress monitoring area is greater than the arrangement density of the fiber gratings in the second stress monitoring area; wherein the stress corresponding to the first stress monitoring area is greater than the stress corresponding to the second stress monitoring area;

[0017] The spacing between adjacent fiber Bragg gratings in the first stress monitoring area is between 3 cm and 5 cm, and the spacing between adjacent fiber Bragg gratings in the second stress monitoring area is between 10 cm and 15 cm.

[0018] In one embodiment, the strain monitoring system further includes a high-frequency current transformer, which is connected to the transformer winding and is used to collect partial discharge signals of the transformer winding;

[0019] And / or, the strain monitoring system further includes a temperature sensor, which is provided on the transformer winding and is used to obtain the temperature of the transformer winding.

[0020] In one embodiment, the strain monitoring system further includes a fault warning model module;

[0021] The fault warning model module is connected to the wireless transmission unit, and is used to compare the current strain information of the transformer winding with pre-stored historical strain data to issue a warning on the fault type of the transformer winding;

[0022] Alternatively, the fault warning model module is connected to the wireless transmission unit, the high-frequency current transformer, and the temperature sensor respectively, and the fault warning model module is used to integrate the current strain information, partial discharge data, and temperature data of the transformer winding, and compare them with pre-stored historical parameter data to issue a warning on the fault type of the transformer winding.

[0023] In a second aspect, an embodiment of the present application provides a strain monitoring method for a transformer winding, the strain monitoring method comprising:

[0024] Establishing a stress identification simulation model and generating stress distribution data of the transformer winding based on a design model of the transformer winding;

[0025] Calibrate multiple monitoring areas of the transformer winding with different stresses based on the stress distribution data, and design fiber Bragg grating arrays with different arrangement densities in each of the monitoring areas;

[0026] Based on the collected strain information, partial discharge data and temperature data of the transformer winding, a multi-parameter fault warning model is established, and multiple parameter data are analyzed to identify the abnormal type of the transformer winding and send a warning signal to the monitoring center in real time.

[0027] In one embodiment, the transformer winding includes a first stress monitoring area and a second stress monitoring area, and the stress corresponding to the first stress monitoring area is greater than the stress corresponding to the second stress monitoring area;

[0028] The step of establishing a multi-parameter fault warning model based on the collected strain information, partial discharge data and temperature data of the transformer winding includes:

[0029] The wavelength offset data of the fiber grating array is obtained, and the partial discharge pulse sequence and temperature distribution spectrum of the transformer winding are obtained to construct the multi-parameter fault warning model.

[0030] In one embodiment, the abnormality type includes at least one of looseness, insulation degradation, and overheating;

[0031] The step of analyzing multiple parameter data to identify the abnormal type of the transformer winding and sending an early warning signal to a monitoring center in real time includes:

[0032] Comparing the plurality of parameter data with pre-stored historical parameter data to output an abnormality type of the transformer winding;

[0033] Integrating the plurality of parameter data and generating a strain deviation;

[0034] The strain deviation is compared with multiple thresholds. If the strain deviation exceeds any threshold, warning signals of different levels are generated.

[0035] The strain monitoring system and method for transformer windings described above utilizes a fiber grating array (FBG) array installed on the transformer windings. The FBG array transmits strain information via optical signals, utilizing the optical fiber's resistance to interference from the transformer's strong electromagnetic environment to suppress electromagnetic noise and improve monitoring accuracy. The FBG array is deployed at multiple points, covering the windings in both the axial and radial directions, to capture the full strain distribution of the windings. This helps capture strain differences between the winding ends and the middle, thereby reducing missed detection rates. This embodiment directly senses the mechanical deformation of the windings through FBGs, reducing measurement errors. The FBG array is sensitive to static strain and facilitates monitoring of minute strain changes. Furthermore, the FBG array in this embodiment can be installed before the transformer leaves the factory, requiring only remote diagnosis via wireless signals for subsequent maintenance. Furthermore, the historical data storage and trend analysis capabilities of the data analysis unit provide early warning of potential failures, reducing unplanned downtime for repairs. Compared to resistance strain gauge solutions, this significantly reduces operational costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a schematic structural diagram of a strain monitoring system for transformer windings provided according to some embodiments of the present application.

[0037] Figure 2 Schematic diagram of a flow chart of a strain monitoring method for transformer windings according to some embodiments of the present application.

[0038] Figure Number:

[0039] 100. Transformer winding;

[0040] 200 , strain monitoring device; 210 , fiber Bragg grating array; 220 , signal processor; 230 , data analysis unit; 240 , wireless transmission unit. DETAILED DESCRIPTION

[0041] To make the above-mentioned objects, features, and advantages of the present application more clearly understood, the specific embodiments of the present application are described in detail below with reference to the accompanying drawings. The following description sets forth many specific details to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar improvements without violating the scope of the present application. Therefore, the present application is not limited to the specific embodiments disclosed below.

[0042] In the description of this application, it should be understood that if the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc. appear, the orientation or position relationship indicated by these terms is based on the orientation or position relationship shown in the accompanying drawings, which is only for the convenience of describing this application and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on this application.

[0043] In addition, if the terms "first" or "second" appear, these terms are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include at least one of such features. In the description of this application, if the term "plurality" appears, "plurality" means at least two, for example, two, three, etc., unless otherwise specifically defined.

[0044] In this application, unless otherwise specified or limited, the terms "mounted," "connected," "connected," "fixed," etc., should be interpreted broadly. For example, these terms may refer to fixed connections, removable connections, or integration; mechanical connections or electrical connections; direct connections or indirect connections through an intermediary; and internal communication between two components or interaction between two components, unless otherwise specified. Those skilled in the art will understand the specific meanings of these terms in this application based on the specific circumstances.

