Method, device and equipment for evaluating insulation state of impregnated fiber sleeve, medium and product

Abstract

The invention discloses a method, a device and equipment for evaluating the insulation state of an impregnated fiber sleeve, a medium and a product, and relates to the technical field of electrical equipment fault diagnosis, and the method comprises the steps: issuing a test configuration instruction to a test terminal; wherein the test terminal comprises a high-frequency square wave power supply, a high-voltage relay, a current measurement unit and a data processing unit; the test configuration instruction comprises indication of applying bipolar square wave voltage excitation with preset parameters to the impregnated fiber sleeve sample, collection of a polarization current response sequence in a polarization stage, and collection of a depolarization current response sequence in a depolarization stage; receiving test data uploaded by the test terminal; wherein the test data comprises a polarization current response sequence acquired in a polarization stage and a depolarization current response sequence acquired in a depolarization stage; performing conjoint analysis on the polarization current response sequence and the depolarization current response sequence to obtain characteristic parameters; and based on the characteristic parameters, determining an insulation state evaluation result of the impregnated fiber sleeve sample. According to the method, the insulation state can be accurately evaluated.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electrical equipment fault diagnosis, and in particular to a method and device for evaluating the insulation state of a fiber-saturated bushing, equipment, medium and product. BACKGROUND

[0002] High-voltage power equipment is the physical cornerstone of safe operation of the power grid, and the reliability of its insulation system directly determines the stability and life of the entire power system. As the main insulation material of transformers, bushings and other equipment, fiber-saturated insulation is subjected to the combined action of multiple stresses such as electricity, heat, mechanics and environment during long-term operation, and will inevitably undergo progressive aging, leading to degradation of its dielectric properties and triggering serious faults such as partial discharge and even insulation breakdown. According to statistics, insulation aging is the primary cause of high-voltage equipment failure. With the deepening of the construction of smart grids and the state maintenance mode of equipment, the power industry is transforming from regular maintenance and post-maintenance to predictive maintenance. The prerequisite for this transformation is to first achieve timely, non-destructive and accurate evaluation of the insulation state to scientifically predict the remaining life and warn of potential risks, thereby ensuring the essential safety of the power grid, optimizing the allocation of operation and maintenance resources, and avoiding catastrophic power outages. Therefore, developing advanced insulation state evaluation technology has become an urgent and major demand in the field of intelligent operation and maintenance of power equipment.

[0003] Currently, the evaluation of fiber-saturated insulation and other solid insulation materials mainly relies on three traditional methods: electrical property method, chemical analysis method and partial discharge detection. However, these methods have problems such as being susceptible to environmental interference and being cumbersome, and cannot accurately evaluate the insulation state.

[0004] Therefore, there is an urgent need for a method that can accurately evaluate the insulation state. SUMMARY

[0005] The present application provides a method, device, equipment, medium and product for evaluating the insulation state of a fiber-saturated bushing, which can accurately evaluate the insulation state.

[0006] To achieve the above-mentioned purpose, the present application adopts the following technical solutions: In a first aspect, the present application provides a method for evaluating the insulation state of a fiber-saturated bushing, comprising: issuing a test configuration instruction to a test terminal; wherein the test terminal includes a high-frequency square wave power supply, a high-voltage relay, a current measurement unit and a data processing unit; the test configuration instruction includes instructions to apply a preset parameter of a bipolar square wave voltage excitation to the fiber-saturated bushing sample, collect a polarization current response sequence during the polarization stage, and collect a depolarization current response sequence during the depolarization stage; receive test data uploaded from the test terminal; wherein the test data comprises a polarization current response sequence collected in a polarization phase and a depolarization current response sequence collected in a depolarization phase; jointly analyze the polarization current response sequence and the depolarization current response sequence to obtain a characteristic parameter; determine an insulation state evaluation result of the impregnated fiber sleeve sample based on the characteristic parameter.

[0007] In some possible implementation manners, the preset parameters include a square wave voltage frequency, a duty cycle, and a voltage amplitude.

[0008] In some possible implementation manners, the characteristic parameter obtained by jointly analyzing the polarization current response sequence and the depolarization current response sequence includes: perform short-time Fourier transform on the polarization current response sequence to obtain a first time-frequency spectrum, and perform short-time Fourier transform on the depolarization current response sequence to obtain a second time-frequency spectrum; extract a relaxation time constant and a weight coefficient corresponding to each relaxation time constant based on the second time-frequency spectrum; calculate a first energy distribution ratio of the polarization current response sequence in a preset frequency band based on the first time-frequency spectrum to obtain a polarization energy feature; calculate a second energy distribution ratio of the depolarization current response sequence in the preset frequency band based on the second time-frequency spectrum to obtain a depolarization energy feature; calculate an integral difference value between the first time-frequency spectrum and the second time-frequency spectrum in a preset time window to obtain an integral charge feature.

[0009] In some possible implementation manners, the impregnated fiber sleeve sample has a preset size and is subjected to vacuum drying treatment.

[0010] In some possible implementation manners, determining the insulation state evaluation result of the impregnated fiber sleeve sample based on the characteristic parameter includes: input the characteristic parameter into a pre-trained insulation state evaluation model to obtain the insulation state evaluation result of the impregnated fiber sleeve sample; wherein the insulation state evaluation model is constructed based on a machine learning algorithm and is trained through historical samples with known aging states; the insulation state evaluation result includes an insulation state category, an aging degree level, and confidence information.

[0011] In some possible implementation manners, the impregnated fiber sleeve sample has a water content lower than a preset water content threshold.

