End screen acquisition unit, signal detection method and oil-immersed bushing

Abstract

The invention relates to an end screen acquisition unit, a signal detection method and an oil-immersed bushing, the end screen acquisition unit comprises a sensing module and a signal processing module electrically connected to the sensing module, the sensing module comprises at least two induction coils configured for different frequency bands, and all the induction coils are arranged in a manner of sharing a magnetic circuit; the signal processing module is used for carrying out independent sampling and bandwidth partitioning on current signals of the induction coils; carrying out weight calculation on the current signals of different frequency bands; and carrying out dynamic weighted fusion and spectrum reconstruction on the frequency domain signal based on a weight calculation result. According to the application, the at least two induction coils configured for different frequency bands are arranged, and the signal processing module for performing independent acquisition and digital fusion processing on output signals of the induction coils is matched, so that high-fidelity acquisition of current signals in a broadband range is realized; the problem of measurement signal distortion caused by simple signal processing of a multi-coil sensor in a complex environment of an oil-immersed bushing end screen is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of high-voltage equipment online monitoring technology, in particular to a terminal screen acquisition unit, a signal detection method and an oil immersed bushing. BACKGROUND

[0002] In the field of high-voltage power equipment online monitoring, the insulation state evaluation of the oil immersed bushing is crucial. The terminal screen terminal provides a standard interface for extracting the operating current signal, and is a key position for monitoring fault phenomena such as partial discharge and transient overvoltage. However, the operating environment of the terminal screen area of the oil immersed bushing is special and harsh, with high electric field strength, strong electromagnetic interference, limited space, and long-term immersion in insulating oil, temperature cycling and mechanical vibration.

[0003] To meet the wideband measurement requirements, a Rogowski coil sensor based on a multi-coil superposition scheme is usually used. The sensor integrates multiple coils with different magnetic core materials to cover different frequency bands for synthesizing a wideband signal. The signal processing uses simple linear superposition, which cannot dynamically compensate for the amplitude and phase distortion of each frequency band signal, resulting in distortion of the fused signal at the frequency band junction, making it difficult to achieve high-fidelity measurement. SUMMARY

[0004] Therefore, it is necessary to provide a terminal screen acquisition unit, a signal detection method and an oil immersed bushing to solve the problem of measurement signal distortion caused by the simple signal processing of the multi-coil superposition Rogowski coil sensor in the complex environment of the terminal screen of the oil immersed bushing.

[0005] In a first aspect, the present application provides a terminal screen acquisition unit, which adopts the following technical solution:

[0006] A terminal screen acquisition unit, comprising a sensing module and a signal processing module, the sensing module is used for sensing the current signal flowing through the terminal screen; the sensing module comprises at least two induction coils configured for different frequency bands, all the induction coils are configured as a common magnetic circuit; the signal processing module is electrically connected with the sensing module, and is used for processing the current signal; wherein the signal processing module is configured to perform the following operations: independently sampling and bandwidth partitioning the current signal of each induction coil; weight calculation for current signals of different frequency bands; dynamic weighted fusion and spectrum reconstruction of frequency domain signals based on the weight calculation result.

[0007] In one of the embodiments, the dynamic weighting fusion of the frequency domain signals based on the weighting calculation result comprises: performing sliding window analysis on the common frequency band of the frequency band boundary area, calculating the amplitude gradient and the phase difference; when the amplitude gradient and / or the phase difference exceeds a preset threshold, adjusting the weight coefficient of the corresponding frequency band; and using a polynomial smoothing fitting algorithm to generate a continuous transition curve in the overlapping frequency band area.

[0008] In one of the embodiments, all the induction coils comprise low-frequency coils, medium-frequency coils and high-frequency coils arranged in sequence, the low-frequency coils, the medium-frequency coils and the high-frequency coils are configured as a common magnetic circuit; the low-frequency coils adopt soft magnetic materials with high saturation magnetic induction intensity and low magnetic loss; the medium-frequency coils adopt composite soft magnetic materials with medium magnetic permeability and wide frequency response characteristics; and the high-frequency coils adopt low-dielectric-constant, low-loss high-frequency dielectric substrates or air core structures.

[0009] In one of the embodiments, at least one of the low-frequency coils, the medium-frequency coils and the high-frequency coils adopts a temperature-stable magnetic core material, the temperature-stable magnetic core material is configured to have a magnetic permeability change amplitude of less than ±3% within a temperature range of -40°C to 120°C.

[0010] In one of the embodiments, the electromagnetic isolation structure is further arranged between any two of the induction coils; the electromagnetic isolation structure comprises at least one or more of any combination of a non-magnetic isolation ring, an air gap and an insulating layer.