[0045] In this application, unless otherwise expressly specified or limited, if a first feature is described as being "above" or "below" a second feature, or similar descriptions, this may mean that the first and second features are in direct contact, or that the first and second features are in indirect contact through an intermediate medium. Furthermore, when a first feature is described as being "above," "above," or "above" a second feature, it may mean that the first feature is directly above or diagonally above the second feature, or simply means that the first feature is at a higher level than the second feature. When a first feature is described as being "below," "below," or "below" a second feature, it may mean that the first feature is directly below or diagonally below the second feature, or simply means that the first feature is at a lower level than the second feature.

[0046] It should be noted that if an element is referred to as being "fixed to" or "disposed on" another element, it may be directly on the other element or there may be an intermediate element. If an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be an intermediate element. If any, the terms "vertical", "horizontal", "upper", "lower", "left", "right" and similar expressions used in this application are for illustrative purposes only and do not represent the only embodiment.

[0047] As mentioned in the background technology, during the operation of the transformer, the windings are subjected to complex mechanical loads, including complex electromagnetic environments, gravity, and centrifugal forces. These loads will cause strain in the transformer windings, which in turn affects their structural integrity and service life. Long-term or excessive strain may cause damage or fatigue to the windings, and may even cause failure of the entire transformer. Therefore, real-time and accurate monitoring of the strain state of the transformer windings is crucial to ensuring the efficient and safe operation of the power system. In the existing transformer winding strain monitoring technology, the resistance strain gauge solution is specifically: measuring local deformation by pasting metal foil, but there are significant defects, such as weak anti-electromagnetic interference ability, including high signal distortion rate in strong electromagnetic environments; and insufficient spatial coverage, including the inability of single-point monitoring to capture the full-domain strain distribution of the winding, resulting in a high missed detection rate; difficult installation and maintenance: power outage construction is required and the service life is short, increasing operation and maintenance costs. Specifically, the vibration sensor solution indirectly infers strain based on an acceleration sensor, which has fundamental limitations, including large indirect measurement errors: the vibration signal is nonlinearly related to strain, and temperature drift causes large errors; it cannot identify static deformation, including the inability to monitor long-term creep failure of the winding; and it has poor environmental adaptability, including a more serious decrease in sensitivity when the humidity is high.

[0048] To address the above-mentioned problems, an embodiment of the present application provides a strain monitoring system and method for transformer windings. By disposing a fiber grating array (FBG) on the transformer windings, the FBG array transmits strain information via optical signals, utilizing the optical fiber's resistance to interference from the transformer's strong electromagnetic environment to suppress electromagnetic noise and improve monitoring accuracy. The FBG array is arranged at multiple points, covering the winding's axial and radial directions, to capture the full strain distribution of the winding, facilitating the capture of strain differences between the winding's ends and center, thereby reducing the missed detection rate. This embodiment directly senses the mechanical deformation of the windings through FBGs, reducing measurement errors. The FBG array is sensitive to static strain and facilitates monitoring of minute strain changes. Furthermore, the FBG array in this embodiment can be completed before the transformer leaves the factory, and subsequent maintenance requires only remote diagnosis via wireless signals. Furthermore, the historical data storage and trend analysis functions of the data analysis unit can provide early warning of potential faults, reducing unplanned downtime for repairs. Compared to resistance strain gauge solutions, this significantly reduces operation and maintenance costs.

[0049] See Figure 1 , Figure 1 The present invention provides a schematic structural diagram of a strain monitoring system for transformer windings according to some embodiments of the present application. One embodiment of the present application first provides a strain monitoring system for transformer windings, which may include a transformer winding 100 and a strain monitoring device 200. The strain monitoring device 200 may include a fiber Bragg grating array 210, a signal processor 220, a data analysis unit 230, and a wireless transmission unit 240.

[0050] A fiber Bragg grating array 210 is provided on the transformer winding 100 and is used to capture the strain signal of the transformer winding 100; a signal processor 220 is connected to the fiber Bragg grating array 210 and is used to receive and process the strain signal to generate an electrical signal; a data analysis unit 230 is connected to the signal processor 220 and is used to receive and analyze the electrical signal to obtain strain information of the transformer winding 100; a wireless transmission unit 240 is connected to the data analysis unit 230 and is used to wirelessly transmit the strain information to a monitoring center.

[0051] As will be understood, transformer winding 100 is a core component of a conventional power transformer. It is constructed from multiple turns of insulated wire (such as copper or aluminum) and typically includes high-voltage and low-voltage windings, nested outside the transformer core. During operation, it converts electrical energy to voltage through electromagnetic induction. During operation, strain (such as tensile or compressive deformation of the wire and creep of the insulation layer) generated by electromagnetic forces and thermal stresses is monitored by this system.

[0052] The fiber grating array 210 consists of a single optical fiber with multiple fiber gratings engraved within it. The grating spacing is 5-10 cm (adjustable based on winding size), covering critical axial and radial areas of the transformer winding 100 (such as the winding ends, the middle, and stress concentration points). The optical fiber is coated with an oil- and high-temperature-resistant polyimide to ensure long-term stable operation in the transformer's oil-immersed environment.

[0053] A fiber Bragg grating (FBG) is a periodic refractive index modulated structure formed within the fiber core by ultraviolet light interference. The central wavelength of its reflected light shifts with external strain (or temperature). Strain changes the grating period, and thus the wavelength. When multiple gratings are connected in series to form an array, the characteristic wavelengths of each grating can be used to distinguish monitoring points, enabling the simultaneous capture of strain at different locations in the winding.

[0054] The fiber grating array 210 in this example can be a distributed fiber grating array 210, which can be set on the surface of the transformer winding 100, or pre-buried in the insulating spacer of the transformer winding 100. During the transformer manufacturing stage, the insulating spacer integrated with the fiber grating array 210 can be assembled synchronously with the winding coil to fully couple the grating period with the winding deformation.