[0012] In a second aspect, the present application provides an impregnated fiber sleeve insulation state evaluation device, comprising: The instruction issuing module is configured to issue a test configuration instruction to a test terminal; the test terminal comprises a high-frequency square wave power supply, a high-voltage relay, a current measurement unit, and a data processing unit; the test configuration instruction comprises instructions to apply a preset parameter bipolar square wave voltage excitation to a gel-impregnated fiber sleeve sample, collect a polarization current response sequence in a polarization stage, and collect a depolarization current response sequence in a depolarization stage; The data receiving module is configured to receive test data uploaded from the test terminal; the test data comprises the polarization current response sequence collected in the polarization stage and the depolarization current response sequence collected in the depolarization stage; The analysis module is configured to jointly analyze the polarization current response sequence and the depolarization current response sequence to obtain a characteristic parameter; The evaluation module is configured to determine an insulation state evaluation result of the gel-impregnated fiber sleeve sample based on the characteristic parameter.

[0013] In a third aspect, the present application provides a computing device comprising a memory and a processor; The memory stores one or more computer programs comprising instructions; when the instructions are executed by the processor, the computing device performs the method of any one of the first aspect.

[0014] In a fourth aspect, the present application provides a computer-readable storage medium for storing a computer program for executing the method of any one of the first aspect.

[0015] In a fifth aspect, the present application provides a computer program product comprising one or more computer instructions; when the computer instructions are executed by a computer, the computer executes the method of any one of the first aspect.

[0016] From the above technical solutions, the present application has at least the following beneficial effects: In the present application, by issuing the test configuration instruction including indicating the bipolar square wave voltage excitation applied to the impregnated fiber sleeve sample with preset parameters, collecting the polarization current response sequence of the polarization stage, and collecting the depolarization current response sequence of the depolarization stage to the test terminal, the evaluation of the insulation state without accelerating insulation aging or causing damage is realized; further, receiving the test data uploaded from the test terminal provides sufficient data conditions for subsequent use of multiple signal processing algorithms; then, the polarization current response sequence and the depolarization current response sequence can be jointly analyzed to obtain characteristic parameters, and the direct current conductance component representing insulation aging or moisture is more accurately extracted, providing multi-dimensional criteria for state evaluation; finally, based on the characteristic parameters, the insulation state evaluation result of the impregnated fiber sleeve sample is determined. The present scheme realizes non-destructive testing by introducing the test terminal, realizes pre-diagnosis; furthermore, by jointly analyzing the polarization current response sequence and the depolarization current response sequence, the wide time domain relaxation spectrum from milliseconds to thousands of seconds can be more completely reconstructed, and finally the accurate evaluation result is obtained.

[0017] It should be understood that the description of technical features, technical solutions, beneficial effects or similar language in the present application does not imply that all features and advantages can be realized in any single embodiment. On the contrary, it can be understood that the description of a feature or beneficial effect means that the specific technical feature, technical solution or beneficial effect is included in at least one embodiment. Therefore, the description of technical features, technical solutions or beneficial effects in the specification does not necessarily refer to the same embodiment. Further, the technical features, technical solutions and beneficial effects described in the present embodiment can be combined in any appropriate manner. Those skilled in the art will understand that the embodiments can be implemented without one or more specific technical features, technical solutions or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects can be identified in specific embodiments that do not embody all embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 An application environment diagram of a kind of impregnated fiber sleeve insulation state evaluation method provided in the present application embodiment; Figure 2 A flowchart of a kind of impregnated fiber sleeve insulation state evaluation method provided in the present application embodiment; Figure 3 A structure diagram of a kind of impregnated fiber sleeve insulation state evaluation device provided in the present application embodiment; Figure 4 A schematic diagram of a kind of computing device provided in the present application embodiment. DETAILED DESCRIPTION

[0019] The terms "first", "second", and "third" and the like in the specification and the drawings of the present application are used to distinguish different objects, rather than to define a particular order.

[0020] In the embodiments of the present application, the words such as "exemplary" or "for example" are used to mean serving as an example, instance, or illustration. Any embodiment or design presented as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or advantageous than other embodiments or design schemes. Rather, the use of such terms is intended to present concepts in a particular way. The use of the words such as "exemplary" or "for example" is intended to present concepts in a particular way.

[0021] The evaluation of the insulation state of the impregnated fiber bushing is one of the important technologies for the condition-based maintenance of high-voltage power equipment, which aims to obtain characteristic information reflecting the internal microstructure changes of the insulation material through non-destructive or slightly destructive testing methods, and then make scientific judgments on the aging degree, moisture condition and remaining life.

[0022] Dielectric relaxation refers to the phenomenon that the polarization or depolarization process is not completed instantaneously, but gradually established or attenuated with a certain time constant. Different relaxation time constants correspond to different microphysical mechanisms (such as dipole turning, ion migration, and interface polarization).

[0023] At present, the evaluation methods based on polarization / depolarization current mostly use direct current or low-frequency sinusoidal wave as the excitation source. This excitation mode has two fundamental limitations, the first is low testing efficiency. Under direct current excitation, the polarization and depolarization processes are slow, and a single test takes tens of minutes or even hours, which completely cannot meet the demand of on-site rapid inspection; the second is incomplete information excitation. Direct current or low-frequency excitation mainly excites slow relaxation processes (such as ion conduction and deep interface polarization), but insufficiently excites the medium-high frequency band dielectric response that can more sensitively reflect the material trap characteristics and shallow interface state changes caused by early thermal aging and electrical aging, resulting in single information dimension and unremarkable fingerprint characteristics of the obtained current signal.

[0024] And the deeper problem is the bottleneck of signal analysis and interpretation. Even if the current response is obtained, the traditional method mostly relies on observing the curve shape or calculating limited macroscopic parameters (such as steady-state current value), and lacks the ability to automatically and accurately decouple and quantify the contributions of different micro relaxation mechanisms from complex and non-stationary time series signals. This makes the evaluation results seriously dependent on expert experience, with strong subjectivity, and the boundary between slight aging and normal state is blurred, making it difficult to achieve standardized and quantitative accurate evaluation.