[0011] In one of the embodiments, the end screen acquisition unit further comprises a packaging shell for accommodating the sensing module and the signal processing module; the packaging shell is configured to be integrally cast by aluminum alloy.

[0012] In one of the embodiments, the end screen acquisition unit further comprises a conductive sliding block, a flow guide and a conductive back cover connected in sequence; the flow guide is configured to be able to stretch and contract along the connection direction of the conductive sliding block and the conductive back cover.

[0013] In a second aspect, the application provides a signal detection method, which comprises the following steps:

[0014] A signal detection method applied to the end screen acquisition unit, the signal detection method comprising the following steps:

[0015] Independently collecting current signals of different frequency bands by the plurality of induction coils;

[0016] Independently digitizing and sampling the current signals of the induction coils to form independent digital signals;

[0017] Bandwidth partitioning and fast Fourier transform are performed on each of the digital signals to obtain corresponding spectral components;

[0018] Based on preset calibration data and real-time calculated signal-to-noise ratio, a weight coefficient is dynamically assigned to each of the spectral components of each frequency band;

[0019] The spectral components are weighted and fused based on the weight coefficient and reconstructed into a spectrum;

[0020] The reconstructed spectrum is inversely transformed to output a time-domain fused signal of the full frequency band.

[0021] In one embodiment, the weighting and fusing of the spectral components based on the weight coefficient comprises:

[0022] The spectral components at the frequency band boundary are subjected to amplitude gradient and phase difference analysis;

[0023] When the analysis result exceeds a set threshold, the weight coefficient is adaptively adjusted;

[0024] A polynomial smoothing fitting algorithm is used to ensure that the amplitude-frequency response at the frequency band boundary is first-order continuous and the phase-frequency response is zero-order continuous.

[0025] In a third aspect, the application provides an oil-immersed bushing, which uses the following technical solution:

[0026] An oil-immersed bushing includes a terminal end and the terminal end collecting unit described above, and the terminal end collecting unit is installed on the terminal end.

[0027] The terminal end collecting unit described above, by being provided with at least two induction coils configured for different frequency bands and cooperating with a signal processing module that independently collects and digitally fuses the output signals of each induction coil, realizes high-fidelity collection of current signals in a wide frequency band range, thereby solving the problem of distortion of measurement signals caused by simple signal processing of multi-coil sensors in the complex environment of the terminal end of an oil-immersed bushing. BRIEF DESCRIPTION OF DRAWINGS

[0028] Figure 1 It is a whole view of the terminal end collecting unit in one embodiment of the application.

[0029] Figure 2 It is a cross-sectional view of the terminal end collecting unit A-A in one embodiment of the application.

[0030] Figure 3 It is a flowchart of the algorithm of the signal processing unit in one embodiment of the application.

[0031] BRIEF DESCRIPTION OF DRAWINGS

[0032] 1, sensing module; 11, low-frequency coil; 12, medium-frequency coil; 13, high-frequency coil; 2, electromagnetic isolation structure; 3, packaging shell; 31, shell; 32, cover; 4, conductive slider; 5, flow guide; 6, conductive back cover; 7, thimble; 8, center conductive rod; 9, end screen grounding lead. DETAILED DESCRIPTION

[0033] In order to make the above objectives, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below in conjunction with the drawings. In the following description, a large number of specific details are set forth in order to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without departing from the spirit of the present application, so the present application is not limited to the specific embodiments disclosed below.

[0034] In the description of the present application, it should be understood that if these terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like appear, these terms indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the purpose of facilitating the description of the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.

[0035] In addition, if these terms "first", "second" appear, these terms are only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features limited by "first", "second" can explicitly or implicitly include at least one of the features. In the description of the present application, if the term "a plurality of" appears, the meaning of "a plurality of" is at least two, such as two, three, etc., unless otherwise explicitly specified.

[0036] In the present application, unless otherwise explicitly specified and limited, if the terms "mounting", "connecting", "connecting", "fixing" and the like appear, these terms should be understood in a broad sense. For example, it can be fixedly connected, or it can be detachably connected, or it can be integrated; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium; it can be the internal communication of two elements or the interaction relationship between two elements, unless otherwise explicitly limited. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0037] In the present application, unless specifically stated and limited otherwise, if there is a description of a first feature "on" or "under" a second feature, it can mean that the first and second features are in direct contact, or the first and second features are indirectly in contact through an intermediate medium. Moreover, the first feature "on", "above" and "over" the second feature can be that the first feature is directly above or obliquely above the second feature, or only indicates that the first feature is higher in horizontal height than the second feature. The first feature "under", "below" and "under" the second feature can be that the first feature is directly below or obliquely below the second feature, or only indicates that the first feature is lower in horizontal height than the second feature.