[0055] Signal processor 220 may include a broadband light source, an optical coupler, a fiber Bragg grating (FBG) demodulation module, and a signal conversion circuit. The entire system is enclosed in a metal shielding box to accommodate the dust and humidity environment near the transformer. The general operating principle of signal processor 220 can be understood as follows: an optical signal emitted by the broadband light source is transmitted to the fiber Bragg grating array 210 via an optical coupler. Each grating reflects an optical signal of a specific wavelength (carrying strain information), and the reflected light is returned to the demodulation module via the coupler. The demodulation module identifies the center wavelength offset of each grating through spectral analysis and converts it into an analog electrical signal. The signal conversion circuit then filters the analog signal to generate a standardized electrical signal.

[0056] Data analysis unit 230 can be an industrial-grade single-chip microcomputer or embedded computer with a built-in strain calculation algorithm and data storage module. The operating principle of data analysis unit 230 can be summarized as follows: after receiving the electrical signal output by signal processor 220, it converts the wavelength offset into actual strain value using a preset calibration coefficient. It also combines the strain data from multiple measurement points to generate a global strain distribution map of the winding. It determines the presence of abnormal strain by setting a threshold and records the strain change curve over time.

[0057] The wireless transmission unit 240 can use an industrial-grade wireless module and be encapsulated in a waterproof housing. It receives the strain information output by the data analysis unit 230, including the implemented strain value, distribution map, and abnormal warning signal light, and transmits it to the remote monitoring center after encryption through the wireless communication protocol.

[0058] When the transformer winding 100 is in operation, strain is generated. For example, radial expansion caused by electromagnetic forces and axial expansion and contraction caused by temperature changes are transmitted to the fiber grating array 210, causing the central wavelength of each grating to shift. The broadband light source of the signal processor 220 transmits an optical signal. After transmission through the fiber grating, the demodulation module identifies the wavelength offset and converts it into an electrical signal. After filtering and amplification, the signal is output to the data analysis unit 230. The data analysis unit 230 converts the electrical signal into a strain value, generates a global strain distribution map, determines whether there is super-threshold strain, for example, a sudden increase in local strain may indicate a short circuit between turns, and stores historical data. The wireless transmission unit 240 sends real-time strain information, warning signals, etc. to the monitoring center, enabling remote real-time monitoring.

[0059] In this embodiment, based on ANSYS finite element simulation, high stress areas such as leakage magnetic field concentration areas at the ends of the transformer winding 100 and pressure plate support points can be identified, so that fiber Bragg gratings of different densities can be set in different stress areas of the transformer winding 100 to reduce the missed detection rate.

[0060] In addition, 2-3 overlapping gratings can be arranged at key locations of the winding, such as end corners, to eliminate abnormal values ​​through data comparison and improve monitoring reliability.

[0061] Furthermore, a temperature-sensitive grating (FBG) can be connected in series to the fiber Bragg grating array 210. The data analysis unit 230 uses a temperature-strain cross-coupling algorithm to correct for temperature interference with wavelength shift, thereby reducing strain measurement errors. Of course, the temperature-sensitive grating in this example is only sensitive to temperature and is not affected by strain.

[0062] Furthermore, the data analysis unit 230 can dynamically adjust the sampling frequency according to the strain change rate, for example, adopting low frequency sampling when the strain is stable and increasing the sampling frequency when the strain suddenly changes, so as to reduce the amount of data transmission.

[0063] Furthermore, a fiber Bragg grating array 210 with a relatively high density may be arranged on the transformer winding 100 and dynamically adjusted to extract data of a portion of the fiber Bragg grating for processing.

[0064] In application, the strain monitoring device 200 in this embodiment is applied to a 500KV transformer in a certain wind farm, which can provide an early warning of winding loosening 72 hours in advance, thereby avoiding large economic losses.

[0065] In summary, the strain monitoring system for transformer windings provided by the embodiments of the present application comprises a fiber Bragg grating array (FBG) 210 disposed on the transformer winding 100. The FBG array 210 transmits strain information via optical signals, utilizing the optical fiber's resistance to interference from the transformer's strong electromagnetic environment to suppress electromagnetic noise and improve monitoring accuracy. The FBG array 210 is arranged at multiple points, covering the winding's axial and radial directions, to capture the full strain distribution of the winding. This facilitates capturing the strain difference between the winding's ends and center, thereby reducing the missed detection rate. This embodiment directly senses the mechanical deformation of the winding through FBGs, reducing measurement errors. The fiber Bragg grating array is sensitive to static strain and facilitates monitoring of minute strain changes. Furthermore, the FBG array 210 in this embodiment can be completed before the transformer leaves the factory, and subsequent maintenance requires only remote diagnosis via wireless signals. Furthermore, the historical data storage and trend analysis functions of the data analysis unit 230 can provide early warning of potential faults, reducing unplanned downtime for repairs. Compared to resistance strain gauge solutions, this significantly reduces operation and maintenance costs.

[0066] Below, we will refer to the attached Figure 1The specific structure of the strain monitoring system for transformer windings provided in an embodiment of the present application is introduced in detail.

[0067] like Figure 1 As shown, in some embodiments, a fiber grating array 210 is disposed on the surface of the transformer winding 100. Specifically, the fiber grating array 210 can be directly adhered to the outer surface of the transformer winding 100 using high-temperature resistant epoxy resin adhesive. The fiber grating array 210 is laid along the winding direction, and the spacing between adjacent gratings is set according to the monitoring accuracy requirements (typically 5-10 cm). At the winding ends (stress concentration areas), a denser arrangement is adopted, and the spacing can be reduced to 3-5 cm. The ends of the fiber grating array 210 are fixed to the insulating bracket at the winding end via a fiber optic fixing base (made of polytetrafluoroethylene). The fixing base and the winding are bonded with insulating adhesive to ensure good tracking of the optical fiber when the winding deforms.