[0025] Therefore, the embodiments of the present application provide an insulation state evaluation method for impregnated fiber bushings. In order to make the technical solutions of the present application clearer and easier to understand, the application scenarios of the technical solutions of the present application are introduced below with reference to the drawings. As shown in Figure 1As shown in the figure, the figure is a schematic diagram of an application scenario provided by the embodiment of the application.

[0026] In this application environment, the sensor 104 (usually a smart server integrated with data analysis function) located in the control center generates a structured test configuration instruction according to the evaluation plan or the trigger condition such as periodic inspection and abnormality early warning. The instruction is encapsulated in the form of a digital protocol package, and the parameters include excitation parameters, test procedures, acquisition parameters and sample identification, etc. The sensor 104 sends the instruction downward to the test terminal 102 deployed on the equipment site or in the laboratory through a reliable communication network such as industrial Ethernet, optical fiber or 5G.

[0027] Further, after the data processing unit of the test terminal 102 receives the instruction, it immediately performs analysis, drives the high-frequency square wave power supply to generate an accurate bipolar square wave voltage signal according to the preset parameters, and controls the high-voltage relay to be in the initial state (usually connected to the output end of the square wave power supply). In the polarization stage, the relay remains connected, the square wave voltage excitation is applied to the sample of the impregnated fiber sleeve, the current measurement unit (picoammeter) is started synchronously, the polarization current response sequence is collected with high precision, and the data is cached locally in real time. Stage switching and depolarization, at the preset end point of the polarization time, the data processing unit immediately sends a control signal to drive the high-voltage relay to switch to the "ground" position at high speed, so that the sample is short-circuited at both ends. In the depolarization stage, the current measurement unit continues to work, and the depolarization current response sequence is collected. After the complete data collection is completed, the local cache stores the original data set of the depolarization current response sequence and the polarization current response sequence.

[0028] Further, the data processing unit of the test terminal 102 packs the collected double-channel current sequence, time stamp, sample number and actual execution parameters (such as actual voltage fluctuation) of this test into a test data package, and transmits the data package upward to the sensor 104 through the communication link.

[0029] After the sensor 104 receives the data, it performs integrity check (such as data length check and CRC check). If the check is passed, an acknowledgement is sent to the test terminal 102; if the data is abnormal or missing, a retransmission request or an error alarm may be triggered.

[0030] In order to make the technical solutions of the application clearer and easier to understand, the following describes a fault recovery method provided by the embodiment of the application in combination with the above application scenario. As shown in the figure, the figure is a flow chart of a fault recovery method provided by the embodiment of the application. Figure 2 As shown in the figure, the figure is a flow chart of a method for evaluating the insulation state of an impregnated fiber sleeve provided by the embodiment of the application. The method for evaluating the insulation state of the impregnated fiber sleeve comprises: S201, a test configuration instruction is issued to a test terminal.

[0031] The test terminal comprises a high-frequency square-wave power supply, a high-voltage relay, a current measuring unit and a data processing unit.

[0032] The test configuration instruction comprises instructions for applying a preset parameter of a bipolar square-wave voltage excitation to the impregnated fiber sleeve sample, collecting a polarization current response sequence of a polarization stage, and collecting a depolarization current response sequence of a depolarization stage.

[0033] The preset parameter comprises a square-wave voltage frequency, a duty cycle and a voltage amplitude.

[0034] The impregnated fiber sleeve sample is of a preset size and has been subjected to vacuum drying treatment.

[0035] The water content of the impregnated fiber sleeve sample is lower than a preset water content threshold.

[0036] The test terminal can be a set of hardware devices deployed at a test site for directly performing the insulation state evaluation test.

[0037] The high-frequency square-wave power supply is a high-voltage power supply with a specific frequency, amplitude and waveform (bipolar square wave), and is an excitation source.

[0038] The high-voltage relay is a high-speed electronic switch for quickly and safely switching circuit connections between two test stages (polarization / depolarization).

[0039] The current measuring unit is usually a picoammeter, an instrument capable of accurately measuring extremely weak current (picoampere, i.e., 10⁻¹² ampere), and is used to collect the response current of the insulating material.

[0040] The data processing unit is an embedded device or industrial computer with data acquisition, caching and preliminary processing capabilities, and is responsible for receiving and executing instructions from the server and uploading raw current data.

[0041] The test configuration instruction is a complete set of operation instructions generated by the control center (such as the server 104) and issued to the test terminal, for accurately controlling the start, parameter setting and data collection process of the test.

[0042] The polarization stage is a period of applying an external square-wave voltage excitation to the impregnated fiber sleeve sample; at this time, the charges (such as dipoles, interface charges and trap charges) inside the impregnated fiber sleeve sample are oriented and arranged or migrated under the action of the external electric field, a polarization state is established, and a polarization current is generated.

[0043] The depolarization stage is a period of instantaneously removing the voltage applied to both ends of the impregnated fiber sleeve sample (by switching to short-circuit ground through the relay); at this time, the external electric field disappears, the charges inside the impregnated fiber sleeve sample established in the polarization stage begin to relax and release, and return to the equilibrium state, and a depolarization current is generated during this process.

[0044] The polarization current response sequence is a series of current values continuously collected by the current measurement unit at a high sampling rate (e.g., microsecond or millisecond level) during the polarization phase, forming a current-time curve that changes over time, which fully records the dynamic dielectric response of the insulating material to external excitation.

[0045] The depolarization current response sequence is a series of current values continuously collected by the current measurement unit at a high sampling rate (e.g., microsecond or millisecond level) during the polarization or depolarization phase, forming a current-time curve that changes over time, which fully records the dynamic dielectric response of the insulating material to external excitation.