[0038] If an element is referred to as being "fixed" or "attached" to another element, it can be directly on the other element or there can be an intervening element. If an element is referred to as being "connected" to another element, it can be directly connected to the other element or intervening elements can be present. As used in this application, the terms "vertical", "horizontal", "up", "down", "left", "right", and similar expressions are for illustrative purposes only and are not meant to be limiting.

[0039] In the field of online monitoring of high-voltage power equipment, the insulation state evaluation of oil-immersed bushings is crucial. The terminal of the end shield provides a standard interface for extracting the operating current signal and is a key position for monitoring fault phenomena such as partial discharge and transient overvoltage. However, the operating environment of the end shield region of the oil-immersed bushing is complex, mainly in the following aspects:

[0040] (1) The amplitude of the partial discharge signal is weak, usually below the millivolt level, and is easily overwhelmed by power frequency and harmonic noise;

[0041] (2) The rising edge of the transient overvoltage is steep, usually on the order of nanoseconds, requiring a high upper limit of the frequency response of the sensor;

[0042] (3) The end shield region is narrow in space and has a complex oil-paper insulation structure, with high electric field strength and severe electromagnetic coupling;

[0043] (4) Over a long period of operation, oil temperature changes and mechanical vibrations cause parameter drift and sealing failure of the sensing coil.

[0044] To meet the wideband measurement requirements, existing technologies have developed a multi-coil superposition scheme for Rogowski coil sensors. This type of sensor arranges multiple sensing coils with different parameters, each optimized for a different frequency band. For example, by increasing the number of turns of the low-frequency coil and using a high-permeability magnetic core, the low-frequency response sensitivity is improved, while by reducing the number of turns of the high-frequency coil and using a special structure, the high-frequency response range is expanded. The output signals of each coil are processed by their respective integration circuits and then simply linearly superimposed, in an attempt to achieve continuous frequency band coverage from power frequency to high frequency.

[0045] However, this traditional method has significant technical limitations. The sensitivity and phase response characteristics of different coils differ, and simple linear superposition introduces amplitude errors and phase distortions at frequency band boundaries. As the number of coils increases, the mutual inductance and distributed capacitance between them become more pronounced, exacerbating signal coupling and noise accumulation problems.

[0046] At the signal processing level, the fixed "integration + summation" algorithm lacks a dynamic compensation mechanism for the sensitivity, phase difference and bandwidth differences of each coil. In particular, signal attenuation or distortion is prone to occur in the high-frequency and mid-frequency boundary region, resulting in discontinuous frequency band splicing and distorted measurement curves.

[0047] Specifically, the shortcomings of the Rogowski coil sensor with the aforementioned multi-coil stacking scheme in the end-screen area are mainly reflected in the following aspects:

[0048] (1) In an oil-immersed, high-electric-field environment, the magnetic core is susceptible to parasitic capacitance and eddy current. Its signal processing adopts a simple linear superposition method, which cannot dynamically compensate for the amplitude and phase distortion of each frequency band signal. This results in distortion of the fused signal at the frequency band boundary, making it difficult to achieve true high-fidelity measurement.

[0049] (2) Insufficient high-frequency response, usually cut off below 200kHz, making it difficult to capture partial discharge and transient signals;

[0050] (3) In terms of mechanical structure, the insulation and packaging structure of existing sensors often fails to fully consider the long-term compatibility with oil media. Under continuous oil immersion, the packaging material (oil paper) may expand or deteriorate, affecting the sealing performance.

[0051] Meanwhile, due to the lack of targeted design for long-term mechanical stress, the internal lead wire connection points are prone to loosening under temperature cycling and mechanical vibration, affecting long-term operational reliability.

[0052] Therefore, the core problem to be solved in this application is: how to achieve high-fidelity detection of current signals across the entire frequency band (from power frequency to MHz level) in the high electric field, strong interference, and narrow space environment of an oil-immersed bushing end screen, while taking into account anti-interference, sealing performance, and structural stability.

[0053] Based on the above requirements, this application provides a terminal screen acquisition unit, a signal detection method applied to the unit, and an oil-immersed bushing containing the unit. The following is in conjunction with the appendix... Figures 1-3 The embodiments of this application will be described in further detail.

[0054] See Figure 1 and Figure 2 As shown, Figure 1 This paper shows an overall diagram of the end-screen acquisition unit in one embodiment of this application.Figure 2 A cross-sectional schematic diagram along the AA direction of the end-screen acquisition unit in one embodiment of this application is shown.