[0068] The signal processor 220 is connected to the fiber grating array 210 through an optical fiber jumper. The optical fiber jumper is led out from the reserved threading hole of the transformer oil tank. The threading hole is sealed with a sealing joint (made of nitrile rubber) to prevent transformer oil leakage.

[0069] In this embodiment, the fiber Bragg grating array 210 directly contacts the winding surface, enabling real-time detection of subtle deformations. This highly efficient strain transfer method improves measurement accuracy compared to indirect measurement methods. Furthermore, the internal structure of the winding does not need to be altered, making it suitable for retrofitting and upgrading existing transformers. The length and number of gratings in the fiber Bragg grating array 210 can be flexibly adjusted based on the winding size, achieving full coverage of the winding and avoiding monitoring blind spots.

[0070] In one example, the transformer winding 100 includes an insulation structure including an insulation spacer and a strain monitoring device 200 embedded in the insulation spacer.

[0071] Specifically, the insulation structure of the transformer winding 100 includes an insulating spacer (made of cardboard or epoxy resin), which is arranged between the turns or cakes of the winding to support the winding and provide insulation isolation. During the manufacturing process of the insulating spacer, the fiber grating array 210 is embedded therein, including: when the insulating spacer is prefabricated, a fiber channel is reserved in the mold, the channel extends along the length direction of the insulating spacer, and maintains a distance of, for example, 2-3 mm from the surface of the spacer. Then, the fiber grating array 210 is inserted into the channel, and the position of the grating corresponds to the stress point of the insulating spacer (such as the two ends and the middle position of the spacer). Subsequently, a small amount of epoxy resin glue is injected into the channel, and after curing, the fiber grating array 210 is firmly bonded to the insulating spacer. When the insulating spacer is installed on the winding, ensure that the embedded fiber grating array 210 is facing the stress direction of the winding to improve the strain capture sensitivity.

[0072] In this embodiment, the insulating spacer provides mechanical protection for the fiber Bragg grating array 210, preventing wear and compression of the optical fiber during winding assembly and operation, thereby increasing the fiber's service life. Furthermore, the integrated design of the fiber Bragg grating array 210 and the insulating spacer reduces interference from external vibrations on the measurement, thereby improving the stability of the measured data. Furthermore, the fiber Bragg grating embedded within the insulating structure can sense the strain state within the winding, more closely resembling actual stress conditions and resulting in more accurate measurement results.

[0073] In some embodiments, the signal processor 220 includes a wavelength demodulation module, which is used to convert the wavelength change of each fiber Bragg grating center in the fiber Bragg grating array 210 into an electrical signal.

[0074] Specifically, the wavelength demodulation module in signal processor 220 includes a broadband light source, an optical coupler, a spectrometer, and a data acquisition card. The broadband light source utilizes an erbium-doped fiber amplifier (EDFA) with a central output wavelength of 1550 nm, a bandwidth of 40 nm, and an output power of 10-20 mW. The optical coupler is a 3 dB bidirectional coupler, one end of which is connected to the broadband light source, one end to the fiber Bragg grating array 210, and the other end to the spectrometer.

[0075] The spectrometer uses a high-resolution grating spectrometer with a resolution of 0.1 μm and a sampling frequency of 100 Hz, capable of real-time acquisition of spectral information from the light reflected by the fiber Bragg grating array 210. A data acquisition card converts the analog signal output by the spectrometer into a digital signal and transmits it to the main control unit of the signal processor 220 via a USB interface.

[0076] The working process of the wavelength demodulation module can be that the optical signal emitted by the broadband light source enters the fiber grating array 210 through the optical coupler, each fiber grating reflects the optical signal of a specific wavelength, and the reflected light returns to the spectrometer through the optical coupler. The spectrometer splits and detects the reflected light to obtain the central wavelength information of each grating. The data acquisition card converts the wavelength information into an electrical signal and outputs it to the data analysis unit 230.

[0077] In this embodiment, the wavelength resolution of 0.1 pm corresponds to a strain resolution of 0.08 με, capable of capturing minute strain changes in the windings and meeting high-precision monitoring requirements. The 100 Hz sampling frequency enables real-time tracking of dynamic changes in winding strain, accurately capturing strain fluctuations caused by transient electromagnetic forces. Furthermore, the all-optical signal processing reduces the impact of electromagnetic interference on the signal, improving the signal-to-noise ratio by over 20 dB compared to electrical signal transmission.

[0078] In one example, the data analysis unit 230 includes a strain calculation module, which is configured to calculate the strain magnitude of the transformer winding 100 according to the electrical signal.

[0079] Specifically, the strain calculation module of data analysis unit 230 is built into an embedded processor and implements strain calculation functionality through a software program. The core algorithm of the strain calculation module is based on the strain-wavelength characteristic formula of fiber Bragg gratings: ε = (Δλ / λ0 - KTΔT) / Kε, where Δλ is the change in the grating's central wavelength, λ0 is the grating's initial central wavelength, KT is the temperature sensitivity coefficient, ΔT is the temperature change, and Kε is the strain sensitivity coefficient (1.2 pm / με).

[0080] The workflow of the strain calculation module may be that, after receiving the wavelength-changing electrical signal output by the signal processor 220, temperature compensation is first performed (ΔT is obtained through the built-in temperature sensor data or temperature compensation grating data), and then the strain value is calculated by substituting the above formula. Finally, the strain values ​​of multiple monitoring points are statistically analyzed to generate the strain distribution data of the winding.

[0081] In addition, the strain calculation module also has a data calibration function, which can correct the nonlinear error of the fiber Bragg grating through the preset calibration coefficient to ensure the accuracy of strain calculation.