[0046] The preset parameters are the key electrical parameters of the square wave excitation that are pre-set according to the characteristics of the impregnated fiber material and the evaluation target; mainly including square wave voltage frequency, duty cycle and voltage amplitude; for example, the square wave voltage frequency is 10 kHz, the high frequency can accelerate the test and excite the dielectric response in the medium and high frequency band, the duty cycle is 50%, which means that the high level (positive voltage) and low level (negative voltage) in a cycle each lasts for half of the time, forming a symmetrical bipolar excitation, which helps to suppress the single direction accumulation of space charge, and the voltage amplitude is 1kV, which is selected according to the thickness of the sample and the breakdown field strength of the material, which needs to be high enough to generate a measurable signal, but cannot cause breakdown.

[0047] The impregnated fiber sleeve sample is a standardized insulating material sample for evaluation, which simulates the structure of the actual impregnated fiber sleeve, but is cut to a pre-set size (e.g., 5cm x 5cm x 0.1cm) to ensure the consistency and comparability of the test.

[0048] Key pretreatment, must be treated by vacuum drying, its water content is strictly controlled below the pre-set water threshold (e.g., less than 0.5%), because water is a strong polar impurity, which will significantly interfere with and mask the intrinsic dielectric response signal of the insulating material, drying is a necessary prerequisite to ensure that the test signal reflects the true aging state of the material (rather than being affected by moisture).

[0049] For example, the control center (server 104) can issue a detailed test configuration instruction to the on-site test terminal according to the evaluation strategy; the specific content of the instruction is: Excitation application: command the high-frequency square wave power supply to output a bipolar square wave voltage with specific preset parameters (e.g., 10 kHz, 50% duty cycle, 1kV), and apply it to the impregnated fiber sleeve sample; Data acquisition: during the polarization phase of applying voltage, command the current measurement unit to start collecting the polarization current response sequence; when the high-voltage relay is quickly switched to the short-circuit state and enters the depolarization phase, continue to collect the depolarization current response sequence.

[0050] Precondition for execution: the successful execution of the instruction depends on the strict standardization of the test object, such as the gel-impregnated fiber sleeve sample must meet two conditions: physical size standardization (preset size) to ensure the comparability of results between different tests, between laboratory and field tests; state standardization (dry, low water content) can reduce water content below the threshold value through vacuum drying, fundamentally eliminating the pollution of the most important interference factor of moisture on the test results, ensuring that the collected current sequence reflects the intrinsic information of material aging (such as molecular chain rupture, interface deterioration), rather than moisture.

[0051] In summary, the server 104 remotely drives the test terminal to automatically complete the whole process from excitation application to double-channel current data acquisition by issuing a fine instruction that integrates incentive parameters, test procedures and sample standards, providing a raw, pure and standardized data basis for subsequent insulation state analysis, and achieving high automation and standardization.

[0052] It should be noted that the test terminal is a hardware system that integrates excitation source, high-speed switch, precise measurement and local computing functions, capable of generating high-frequency square wave power of specific waveform, high-voltage relay for quickly switching circuit state between polarization and depolarization stages, current measurement unit (such as picoammeter) for capturing extremely weak current response, and data processing unit responsible for instruction reception and data preliminary processing.

[0053] The test configuration instruction issued is not a simple instruction, but a programmed operation sequence that explicitly instructs the terminal: first, apply a preset parameter bipolar square wave voltage excitation to the prepared gel-impregnated fiber sleeve sample; second, during the polarization stage of applying voltage, synchronize the collection of polarization current response sequence; third, during the depolarization stage of removing voltage and short-circuiting the sample, synchronize the collection of depolarization current response sequence. The preset parameters are the core of the method, usually including the frequency (such as 10 kHz), duty cycle (such as 50%) and voltage amplitude (such as 1 kV) of the square wave voltage, which are specially designed to efficiently excite various dielectric relaxation processes inside the insulation material.

[0054] It should be noted that through programmed instructions, the test system is instructed to apply a preset parameter bipolar square wave voltage excitation to the sample and synchronize the collection of polarization and depolarization currents. Its beneficial effects are to achieve non-destructive, low-field safety evaluation, avoiding the insulation damage that may be caused by traditional high-voltage testing; at the same time, the highly controllable and information-rich excitation signal can flexibly excite various polarization processes from fast to slow inside the insulation material; the clear stage division lays a solid foundation for subsequent separation and acquisition of pure polarization and depolarization responses, providing a clear physical meaning data source for in-depth analysis.

[0055] S202. Receive test data uploaded from the test terminal.

[0056] The test data includes polarization current response sequences collected during the polarization phase and depolarization current response sequences collected during the depolarization phase.

[0057] Test data refers to raw electrical measurement data that directly reflects the dielectric response of insulating materials.

[0058] The polarization stage is a specific time period during which an external square wave voltage excitation is applied to the impregnated fiber sheath sample. During this stage, the electric field force drives the microscopic charge carriers (such as dipoles, ions, and interface bound charges) inside the insulating medium to oriented or migrate, thereby establishing a macroscopic polarization state inside the medium.

[0059] The polarization current response sequence is a series of time-varying current values ​​continuously measured and recorded by a current measurement unit at a high sampling rate during the polarization phase. This current is an instantaneous superposition of multiple physical processes, containing complete dynamic information about the polarization establishment process of the insulating material. Its amplitude, attenuation mode, and steady-state value are directly related to the material's dielectric constant, polarizability, and conductivity.

[0060] The depolarization stage is a specific time period after the polarization stage ends, during which the two ends of the sample are quickly short-circuited to ground by a high-voltage relay; at this time, the external excitation electric field is removed instantaneously.