[0055] This application provides a terminal screen acquisition unit, including a housing 3 and a sensing module 1 and a signal processing module (not shown) encapsulated within the housing 3. The sensing module 1 is used to sense the current signal flowing through the terminal screen area. The signal processing module is electrically connected to the sensing module 1 and is used to process the sensed current signal. In this embodiment, the sensing module 1 adopts a multi-coil common magnetic circuit layout structure, including multiple induction coils set for different frequency bands. All induction coils are constructed to share the same magnetic circuit path to improve the consistency of signal acquisition across different frequency bands.

[0056] In some embodiments, the sensing module 1 may include two induction coils (not shown), optimized for low and high frequencies respectively, suitable for applications where mid-frequency response requirements are not high.

[0057] by Figure 2 As shown in the example, in this embodiment of the application, the sensing module 1 includes three induction coils, specifically a low-frequency coil 11, a medium-frequency coil 12, and a high-frequency coil 13 arranged sequentially. These three induction coils are arranged in a coaxial nested manner to form a compact common magnetic circuit structure. Among them, the low-frequency coil 11 is located in the innermost layer and is directly fitted onto the outside of the central conductive rod 8.

[0058] In this embodiment, the low-frequency coil 11 uses a soft magnetic material with high saturation magnetic induction (Bs≥1.2T) and low magnetic loss. Such soft magnetic materials include, but are not limited to, microcrystalline alloys, high-permeability magnetic-based soft magnetic alloys, iron-nickel alloys, or other high-magnetic-induction magnetic core composite materials. These materials have high permeability and saturation magnetic induction, which can effectively improve the detection sensitivity of power frequency and low-frequency components, while suppressing noise interference.

[0059] The intermediate frequency coil 12 is sleeved around the low frequency coil 11, maintaining a certain distance from the outer wall of the low frequency coil 11. The intermediate frequency coil 12 is made of a composite soft magnetic material with moderate permeability and wide frequency response characteristics, including but not limited to any one of the following: a multilayer structure composed of amorphous and nanocrystalline materials, a magnetic ceramic composite core, or an equivalent soft magnetic multilayer structure. This composite soft magnetic material can achieve smooth response characteristics and linear output in the transition frequency band, effectively bridging the low-frequency and high-frequency measurement ranges.

[0060] The high-frequency coil 13 is located on the outermost layer and adopts a low dielectric constant, low-loss high-frequency dielectric substrate or an air core structure. Representative materials include, but are not limited to, high-frequency PTFE composite dielectric substrate, low-loss epoxy fiberglass board, or low-permeability non-magnetic material structure. This design avoids magnetic saturation under high-frequency conditions and maintains the waveform fidelity of high-frequency signals.

[0061] Specifically, the outer ring of the high-frequency coil 13 is positioned close to the inner wall of the package housing 3, and a certain distance is maintained between the inner ring of the high-frequency coil 13 and the outer ring of the intermediate frequency coil 12 to ensure the electromagnetic independence between each induction coil and avoid electromagnetic interference.

[0062] In this embodiment, the three layers of induction coils are sequentially nested along the axial direction to form a "sleeve-like nested" structure, arranged coaxially in layers, specifically with the low-frequency coil 11 at the innermost position, the mid-frequency coil 12 in the middle, and the high-frequency coil 13 at the outermost position. Specifically, the sequential nesting of the three layers of induction coils along the axial direction, with the low-frequency coil 11 at the innermost position, the mid-frequency coil 12 in the middle, and the high-frequency coil 13 at the outermost position, was determined through comprehensive optimization of four aspects: magnetic flux distribution, frequency response linearity, mechanical structure, and shielding characteristics. This sequential arrangement optimizes the magnetic circuit structure and spatial coupling relationship, helping to achieve balanced response across the entire frequency band, a balance between high and low frequency performance, and suppression of electromagnetic interference, while also simplifying the structural assembly and insulation encapsulation process.

[0063] In some other embodiments, along the axial direction of the end-screen acquisition unit, the sensing module 1 includes a low-frequency coil 11, a medium-frequency coil 12 and a high-frequency coil 13 sequentially sleeved on the central conductive rod 8. The three sensing coils have the same central aperture and are arranged at intervals along the axial direction to form a compact common magnetic circuit structure.

[0064] Along the axial direction of the end-screen acquisition unit, the low-frequency coil 11 is located at the front end of the end-screen acquisition unit, the intermediate-frequency coil 12 is located in the middle section of the end-screen acquisition unit, and the high-frequency coil 13 is located at the end of the end-screen acquisition unit. The materials used for the low-frequency coil 11, intermediate-frequency coil 12, and high-frequency coil 13 are the same as those in the above embodiment, which also helps to achieve balanced response of full-band signals, balance of high and low frequency performance, suppression of electromagnetic interference, and simplifies the structural assembly and insulation packaging process.