[0082] This embodiment uses temperature compensation and nonlinear correction to control strain measurement errors within ±5με, improving the accuracy of strain calculations. Furthermore, the high-speed computing power of the embedded processor ensures strain calculations are completed within 1ms, meeting the time requirements of real-time monitoring. Furthermore, wavelength changes are converted into intuitive strain values, facilitating subsequent data analysis and fault diagnosis.

[0083] In some embodiments, the transformer winding 100 has multiple monitoring regions, each corresponding to a different stress. Each monitoring region is provided with a strain monitoring device 200, and the arrangement density of the fiber Bragg gratings in the fiber Bragg grating array 210 within each monitoring region is different. In one example, the transformer winding 100 has a first stress monitoring region and a second stress monitoring region, and the arrangement density of the fiber Bragg gratings within the first stress monitoring region is greater than the arrangement density of the fiber Bragg gratings within the second stress monitoring region. The stress corresponding to the first stress monitoring region is greater than the stress corresponding to the second stress monitoring region. The spacing between adjacent fiber Bragg gratings within the first stress monitoring region is between 3 cm and 5 cm, and the spacing between adjacent fiber Bragg gratings within the second stress monitoring region is between 10 cm and 15 cm.

[0084] It is understood that based on the stress distribution characteristics of the transformer winding 100, the winding is divided into multiple monitoring areas: the first stress monitoring area (high stress area) includes stress concentration areas such as the winding ends, lead wire connections, and insulation spacer support points; the second stress monitoring area (low stress area) includes areas with lower stress, such as the straight section in the middle of the winding. In the first stress monitoring area, the fiber Bragg gratings are arranged at a high density, with adjacent gratings spaced 3-5 cm apart, for example, 3 cm, 4 cm, 5 cm, etc., to ensure detailed capture of the strain distribution details in the stress concentration areas. In the second stress monitoring area, adjacent gratings are spaced 10-15 cm apart, for example, 10 cm, 11 cm, 12 cm, 13 cm, 14 cm, 15 cm, etc., to reduce the number of gratings while ensuring monitoring coverage and lowering costs.

[0085] The fiber Bragg grating array 210 is laid out in a segmented design. The gratings in the high stress area and the low stress area are connected in series through the same optical fiber. The gratings in different areas are distinguished by different initial center wavelengths (with intervals of more than 2nm) to avoid wavelength overlap interference.

[0086] This embodiment reduces system costs while ensuring monitoring accuracy in critical areas by increasing the number of monitoring points in high-stress areas and reducing them in low-stress areas. Furthermore, the denser monitoring points in high-stress areas accurately capture the location and extent of strain anomalies, facilitating precise fault location and analysis. Furthermore, the varying density of monitoring points balances monitoring accuracy and breadth, fully reflecting the overall strain state of the winding.

[0087] In some embodiments, the strain monitoring system further includes a high-frequency current transformer, which is connected to the transformer winding 100 and is used to collect partial discharge signals of the transformer winding 100 .

[0088] Specifically, the high-frequency current transformer uses a Rogowski coil current transformer, capable of collecting high-frequency partial discharge current signals from the transformer winding 100. The high-frequency current transformer is installed at the lead end of the transformer winding 100. Its primary side passes through the winding lead, and its secondary side is connected to the partial discharge signal processing unit of the signal processor 220 via a coaxial cable. The data analysis unit 230 correlates and analyzes the partial discharge signal with the strain signal. When the partial discharge exceeds a threshold and is accompanied by a sudden increase in strain, it is determined to be a potential fault.

[0089] This embodiment combines partial discharge signals with strain signals to achieve multi-dimensional diagnosis of winding faults, improving fault identification accuracy. Partial discharge is often an early sign of winding insulation aging. Combined with strain monitoring, it can provide early warning of potential faults 3-6 months in advance. Furthermore, correlation analysis between partial discharge and strain can distinguish between faults caused by insulation aging and excessive mechanical stress, facilitating targeted maintenance.

[0090] In one example, the strain monitoring system further includes a temperature sensor, which is disposed on the transformer winding 100 and is used to obtain the temperature of the transformer winding 100 .

[0091] Specifically, the temperature sensor uses a platinum resistance temperature sensor with a measurement range of -50°C to 200°C. The temperature sensors are installed at different heights (top, middle, and bottom) of the winding and connected to the signal processor 220 via high-temperature-resistant wires. Within the fiber grating array 210, a portion of the gratings are selected as temperature compensation gratings (unaffected by strain and sensitive only to temperature), forming a dual temperature monitoring system with the temperature sensor. The signal processor 220 transmits the temperature sensor measurement data to the data analysis unit 230 for temperature compensation in strain calculations.

[0092] The data analysis unit 230 establishes a temperature-strain correlation model and automatically adjusts the strain monitoring threshold when the temperature change rate exceeds, for example, 5°C / h to prevent thermal expansion and contraction caused by temperature from being misjudged as fault strain.

[0093] Dual temperature monitoring in this embodiment ensures accurate temperature data, helping to reduce strain measurement errors after temperature compensation. It also adapts to temperature fluctuations during transformer operation, maintaining monitoring accuracy over a wide temperature range. Combining temperature and strain data allows differentiation between winding strain caused by mechanical forces and temperature fluctuations, improving the accuracy of fault diagnosis.

[0094] In some embodiments, the strain monitoring system also includes a fault warning model module; the fault warning model module is connected to the wireless transmission unit 240, and the fault warning model module is used to compare the strain information of the current transformer winding 100 with the pre-stored historical strain data to warn of the fault type of the transformer winding 100.

[0095] It is understood that the fault warning model module in this embodiment uses a deep learning-based neural network model and is deployed on an edge computing server or cloud platform. The training data for the fault warning model includes historical strain data, temperature data, partial discharge data of the transformer winding 100, and corresponding fault records.