[0061] The depolarization current response sequence is a series of time-varying current values ​​continuously measured and recorded by a current measurement unit during the depolarization phase. This current is generated by the relaxation and release of various polarization charges and space charges "frozen" or trapped within the dielectric during the polarization phase, after the removal of the external field, towards thermal equilibrium. This sequence is an important fingerprint for diagnosing the aging state of insulation. Aging (such as the formation of traps and interface degradation) significantly alters the charge trapping and releasing dynamics, thus leaving unique marks on the amplitude, decay time constant, and curve shape of the depolarization current. The depolarization current is more sensitive than the polarization current in reflecting the microscopic defects and aging degree of the material.

[0062] Exemplarily, the server 104, as the core of data processing and intelligent analysis, receives the original measurement data actively uploaded from the test terminal through the communication network. The received test data is not a single value, but two complete and synchronous time series data sets. One is the polarization current response sequence, which records the instantaneous and decaying current accompanied by the establishment of the macroscopic polarization state in the insulation sample due to the processes of dipole turning and interface polarization under external square wave voltage excitation. This sequence contains the dynamic information of the polarization establishment and the intrinsic dielectric properties of the material. The other is the depolarization current response sequence, which records the current generated in the relaxation and release process of the charges (such as trap charges and interface charges) bound or captured in the polarization phase to the thermal equilibrium state after the external excitation is removed and the sample is short-circuited. This sequence is a sensitive fingerprint reflecting the micro-defects, trap distribution and aging degree of the insulation material.

[0063] The two sequences form a set of complementary and inseparable twin data, which together complete the integrated observation of the dynamic dielectric behavior of the insulation sample under the same excitation. The successful reception of the server 104 of this pair of data sequences marks the formal entry of the test into the data analysis stage, providing a unique and complete original data basis for all subsequent in-depth processing such as signal denoising, feature extraction, model comparison and insulation state classification evaluation. It is an important data input point for realizing the conversion of physical measurement into insulation state knowledge.

[0064] It should be noted that the polarization and depolarization current response original sequences uploaded by the test terminal are obtained. The direct beneficial effect is to capture the original time-domain dielectric response fingerprint of the insulation material under a specific electrical excitation. These data completely contain the comprehensive information of the micro-processes inside the insulation. Receiving two sets of data realizes the complementation and mutual verification of information, improves the data reliability and helps to identify noise. At the same time, the complete and high-precision time-domain sequence provides sufficient data conditions for subsequent in-depth analysis and feature mining using various advanced algorithms.

[0065] S203, jointly analyzing the polarization current response sequence and the depolarization current response sequence to obtain a characteristic parameter.

[0066] Among them, the characteristic parameter includes a relaxation time constant, a weight coefficient corresponding to each relaxation time constant, a polarization energy feature, a depolarization energy feature, and an integral charge feature.

[0067] In an implementation, a short-time Fourier transform is performed on the polarization current response sequence to obtain a first time-frequency spectrum, and a short-time Fourier transform is performed on the depolarization current response sequence to obtain a second time-frequency spectrum; based on the second time-frequency spectrum, a relaxation time constant and a weight coefficient corresponding to each relaxation time constant are extracted; based on the first time-frequency spectrum, a first energy distribution ratio of the polarization current response sequence in a preset frequency band is calculated to obtain a polarization energy feature; based on the second time-frequency spectrum, a second energy distribution ratio of the depolarization current response sequence in the preset frequency band is calculated to obtain a depolarization energy feature; and the preset frequency band includes a first frequency band and a second frequency band; and an integral difference value between the first time-frequency spectrum and the second time-frequency spectrum is calculated in a preset time window to obtain an integral charge feature.

[0068] In the joint analysis, the polarization current response sequence and the depolarization current response sequence are not considered in isolation, but are regarded as two stages of the same physical process (excitation-response-relaxation) and are correlated, compared, and comprehensively interpreted to extract more comprehensive and stable state features.

[0069] The characteristic parameter is a mathematical or physical quantity extracted from the original data, which can quantitatively represent the specific physical properties or aging state of the insulation material, and is a bridge connecting the original signal and the final state evaluation conclusion.

[0070] The relaxation time constant is a physical quantity describing the speed of a certain polarization or depolarization process in the insulation medium. A larger relaxation time constant usually corresponds to a slower relaxation process (such as deep trap charge release, interface polarization), which is often related to material aging (increased trap density, interface deterioration).

[0071] The weight coefficient is a coefficient in front of each exponential term when the depolarization current curve is fitted as a superposition model of multiple exponential decay terms (corresponding to different relaxation time constants). It represents the weight or intensity of the corresponding relaxation process to the total depolarization current, and reflects the relative activity of different microscopic mechanisms (such as traps of different depths).

[0072] The polarization energy feature is a feature obtained by calculating the energy distribution ratio of the first time-frequency spectrum in a certain preset frequency band. It reflects the relative proportion of energy consumed or released by the polarization process at different time scales (corresponding to different frequencies). For example, aging may cause the energy proportion of low-frequency components (slow processes) to rise.

[0073] The depolarization energy feature is a feature obtained by calculating the energy distribution ratio of the first time-frequency spectrum in a certain preset frequency band. It reflects the relative proportion of energy consumed or released by the depolarization process at different time scales (corresponding to different frequencies).

[0074] Integral charge characteristic is the integral difference of the physical quantity (such as time-frequency spectrum) related to the polarization process and the depolarization process in the preset time window. It directly quantifies the amount of charge that is captured in the medium and cannot be completely released in a complete excitation-relaxation cycle, is directly related to the total amount of trap charge, and is an important indicator for characterizing insulation aging (increase of traps).