[0065] Furthermore, in some embodiments, an electromagnetic isolation structure 2 is provided between each induction coil to reduce mutual interference. The electromagnetic isolation structure 2 includes one or more combinations of a non-magnetic isolation ring, an air gap, and an insulating layer. This design effectively reduces the mutual inductance effect between the individual induction coils and improves the independence of signals in each frequency band.

[0066] Combination Figure 1 and Figure 2As shown, the encapsulation housing 3 includes an outer shell 31 and a cover 32 that can be interconnected. In the actual assembly process, the low-frequency coil 11 is first fitted into the central conductive rod 8, and an electromagnetic isolation structure 2 is installed on the side of the low-frequency coil 11 near the intermediate frequency coil 12. Then, the intermediate frequency coil 12 is assembled and another electromagnetic isolation structure 2 is installed. Next, the high-frequency coil 13 is assembled and fixed to the insulating frame (not shown). The whole assembly is then installed into the cavity of the outer shell 31. Finally, the cover 32 is encapsulated in the outer shell 31 and pressed tightly to form the whole end-screen acquisition unit.

[0067] In this embodiment, the encapsulation housing 3 is constructed from a single piece of cast aluminum alloy, ensuring both lightweight construction and excellent electromagnetic shielding performance and mechanical strength. It is understood that, for specific application environments, the encapsulation housing 3 of the end-screen acquisition unit can also be made of different metal materials, such as magnesium-aluminum alloy or copper alloy, to meet specific weight, thermal conductivity, or electromagnetic shielding requirements. The sealing structure can also be adjusted according to the specific characteristics of the oil medium to ensure long-term compatibility and reliability.

[0068] Furthermore, in terms of the conductive structure design, the conductive structure includes a conductive slider 4, a flow guide 5, and a conductive back cover 6 connected in sequence. The conductive slider 4 is configured to slide on the ejector pin 7 and is electrically connected to the ejector pin 7. The flow guide 5 can extend and retract along the connection direction between the conductive slider 4 and the conductive back cover 6, forming a dynamically adaptive conductive path.

[0069] Specifically, the conductive slider 4 is made of a highly conductive material, and its surface is specially treated to reduce contact resistance and improve wear resistance. The flow guide 5 is made of an alloy material with good conductivity and elasticity to ensure that it can maintain stable elasticity and conductivity under long-term compression. In this embodiment, the flow guide 5 can specifically be a conductive spring connected between the conductive slider 4 and the conductive back cover 6. It is compressed during assembly to form a preload. The preload in this application is set to about 10N-15N, thereby forming an adjustable elastic gap of about 0.3mm-0.5mm between the slider and the conductive back cover 6.

[0070] After the end-screen acquisition unit is assembled, the pre-pressure of the guide component 5 is transmitted to the front pin 7 through the conductive slider 4, achieving stable conductivity between the main body (not shown) and the conductive back cover 6. When the end-screen acquisition unit is subjected to mechanical vibration or temperature changes during operation, the slight relative displacement between the conductive slider 4 and the conductive back cover 6 is automatically absorbed by the elastic deformation of the guide component 5. The compression or release stroke of the guide component 5 is used to compensate for assembly tolerances and thermal expansion gaps, preventing possible contact interruptions under rigid connections. The continuous axial elastic force provided by the guide component 5 keeps the conductive path in a dynamic equilibrium state of "contact-return-re-contact," avoiding current fluctuations caused by micro-arcs or vibrations.

[0071] Furthermore, in terms of material selection, at least one of the induction coils is constructed using a temperature-stable magnetic core material. The permeability of this temperature-stable core material varies by less than ±3% within a temperature range of -40°C to 120°C. This temperature stability allows the terminal screen acquisition unit to maintain stable sensitivity and linearity even under fluctuating temperatures in an oil-immersion environment. Suitable materials include, but are not limited to, Fe-Ni-Mo permalloy, Fe-Si-B-Nb-Cu amorphous alloys, and Fe-Cu-Nb-Si-B nanocrystalline alloys.

[0072] Combination Figure 2 and Figure 3 As shown, the signal processing module is electrically connected to the sensing module 1 to process the output signals of each induction coil. Depending on the requirements, the signal processing module can be implemented using different hardware platforms, such as digital signal processors or dedicated integrated circuits, to provide different performance and cost balance options. In this embodiment, the signal processing module employs an advanced algorithm of independent sampling and dynamic weighted fusion to achieve high-fidelity reconstruction of the full-band signal.

[0073] Specifically, the signal processing module is configured to perform the following operations:

[0074] The current signals of each induction coil are sampled and bandwidth partitioned independently;

[0075] Weighting is performed on current signals of different frequency bands;

[0076] Dynamic weighted fusion and spectrum reconstruction of frequency domain signals are performed based on the weight calculation results.