[0096] In one example, the fault warning model module is connected to the wireless transmission unit 240, the high-frequency current transformer, and the temperature sensor, respectively. The fault warning model module is used to integrate the strain information, partial discharge data, and temperature data of the current transformer winding 100, and compare them with the pre-stored historical parameter data to issue a warning on the fault type of the transformer winding 100.

[0097] More specifically, the fault warning model module can operate as follows: after receiving the current strain information, temperature data, and partial discharge data sent by the wireless transmission unit 240, it is input into a pre-trained neural network model. The fault warning model then outputs the failure probability and remaining life prediction value of the winding. When the failure probability exceeds a set threshold, a warning signal is issued and pushed to the monitoring center via the wireless transmission unit 240. The fault warning model in this embodiment has a self-learning function, which can continuously optimize model parameters based on new operating data to improve warning accuracy. At the same time, the fault warning model stores characteristic maps of different types of faults and can identify various fault types such as inter-turn short circuits, insulation aging, and winding deformation.

[0098] In this embodiment, the deep learning-based model can mine potential correlations in data, helping to improve the accuracy of fault warnings. By analyzing strain trends over time, the remaining life of the winding can be predicted, providing a basis for planned maintenance and reducing unplanned downtime. Furthermore, it can accurately identify different types of faults, guiding maintenance personnel to take targeted repair measures and improving maintenance efficiency.

[0099] Based on the same application concept, the embodiment of the present application also provides a strain monitoring method for transformer windings, such as Figure 2 As shown, Figure 2 This is a flow chart of a strain monitoring method for transformer windings according to some embodiments of the present application. The strain monitoring method includes:

[0100] Step S101, establishing a stress identification simulation model, and generating stress distribution data of the transformer winding 100 based on a design model of the transformer winding 100;

[0101] Step S102, calibrating multiple monitoring areas of the transformer winding 100 with different stresses based on the stress distribution data, and designing fiber grating arrays 210 with different arrangement densities in each monitoring area;

[0102] Step S103: Based on the collected strain information, partial discharge data and temperature data of the transformer winding 100, a multi-parameter fault warning model is established, and multiple parameter data are analyzed to identify the abnormal type of the transformer winding 100, and a warning signal is sent to the monitoring center in real time.

[0103] It is understood that in step S101, the design parameters of the transformer winding 100, including the winding material, structural dimensions, winding method, rated voltage, rated current, etc., can be collected. At the same time, the boundary conditions of the transformer during operation are obtained, such as the number of winding turns required for electromagnetic force calculation, the magnetic permeability of the core, and the heat dissipation coefficient related to thermal stress. Then, finite element simulation software, such as ANSYS Maxwell and Mechanical coupling simulation, is set up. First, an electromagnetic simulation module is constructed in Maxwell to simulate the magnetic field distribution when current flows through the winding, calculate the radial electromagnetic force and the axial electromagnetic force, and obtain a distribution cloud map of the electromagnetic force on the winding; the electromagnetic force results are imported into the Mechanical module, and combined with gravity and thermal stress, a multi-physics field coupled stress simulation model is established; simulation parameters are set, including a mesh size of 5mm in the stress concentration area of ​​the winding end and 10mm in other areas; the solver uses an implicit dynamics algorithm, and the simulation time covers typical operating conditions such as no-load, rated load, and short-circuit impact of the transformer.

[0104] The simulation model is run to output stress distribution data of the transformer winding 100 under different working conditions, including the maximum stress value, stress direction, and stress gradient. To ensure the accuracy of the data, physical experiments can be used for verification. This involves building a scaled-down winding model in the laboratory, applying equivalent electromagnetic force and thermal load, and using resistance strain gauges to measure stress at key points (such as end corners). The test data is compared with the simulation results, and the model parameters are corrected to ultimately generate a stress distribution database with an error of ≤10% from the actual working conditions.

[0105] In this embodiment, the simulation model can quantify the stress differences in various parts of the winding, clearly identify high-stress areas (such as ends and lead connections) and low-stress areas (such as the middle straight section), provide data support for subsequent monitoring area division, and avoid blind placement of monitoring points.

[0106] In step S102, based on the stress distribution data generated in step S101, stress thresholds are set to divide monitoring areas, including marking the boundaries of each area on the winding digital model using 3D modeling software (such as SolidWorks) to generate a visual monitoring area distribution map.

[0107] All gratings are coated with polyimide. The fiber routing avoids obstructions such as winding terminals. The entire array is converged to the signal processor 220 via four main optical fibers, each corresponding to a quadrant of the winding area. In the first stress monitoring zone, the fiber Bragg gratings are arranged with a spacing of 3-5 cm between adjacent gratings. The grating orientation aligns with the primary stress direction (e.g., if radial stress is dominant at the end, the gratings are arranged radially). The gratings are embedded in insulating spacers (refer to the aforementioned insulation structure embedding scheme). In the second stress monitoring zone, adjacent gratings are spaced 10-15 cm apart and are surface-mounted (using high-temperature resistant epoxy adhesive). The gratings are arranged axially to capture axial strain.

[0108] In this embodiment, denser monitoring points in high-stress areas capture subtle stress changes, while the number of gratings is reduced in low-stress areas. While maintaining monitoring accuracy, this reduces the total number of gratings compared to a uniformly distributed arrangement across the entire area, lowering system costs. Precise monitoring of stress concentration areas reduces missed detection rates for faults such as turn-to-turn short circuits and loose insulation spacers, and in particular, enables real-time detection of early deformation of winding ends.

[0109] In step S103, the collected data undergoes preprocessing, including outlier removal, time synchronization, and normalization. Strain information is collected by the fiber Bragg grating array 210 and demodulated by the signal processor 220 to obtain strain values ​​at each monitoring point. Partial discharge data is collected by a high-frequency current transformer and processed to obtain partial discharge pulse signals. Temperature data is collected by a temperature sensor at different winding heights and double-checked using the temperature sensitivity of the fiber Bragg grating.