[0075] Short-Time Fourier Transform (STFT) is a time-frequency analysis tool. It converts a one-dimensional time signal (such as a polarization current response sequence) into a two-dimensional time-frequency spectrum through a sliding window function, which can simultaneously show the frequency components of the signal and their evolution over time (or within the test stage), and is very suitable for analyzing non-stationary relaxation current signals.

[0076] The time-frequency spectrum is the result of the STFT transform, which is a two-dimensional matrix or image. The horizontal axis represents time, and the vertical axis represents frequency. The intensity (color) of each point represents the energy or amplitude of a specific frequency component at a specific time.

[0077] The preset frequency band is an artificially divided frequency band according to the physical meaning of the relaxation of the insulating material, usually including a first frequency band and a second frequency band; wherein the first frequency band (high frequency band) corresponds to fast polarization / depolarization processes (such as electronic polarization, rapid reorientation of dipoles); the second frequency band (low frequency band) corresponds to slow polarization / depolarization processes (such as interfacial polarization, deep trap charge release). Aging often causes signal energy to shift to the second frequency band (low frequency).

[0078] The time window is a specific time period selected for integral calculation, usually selected to cover the main relaxation process to ensure that most of the response information is captured.

[0079] For example, after receiving the polarization current response sequence and the depolarization current response sequence from the test terminal, the joint analysis stage is entered, which aims to extract a set of characteristic parameters from the pair of twin data (polarization current response sequence and depolarization current response sequence) that can comprehensively and deeply characterize the insulation state of the impregnated fiber sleeve. The process first uses the time-frequency analysis tool of the short-time Fourier transform to process the polarization current response sequence and the depolarization current response sequence, respectively, to convert the one-dimensional time signal into a two-dimensional time-frequency spectrum (first time-frequency spectrum and second time-frequency spectrum), thereby clearly revealing the evolution law of different frequency components (corresponding to different speed physical processes) in the current signal over time in the time-frequency joint domain.

[0080] Further, multi-dimensional feature mining is performed based on the two time-frequency spectrums (the first time-frequency spectrum and the second time-frequency spectrum); first, for the second time-frequency spectrum (the depolarization process) that can more sensitively reflect aging-induced trap information, relaxation time constants and corresponding weight coefficients that characterize different speed charge release processes are extracted through spectral analysis or curve fitting technology, and the two parameters directly quantify the depth distribution and density of traps in the medium; second, to evaluate the evolution of energy distribution in the polarization and depolarization processes, the energy distribution ratios of the signals in preset frequency bands (usually including a high-frequency first frequency band corresponding to a fast process and a low-frequency second frequency band corresponding to a slow process) that are preliminarily divided according to physical meanings are calculated based on the first time-frequency spectrum and the second time-frequency spectrum, so as to obtain polarization energy features and depolarization energy features; aging often leads to energy aggregation to low frequencies (the second frequency band), and this feature can effectively capture this trend.

[0081] Finally, in order to directly quantify the unrecoverable charge loss in the medium, the integral difference between the first time-frequency spectrum and the second time-frequency spectrum is calculated in a preset time window covering the main relaxation process, so as to obtain an integral charge feature, which corresponds to the net charge amount permanently captured by deep-level traps in the medium in a complete test cycle and is a direct evidence of insulation degradation.

[0082] Through a series of consecutive operations, the original and massive time-series current data is compressed and refined into a set of multi-dimensional feature parameters with clear physical meaning and strong representation ability, thereby laying a solid data foundation for final insulation state classification and evaluation based on the features.

[0083] It should be noted that the polarization current sequence and the depolarization current sequence are fused and analyzed, and the feature parameters representing the insulation state are extracted. The beneficial effects are that the information is effectively fused and enhanced, the interference conductive current can be more accurately separated through joint analysis, the intrinsic polarization characteristics are obtained, and the time window for effective analysis is widened. On this basis, the method can extract multi-dimensional feature parameters (such as relaxation time spectrum, model equivalent parameters, etc.) that are strongly related to insulation aging, moisture and other states. These features with clear physical meaning constitute a bridge from raw data to state evaluation, and provide multi-dimensional quantitative criteria for accurate diagnosis.

[0084] S204, determining the insulation state evaluation result of the impregnated fiber sleeve sample based on the feature parameters.

[0085] In an implementation manner, the feature parameters are input into a pre-trained insulation state evaluation model to obtain the insulation state evaluation result of the impregnated fiber sleeve sample.

[0086] The insulation state evaluation model is constructed based on a machine learning algorithm and trained by historical samples with known aging states. The insulation state evaluation result includes insulation state category, aging degree level, and confidence information. The function of the insulation state evaluation model is to map the input feature parameters to the corresponding insulation state conclusion, which is to establish a complex nonlinear relationship between features and states by learning the patterns in a large number of historical data.

[0087] The machine learning algorithm is a mathematical and computational framework used to construct the insulation state evaluation model, such as a classification algorithm (e.g., support vector machine, random forest, gradient boosting tree, or neural network).

[0088] The historical samples with known aging states are the "teaching materials" for training the model. These samples include two parts: one is the feature parameters obtained from real or accelerated aging test samples of different aging degrees (e.g., no aging, slight aging, severe aging); the other is the real aging state label (i.e., the correct answer) of these test samples determined by authoritative methods (e.g., chemical analysis, long-term operation record).

[0089] Training is an important process for constructing the insulation state evaluation model. Through automatic adjustment of the parameters inside the model by the algorithm, the model can accurately predict the corresponding aging state label of the input historical sample features to the maximum extent. This process essentially learns the correlation rules between features and states.

[0090] The insulation state category is a qualitative judgment output by the insulation state evaluation model, usually a discrete category label, such as good, slight aging, severe aging, or normal, warning, and abnormal.