[0077] In actual processing, the signal processing module first independently samples and partitions the bandwidth of the current signals from each induction coil. Each induction coil is connected to an independent sampling channel, and each channel contains a dedicated front-end conditioning circuit and an independent analog-to-digital converter (ADC). The front-end conditioning circuit includes a bandpass filter and a programmable gain amplifier, which can be optimized and adjusted for the characteristics of each frequency band. Sampling of each channel is triggered by a unified synchronous clock signal to ensure time synchronization.

[0078] After sampling, the signal processing module performs a Fast Fourier Transform (FFT) on each digital signal, converting the time-domain signal to the frequency domain and extracting the spectral components of each frequency band. Subsequently, the module dynamically assigns weight coefficients to the spectral components of each frequency band based on preset calibration data and real-time calculated signal-to-noise ratio (SNR). The weight calculation is based on the SNR characteristics of each channel at different frequencies, ensuring that higher SNR frequency bands are assigned greater weights, thereby improving overall signal quality.

[0079] Based on the weight calculation results, the signal processing module performs dynamic weighted fusion and spectrum reconstruction on the frequency domain signal. This process includes performing sliding window analysis on the shared frequency bands in the frequency band boundary area to calculate the amplitude gradient and phase difference. When the amplitude gradient and / or phase difference exceed a preset threshold, the system automatically adjusts the weight coefficients of the corresponding frequency band. A polynomial smoothing fitting algorithm is used to generate a continuous transition curve in the overlapping frequency band area to ensure that the amplitude frequency response is first-order continuous and the phase frequency response is zero-order continuous at the frequency band boundary.

[0080] After completing frequency domain fusion, the signal processing module performs inverse Fourier transform (IFFT) on the reconstructed spectrum, outputting a time-domain fused signal across the entire frequency band. This dynamic weighted fusion method effectively avoids the discontinuous frequency band splicing problem caused by traditional linear superposition, achieving a smooth response across the entire frequency band from 50Hz to 1MHz.

[0081] The aforementioned end-screen acquisition unit, by setting at least two induction coils configured for different frequency bands and cooperating with a signal processing module that independently acquires and digitally fuses the output signals of each induction coil, achieves high-fidelity acquisition of current signals over a wide frequency range. This solves the problem of measurement signal distortion caused by simple signal processing in the complex environment of oil-immersed bushing end screens for multi-coil sensors.

[0082] See Figure 3 As shown, in some embodiments, this application also provides a signal detection method applied to the above-mentioned end-screen acquisition unit. The signal detection method achieves high-precision acquisition and processing of full-frequency current signals through multiple steps, specifically including the following steps:

[0083] First, current signals at different frequency bands are independently acquired using multiple induction coils. Each induction coil is optimized for a specific frequency band to ensure optimal response characteristics within its target frequency range. The analog signals output from each coil are processed by a front-end conditioning circuit to achieve appropriate amplification and anti-aliasing filtering.

[0084] Subsequently, the current signals of each induction coil are independently digitally sampled. Each channel uses an independent analog-to-digital converter, controlled by a unified synchronous clock to ensure the consistency of sampling time for each channel. The sampled digital signals form their own independent data streams, which are transmitted to the signal processing module for further processing.

[0085] This architecture ensures that each channel is synchronized in time and independent in hardware, avoiding the problems of mutual interference, crosstalk, and quantization noise superposition caused by traditional co-channel sampling. The sampled data is transmitted to the signal processing module for parallel processing in the form of independent data streams, thereby achieving true "independent acquisition".

[0086] Within the signal processing module, each digital signal is partitioned into bandwidths and subjected to Fast Fourier Transform (FFT). Based on the design characteristics of each induction coil, a corresponding frequency band is set; for example, low-frequency coil 11 handles 50Hz to 20kHz, intermediate-frequency coil 12 handles 20kHz to 200kHz, and high-frequency coil 13 handles 200kHz to 1MHz. The time-domain signal of each channel is converted to a frequency-domain representation using FFT to obtain the corresponding spectral components.

[0087] Subsequently, the signal processing module performs sensitivity normalization. Based on preset calibration data and the real-time calculated signal-to-noise ratio, the signal processing module dynamically assigns weighting coefficients Wi(f) to the spectral components of each frequency band. The calculation formula is as follows:

[0088]

[0089] Where SNR_i(f) is the signal-to-noise ratio of the i-th channel at frequency f.