[0110] Neural networks can be used to extract the spatial characteristics of partial discharge pulse waveforms and strain distribution maps, and historical data can be input to output fault types and failure probabilities. The model receives preprocessed multi-parameter data in real time. For example, when a single parameter exceeds the standard, a first-level warning is triggered; when two or more parameters are abnormally correlated, a second-level warning is triggered, and the model outputs the fault type. The warning signal is transmitted to the monitoring center via wireless transmission unit 240, and the original data is stored locally for traceability.

[0111] In this embodiment, multi-parameter fusion avoids misjudgment based on a single parameter, improving fault identification accuracy compared to single-strain monitoring. Furthermore, the model outputs fault type and location information, enabling maintenance personnel to conduct targeted inspections and reducing blind inspection time.

[0112] The strain monitoring method for transformer windings 100 provided in this embodiment, from simulation modeling to guide monitoring layout to multi-parameter fusion early warning, forms a closed-loop design-monitoring-diagnosis system. This upgrades transformer winding 100 strain monitoring from passive to active early warning, meeting the high requirements of new power systems for equipment status awareness. Furthermore, a differentiated monitoring design based on stress simulation reduces system costs while ensuring monitoring accuracy in high-stress areas; a multi-parameter model reduces ineffective operation and maintenance. Furthermore, early warning of winding faults can avoid sudden power outages (such as regional blackouts caused by transformer burnouts), reduce economic losses caused by faults, and improve grid power reliability.

[0113] In some embodiments, the transformer winding 100 includes a first stress monitoring area and a second stress monitoring area, and the stress corresponding to the first stress monitoring area is greater than the stress corresponding to the second stress monitoring area; in one example, step S103 may include step S1031, obtaining wavelength offset data of the fiber grating array 210, and obtaining a partial discharge pulse sequence and a temperature distribution map of the transformer winding 100 to construct a multi-parameter fault warning model.

[0114] Specifically, the wavelength demodulation module of the signal processor 220 collects the center wavelength offset of each fiber Bragg grating array 210 in real time for both the first stress monitoring region (high stress zone) and the second stress monitoring region (low stress zone). The gratings in the first stress monitoring region are spaced 3-5 cm apart, with data collected every 100 ms, focusing on capturing high-frequency strain fluctuations (such as transient deformation during short-circuit shock). The gratings in the second stress monitoring region are spaced 10-15 cm apart, with data collected every 500 ms, primarily monitoring steady-state strain (such as long-term creep).

[0115] A high-frequency current transformer is used to collect partial discharge signals from the transformer winding 100 to capture local discharges that are prone to occur in high-stress areas (such as corona discharge at the edges of insulating spacers). A temperature distribution map is constructed using temperature sensors (placed at the top, middle, and bottom of the winding) and the temperature-sensitive properties of fiber Bragg gratings (FBGs).

[0116] The model inputs data in the form of a three-dimensional tensor: the first dimension is the wavelength offset sequence of the first stress monitoring area, the second dimension is the wavelength offset sequence of the second stress monitoring area, and the third dimension is the fusion feature of the partial discharge pulse sequence and the temperature distribution map. An attention mechanism assigns higher weight to the data of the first stress monitoring area, strengthening the ability to identify anomalies in high-risk areas. During model training, typical fault data from different stress areas (such as loose insulation pads in the first area and overheated wires in the second area) is included to enable the model to distinguish region-specific faults.

[0117] In some embodiments, the abnormality type includes at least one of looseness, insulation degradation, and overheating; step S103 may also include step S1032, comparing multiple parameter data with pre-stored historical parameter data to output the abnormality type of the transformer winding 100; step S1033, integrating the multiple parameter data and generating a strain deviation; step S1034, comparing the strain deviation with multiple thresholds, and if the strain deviation exceeds any level of threshold, generating different levels of warning signals.

[0118] Specifically, normal parameter thresholds and fault samples of the transformer winding 100 under different operating conditions, including looseness, insulation degradation, and overheating, are pre-stored. Real-time multi-parameter data (strain, partial discharge, and temperature of the first and second regions) collected are compared with a historical database.

[0119] The multi-parameter data of the first stress monitoring area and the second stress monitoring area are aligned in time and space, integrated into a unified eigenvector, and the strain deviation is calculated based on the integrated eigenvector.

[0120] According to the safe operation requirements of the transformer winding 100 and the fault risks of the first stress monitoring area and the second stress monitoring area, three threshold levels are set: Level 1 threshold: strain deviation >5%, corresponding to slight overheating of the second stress monitoring area or slight loosening of the first stress monitoring area (no immediate risk); Level 2 threshold: strain deviation >15%, corresponding to the early stage of insulation degradation in the first stress monitoring area or significant overheating in the second stress monitoring area (needing attention); Level 3 threshold: strain deviation >30%, corresponding to serious loosening or insulation breakdown risk in the first stress monitoring area (emergency treatment).

[0121] Warning signal generation and transmission: If the strain deviation is greater than 5% and less than or equal to 15%, a Level 1 warning signal is generated, containing the message "The temperature in the second stress monitoring area is slightly high (or the first stress monitoring area is slightly loose), and closer observation is recommended." This signal is sent to the operation and maintenance personnel via the wireless transmission unit 240 in the form of a text message.

[0122] If the deviation is greater than 15% and less than or equal to 30%, a secondary warning signal is generated, along with the abnormality type (e.g., "insulation degradation in the first stress monitoring area") and strain distribution map, and is pushed to the monitoring center's audio-visual alarm system via the 4G module.

[0123] If the deviation is greater than 30%, a level 3 warning signal will be generated, triggering an emergency shutdown recommendation, uploading real-time data to the cloud for backup, and automatically calling the person in charge of operations and maintenance.