[0091] The aging degree level is a quantitative or semi-quantitative judgment output by the insulation state evaluation model, which is more detailed than the category. For example, within the slight aging category, it is further divided into level 1, level 2, and level 3, or represented in percentage form (e.g., aging index 70%).

[0092] The confidence information is a quantitative output of the model's judgment certainty. It is usually represented in the form of probability (e.g., the confidence of judging as severe aging is 92%). High confidence means that the model's features are highly consistent with the learned patterns; lower confidence suggests that the result may have uncertainty and needs manual review, which increases the reliability and interpretability of the evaluation result.

[0093] Exemplarily, after successfully extracting the set of characteristic parameters representing the dielectric behavior of the impregnated fiber, the final intelligent diagnosis and decision stage is entered, which is mainly based on a pre-trained insulation state evaluation model to determine the insulation state evaluation result of the sample. The insulation state evaluation model is a data-driven artificial intelligence model based on a machine learning algorithm, which obtains diagnostic ability by training on a large number of historical samples with known aging states. Since these historical samples include the characteristic parameters measured from samples of various aging states and the corresponding artificially labeled true state labels, the insulation state evaluation model can learn and internalize the complex mapping relationship between the characteristic parameters and the insulation aging.

[0094] During specific evaluation, the characteristic parameters of the current sample are input into the trained insulation state evaluation model; the insulation state evaluation model then intelligently analyzes and matches the input characteristics based on the learned knowledge, and outputs a structured insulation state evaluation result, which includes a clear insulation state category (such as good, slight aging or severe aging) to give a qualitative diagnostic conclusion. For further fine description, the insulation state evaluation result usually also outputs an aging degree level to realize semi-quantitative or quantitative differentiation within the qualitative category (for example, indicating the relative severity in slight aging); to enhance the credibility and practicality of the result, the insulation state evaluation result will also provide confidence information in the form of probability to reflect the degree of certainty of the judgment, and high-confidence results can be directly used to guide operation and maintenance decisions, while low-confidence results can trigger an alarm and suggest manual review.

[0095] At this point, through the automated process from feature extraction to model inference, the intelligent transformation from the original current signal to the insulation state evaluation result with clear physical meaning and directly usable for decision-making is completed, and the rapid, objective and quantifiable accurate evaluation of the insulation state of the impregnated fiber sleeve is realized.

[0096] It should be noted that the insulation state evaluation result of the sample is finally determined by threshold comparison, trend analysis or intelligent algorithm mapping based on the extracted characteristic parameters. Its beneficial effect is to complete the intelligent mapping and comprehensive diagnosis from physical characteristics to engineering state, and to realize the quantification and refinement of the evaluation (such as determining the degree of moisture, aging level), rather than a simple pass / fail binary judgment. This enhances the ability of early fault warning and can directly provide actionable decision guidance such as enhanced monitoring and planned maintenance for predictive maintenance of the equipment, forming a closed loop from testing to maintenance.

[0097] Based on the above, the impregnated fiber bush insulation state evaluation method, by issuing test configuration instructions including indicating the application of a preset parameter of a bipolar square wave voltage excitation to the impregnated fiber bush sample, collecting the polarization current response sequence of the polarization stage, and collecting the depolarization current response sequence of the depolarization stage to the test terminal, realizes the evaluation of the insulation state without accelerating the insulation aging or causing damage. Further, the test data uploaded from the test terminal is received, providing sufficient data conditions for subsequent use of multiple signal processing algorithms. Then, the polarization current response sequence and the depolarization current response sequence can be jointly analyzed to obtain characteristic parameters, more accurately extract the direct current component representing insulation aging or moisture, and provide multi-dimensional criteria for state evaluation. Finally, based on the characteristic parameters, the insulation state evaluation result of the impregnated fiber bush sample is determined. The scheme realizes non-destructive testing and pre-diagnosis by introducing the test terminal. Moreover, by jointly analyzing the polarization current response sequence and the depolarization current response sequence, the wide time domain relaxation spectrum from milliseconds to thousands of seconds can be more completely reconstructed, and finally the accurate evaluation result is obtained.

[0098] The above Figures 1 to 2 The impregnated fiber bush insulation state evaluation method provided by the embodiments of the present application is described in detail, and the device and equipment provided by the embodiments of the present application will be introduced below with reference to the drawings.

[0099] The embodiments of the present application also provide an impregnated fiber bush insulation state evaluation device, as shown in Figure 3 The device includes: The instruction issuing module 301 is configured to issue test configuration instructions to the test terminal. The test terminal includes a high-frequency square wave power supply, a high-voltage relay, a current measurement unit, and a data processing unit. The test configuration instructions include instructions for applying a preset parameter of a bipolar square wave voltage excitation to the impregnated fiber bush sample, collecting the polarization current response sequence of the polarization stage, and collecting the depolarization current response sequence of the depolarization stage. The data receiving module 302 is configured to receive test data uploaded from the test terminal. The test data includes the polarization current response sequence collected in the polarization stage and the depolarization current response sequence collected in the depolarization stage. The analysis module 303 is configured to jointly analyze the polarization current response sequence and the depolarization current response sequence to obtain characteristic parameters. The evaluation module 304 is configured to determine the insulation state evaluation result of the impregnated fiber bush sample based on the characteristic parameters.

[0100] In some possible implementation manners, the preset parameters include a frequency, a duty cycle, and a voltage amplitude of the square wave voltage.

[0101] In some possible implementation manners, the analysis module 303 is specifically configured to: perform short-time Fourier transform on the polarization current response sequence to obtain a first time-frequency spectrum, and perform short-time Fourier transform on the depolarization current response sequence to obtain a second time-frequency spectrum; extract, based on the second time-frequency spectrum, relaxation time constants and weight coefficients corresponding to the relaxation time constants; based on the first time-frequency spectrum, calculate a first energy distribution ratio of the polarization current response sequence in a preset frequency band to obtain a polarization energy feature; the preset frequency band includes a first frequency band and a second frequency band; based on the second time-frequency spectrum, calculate a second energy distribution ratio of the depolarization current response sequence in the preset frequency band to obtain a depolarization energy feature; in a preset time window, calculate an integral difference value between the first time-frequency spectrum and the second time-frequency spectrum to obtain an integral charge feature.