[0090] Based on the weight calculation results, the signal processing module performs dynamic weighted fusion and spectrum reconstruction on the frequency domain signal. That is, the signals of each channel are dynamically weighted according to Wi(f) and then superimposed in the frequency domain. The time-domain fused signal is obtained by inverse fast Fourier transform. This algorithm can adjust the weights in real time under different temperature, noise and amplitude conditions to achieve adaptive compensation and dynamic smooth transition.

[0091] Specifically, this process involves performing a sliding window analysis on the shared frequency band at the frequency band boundary (e.g., the boundary between MF and HF) to calculate the amplitude gradient ΔA(f) and phase difference Δφ(f).

[0092] When the amplitude gradient and / or phase difference exceed a preset threshold, in this application, the preset threshold is set to ΔA>0.5dB or Δφ>3°, and the system automatically adjusts the weighting coefficients of the corresponding frequency band. A polynomial smoothing fitting algorithm is used to generate a continuous transition curve in the overlapping frequency band area to ensure that the amplitude frequency response is first-order continuous and the phase frequency response is zero-order continuous at the frequency band boundary.

[0093] After completing frequency domain fusion, the signal processing module performs an inverse Fourier transform on the reconstructed spectrum, outputting a time-domain fused signal across the entire frequency band. This dynamic weighted fusion method effectively avoids the discontinuous frequency band splicing problem caused by traditional linear superposition, achieving a smooth response across the entire frequency band from 50Hz to 1MHz.

[0094] The final output time-domain fusion signal across the entire frequency band contains complete current information from power frequency to high frequency, which can accurately reflect the electrical status of the oil-immersed bushing end screen.

[0095] In practice, the induction coils for each frequency band undergo three-dimensional calibration of permeability, temperature, and frequency response both at the factory and during field operation. The calibration results are stored in a parameter table built into the signal processing module. During the acquisition process, the signal processing module reads the operating temperature and spectral response characteristics of each coil in real time and compares them with the calibration database.

[0096] When the signal processing module detects that the temperature drift rate |Δμ / μ| of the permeability μ exceeds a set threshold (e.g., ±2%), the system automatically triggers the algorithm weight correction process:

[0097] Update the weighting coefficients Wi(f,T) for this frequency band;

[0098] Adjust the correction function K(f,μ) for spectrum reconstruction to compensate for sensitivity attenuation or phase shift;

[0099] The amplitude-frequency response curve of the fused signal is recalculated in the next sampling period.

[0100] Through the aforementioned feedback loop of real-time detection—parameter update—algorithm correction—output re-detection, changes in material parameters can be corrected instantly at the algorithm level, thereby maintaining continuous response and amplitude-phase consistency across different frequency bands. Experimental verification shows that this collaborative closed-loop mechanism can control the overall sensitivity variation of the sensor within ±1.5% in the temperature range of −40°C to 120°C, with a cross-frequency amplitude-frequency response error of less than 0.5dB, achieving dynamic consistency optimization of structure and algorithm.

[0101] In some embodiments, this application also provides an oil-immersed sleeve comprising a terminal screen acquisition unit as shown in any of the above embodiments. The oil-immersed sleeve includes a terminal screen terminal (not shown) and a terminal screen acquisition unit, wherein the terminal screen acquisition unit is directly mounted on the terminal screen terminal.

[0102] The sensing module 1 of the end-screen acquisition unit is housed within a metal cavity formed by the end-screen terminals, and is coaxially arranged with and signal-connected to the end-screen grounding lead 9. This arrangement ensures accurate acquisition of the current signal while making full use of the existing structural space of the end-screen terminals.

[0103] The end-screen acquisition unit and the end-screen terminals are mechanically connected and fixed via threads or flanges. This connection method ensures both installation stability and ease of on-site installation and maintenance. A sealing ring is installed between the encapsulation housing 3 of the end-screen acquisition unit and the inner wall of the metal cavity, forming a sealing structure suitable for oil immersion environments, effectively preventing leakage of insulating oil.

[0104] The enclosure 3 of the final screen acquisition unit and the metal cavity of the final screen terminal together form a continuous electromagnetic shielding structure. This design effectively suppresses the influence of external electromagnetic interference on signal acquisition and improves the reliability of measurement results.

[0105] In summary, the wideband response characteristics of the end-screen acquisition unit shown in this application cover the entire range from power frequency to high frequency, enabling simultaneous capture of electrical phenomena with different frequency characteristics, such as partial discharge and transient overvoltage. The dynamic weighted fusion algorithm effectively solves the problem of discontinuous frequency band splicing caused by traditional linear superposition, improving the accuracy and reliability of the measurement results.

[0106] In terms of structural design, the common magnetic circuit layout of the induction coil structure and the application of temperature-stable magnetic core materials ensure the long-term stability of the sensor under oil immersion and temperature fluctuation conditions. The electromagnetic isolation structure and shielding design effectively suppress internal and external electromagnetic interference, improving the signal-to-noise ratio of the acquired signal.