[0124] In this embodiment, thresholds are set based on the risk differences between the first and second stress monitoring areas to avoid a one-size-fits-all warning. Level 1 warnings reduce unnecessary emergency responses, while level 3 warnings ensure timely handling of high-risk faults. Different levels of warning correspond to differentiated processing flows, focusing O&M resources on high-priority faults and shortening the mean fault handling time. Emergency warnings for high-risk faults in the first stress monitoring area prevent escalation and reduce the incidence of sudden transformer failures.

[0125] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0126] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A strain monitoring system for transformer windings, characterized in that: The strain monitoring system comprises a transformer winding (100) and a strain monitoring device (200); Wherein, the strain monitoring device (200) comprises: A fiber grating array (210) is provided on the transformer winding (100), and the fiber grating array (210) is used to capture the strain signal of the transformer winding (100); a signal processor (220) connected to the fiber grating array (210), the signal processor (220) being used to receive and process the strain signal to generate an electrical signal; a data analysis unit (230), connected to the signal processor (220), the data analysis unit (230) being used to receive and analyze the electrical signal to obtain strain information of the transformer winding (100); A wireless transmission unit (240) is connected to the data analysis unit (230), and the wireless transmission unit (240) is used to wirelessly transmit the strain information to a monitoring center.

2. The strain monitoring system for transformer windings according to claim 1, characterized in that: The fiber grating array (210) is provided on the surface of the transformer winding (100); And / or, the transformer winding (100) comprises an insulating structure, the insulating structure comprising an insulating spacer and the strain monitoring device (200) embedded in the insulating spacer.

3. The strain monitoring system for transformer windings according to claim 2, characterized in that: The signal processor (220) comprises a wavelength demodulation module, and the wavelength demodulation module is used to convert the wavelength change of each fiber grating center in the fiber grating array (210) into the electrical signal; And / or, the data analysis unit (230) includes a strain calculation module, the strain calculation module being used to calculate the strain magnitude of the transformer winding (100) based on the electrical signal.

4. The strain monitoring system for transformer windings according to any one of claims 1 to 3, characterized in that: The transformer winding (100) has a plurality of monitoring areas, each of the monitoring areas corresponding to a different stress; each of the monitoring areas is provided with the strain monitoring device (200), and the arrangement density of the individual fiber gratings in the fiber grating array (210) in each of the monitoring areas is different.

5. The strain monitoring system for transformer windings according to claim 4, characterized in that: The transformer winding (100) has a first stress monitoring area and a second stress monitoring area, the arrangement density of the optical fiber gratings in the first stress monitoring area is greater than the arrangement density of the optical fiber gratings in the second stress monitoring area; wherein the stress corresponding to the first stress monitoring area is greater than the stress corresponding to the second stress monitoring area; The spacing between adjacent fiber Bragg gratings in the first stress monitoring area is between 3 cm and 5 cm, and the spacing between adjacent fiber Bragg gratings in the second stress monitoring area is between 10 cm and 15 cm.

6. The strain monitoring system for transformer windings according to any one of claims 1 to 3, characterized in that: The strain monitoring system further comprises a high-frequency current transformer, which is connected to the transformer winding (100) and is used to collect a partial discharge signal of the transformer winding (100); And / or, the strain monitoring system further comprises a temperature sensor, the temperature sensor being provided on the transformer winding (100) and being used to obtain the temperature of the transformer winding (100).

7. The strain monitoring system for transformer windings according to claim 6, characterized in that: The strain monitoring system also includes a fault warning model module; The fault warning model module is connected to the wireless transmission unit (240), and the fault warning model module is used to compare the current strain information of the transformer winding (100) with pre-stored historical strain data to provide a warning of the fault type of the transformer winding (100); Alternatively, the fault warning model module is connected to the wireless transmission unit (240), the high-frequency current transformer, and the temperature sensor, respectively, and the fault warning model module is used to integrate the strain information, partial discharge data, and temperature data of the current transformer winding (100), and compare them with pre-stored historical parameter data to provide a warning of the fault type of the transformer winding (100).

8. A strain monitoring method for transformer windings, characterized in that: The strain monitoring method comprises: Establishing a stress identification simulation model and generating stress distribution data of the transformer winding based on a design model of the transformer winding; Calibrate multiple monitoring areas of the transformer winding with different stresses based on the stress distribution data, and design fiber Bragg grating arrays with different arrangement densities in each of the monitoring areas; Based on the collected strain information, partial discharge data and temperature data of the transformer winding, a multi-parameter fault warning model is established, and multiple parameter data are analyzed to identify the abnormal type of the transformer winding and send a warning signal to the monitoring center in real time.

9. The strain monitoring method for transformer windings according to claim 8, characterized in that: The transformer winding includes a first stress monitoring area and a second stress monitoring area, and the stress corresponding to the first stress monitoring area is greater than the stress corresponding to the second stress monitoring area; The step of establishing a multi-parameter fault warning model based on the collected strain information, partial discharge data and temperature data of the transformer winding includes: The wavelength offset data of the fiber grating array is obtained, and the partial discharge pulse sequence and temperature distribution spectrum of the transformer winding are obtained to construct the multi-parameter fault warning model.

10. The strain monitoring method for transformer windings according to claim 8, characterized in that: The abnormality type includes at least one of looseness, insulation degradation, and overheating; The step of analyzing multiple parameter data to identify the abnormal type of the transformer winding and sending an early warning signal to a monitoring center in real time includes: Comparing the plurality of parameter data with pre-stored historical parameter data to output an abnormality type of the transformer winding; Integrating the plurality of parameter data and generating a strain deviation; The strain deviation is compared with multiple thresholds. If the strain deviation exceeds any threshold, warning signals of different levels are generated.

Images