[0107] The optimized design of the mounting structure allows the terminal acquisition unit to make full use of the limited space of the terminal, while ensuring ease of installation and reliable connection. The adaptive design of the conductive structure effectively addresses the challenges posed by mechanical vibration and temperature changes, ensuring the stability of the electrical connection during long-term operation.

[0108] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above 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.

[0109] The embodiments described above are merely illustrative of several implementation methods of this application, and 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 those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A terminal screen acquisition unit, characterized in that, include: A sensing module for sensing current signals flowing through the end screen; the sensing module includes at least two induction coils configured for different frequency bands, and all the induction coils are configured to be arranged in a common magnetic circuit. and A signal processing module, electrically connected to the sensing module, is used to process the current signal; The signal processing module is configured to perform the following operations: The current signals of each of the aforementioned induction coils are independently sampled and bandwidth partitioned; Weighting is performed on current signals of different frequency bands; Dynamic weighted fusion and spectrum reconstruction of frequency domain signals are performed based on the weight calculation results.

2. The end-screen acquisition unit according to claim 1, characterized in that, The dynamic weighted fusion of frequency domain signals based on weight calculation results includes: Perform sliding window analysis on the shared frequency bands at the frequency band boundary to calculate the amplitude gradient and phase difference; When the amplitude gradient and / or the phase difference exceed a preset threshold, the weighting coefficient of the corresponding frequency band is adjusted. A polynomial smoothing fitting algorithm is used to generate continuous transition curves in the overlapping frequency band region.

3. The end-screen acquisition unit according to claim 1, characterized in that, All of the aforementioned induction coils include a low-frequency coil, an intermediate-frequency coil, and a high-frequency coil arranged sequentially, wherein the low-frequency coil, intermediate-frequency coil, and high-frequency coil are configured to be arranged in a common magnetic circuit. The low-frequency coil is made of a soft magnetic material with high saturation magnetic induction intensity and low magnetic loss. The intermediate frequency coil is made of a composite soft magnetic material with medium permeability and wide frequency response characteristics. The high-frequency coil adopts a low dielectric constant, low loss high-frequency dielectric substrate or an air core structure.

4. The end-screen acquisition unit according to claim 3, characterized in that, At least one of the low-frequency coil, intermediate-frequency coil, and high-frequency coil uses a temperature-stable magnetic core material, which is configured such that the permeability variation is less than ±3% in the temperature range of -40°C to 120°C.

5. The end-screen acquisition unit according to any one of claims 1-4, characterized in that, It also includes an electromagnetic isolation structure disposed between any two of the induction coils; the electromagnetic isolation structure includes at least one or any combination of a non-magnetic isolation ring, an air gap, and an insulating layer.

6. The end-screen acquisition unit according to any one of claims 1-4, characterized in that, The end-screen acquisition unit also includes a housing for accommodating the sensing module and the signal processing module; the housing is configured to be integrally cast from aluminum alloy.

7. The end-screen acquisition unit according to any one of claims 1-4, characterized in that, The end-screen acquisition unit also includes a conductive slider, a flow guide, and a conductive back cover that are electrically connected in sequence. The flow guide is configured to extend and retract along the connection direction between the conductive slider and the conductive back cover.

8. A signal detection method, applied to the end-screen acquisition unit as described in any one of claims 1-7, characterized in that, The signal detection method includes the following steps: Current signals of different frequency bands are collected independently by multiple induction coils; The current signal of each of the aforementioned induction coils is independently digitally sampled to form an independent digital signal; The bandwidth of each digital signal is partitioned and subjected to Fast Fourier Transform to obtain the corresponding spectral components; Based on the preset calibration data and the real-time calculated signal-to-noise ratio, weighting coefficients are dynamically assigned to the spectral components of each frequency band. The spectral components are weighted and fused and reconstructed based on the weighting coefficients. The reconstructed spectrum is inversely transformed to output a time-domain fused signal across the entire frequency band.

9. The signal detection method according to claim 8, characterized in that, The weighted fusion of the spectral components based on the weighting coefficients includes: Amplitude gradient and phase difference analysis of spectral components in the frequency band boundary region; When the analysis results exceed the set threshold, the weighting coefficients are adaptively adjusted. A polynomial smoothing fitting algorithm is used to ensure that the amplitude frequency response is continuous at the first order and the phase frequency response is continuous at the zero order at the frequency band boundary.

10. An oil-immersed bushing, comprising a final shield terminal, characterized in that, It also includes the end-screen acquisition unit as described in any one of claims 1-7, wherein the end-screen acquisition unit is installed on the end-screen terminal.

Images