[0102] In some possible implementation manners, the impregnated fiber bushing sample is of a preset size and is subjected to vacuum drying treatment.

[0103] In some possible implementation manners, the evaluation module 304 is specifically configured to: input the feature parameters into a pre-trained insulation state evaluation model to obtain an insulation state evaluation result of the impregnated fiber bushing sample; the insulation state evaluation model is constructed based on a machine learning algorithm and is trained through historical samples of known aging states; the insulation state evaluation result includes an insulation state category, an aging degree grade, and confidence information.

[0104] In some possible implementation manners, the impregnated fiber bushing sample has a water content lower than a preset water content threshold.

[0105] The impregnated fiber bushing insulation state evaluation apparatus according to the embodiments of the present application can correspond to performing the methods described in the embodiments of the present application, and the above other operations and / or functions of each module / unit of the impregnated fiber bushing insulation state evaluation apparatus are respectively for realizing Figure 1 the corresponding procedures of the methods in the embodiments, which will not be described herein again for brevity.

[0106] The embodiments of the present application further provide a computing device. As Figure 4 shown, the figure is a schematic diagram of a computing device according to an embodiment of the present application. The computing device 400 includes a bus 401, a processor 402, a communication interface 403, and a memory 404. The processor 402, the memory 404, and the communication interface 403 communicate through the bus 401.

[0107] Bus 401 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 4 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0108] Processor 402 can be any one or more of the following processors: central processing unit (CPU), graphics processing unit (GPU), microprocessor (MP), or digital signal processor (DSP).

[0109] Communication interface 403 is used for communication with external devices.

[0110] Memory 404 may include volatile memory, such as random access memory (RAM). Memory 404 may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state drive (SSD).

[0111] The memory 404 stores executable code, and the processor 402 executes the executable code to perform the aforementioned method for evaluating the insulation status of the adhesive-impregnated fiber optic sleeve.

[0112] Specifically, in achieving Figure 3 In the case of the illustrated embodiment, and Figure 3 When the modules or units of the adhesive-impregnated fiber optic sleeve insulation condition assessment device described in the embodiments are implemented by software, the following steps are performed: Figure 3 The software or program code required for the functions of each module / unit can be partially or entirely stored in memory 404. Processor 402 executes the program code corresponding to each unit stored in memory 404 to perform the aforementioned method for evaluating the insulation status of adhesive-impregnated fiber optic sleeves.

[0113] The embodiments of the present application also provide a computer readable storage medium. The computer readable storage medium can be any available medium or data storage device that can be used to store instructions that can be executed by a computing device. The computer readable storage medium can be a magnetic-based medium (e.g., a floppy diskette, a hard disk drive, a magnetic tape), an optical-based medium (e.g., a compact disc, a DVD), or a semiconductor-based medium (e.g., a solid-state drive), etc. The computer readable storage medium includes instructions that are executable by a computing device to perform the insulation state evaluation method of the impregnated fiber sleeve.

[0114] The embodiments of the present application also provide a computer program product including one or more computer instructions. When the computer instructions are loaded and executed on a computing device, the computer instructions generate, in whole or in part, the processes or functions described in the embodiments of the present application.

[0115] The computer instructions can be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions can be transmitted from one website, computer or data center to another website, computer or data center through a wired (e.g., coaxial cable, optical fiber) or wireless (e.g., infrared, wireless, microwave, etc.) manner.

[0116] The computer program product is executed by a computer, and the computer executes any of the insulation state evaluation methods of the impregnated fiber sleeve. The computer program product can be a software installation package, and when any of the insulation state evaluation methods of the impregnated fiber sleeve is needed, the computer program product can be downloaded and executed on the computer.

[0117] The descriptions of the processes or structures corresponding to the above respective figures are each focused on, and the parts not described in detail in a certain process or structure can be referred to the related descriptions of other processes or structures.

[0118] The above only describes the specific embodiments of the present application, but the protection scope of the present application is not limited to this. Any changes or replacements within the technical scope disclosed in the present application should be covered within the protection scope of the present application.

Claims

1. A method for assessing the insulation condition of resin-impregnated fiber optic sleeves, characterized in that, The method includes: Test configuration instructions are issued to the test terminal; wherein the test terminal includes a high-frequency square wave power supply, a high-voltage relay, a current measurement unit and a data processing unit; the test configuration instructions include instructing the application of a bipolar square wave voltage excitation with preset parameters to the adhesive-impregnated fiber sleeve sample, acquiring the polarization current response sequence during the polarization stage, and acquiring the depolarization current response sequence during the depolarization stage. Receive test data uploaded from the test terminal; wherein the test data includes a polarization current response sequence collected during the polarization phase and a depolarization current response sequence collected during the depolarization phase; The characteristic parameters are obtained by jointly analyzing the polarization current response sequence and the depolarization current response sequence. Based on the aforementioned characteristic parameters, the insulation status assessment results of the impregnated fiber sheath sample are determined.

2. The method according to claim 1, characterized in that, The preset parameters include square wave voltage frequency, duty cycle, and voltage amplitude.

3. The method according to claim 1, characterized in that, in, The characteristic parameters include relaxation time constants, weighting coefficients corresponding to each relaxation time constant, polarization energy characteristics, depolarization energy characteristics, and integrated charge characteristics. The characteristic parameters obtained by jointly analyzing the polarization current response sequence and the depolarization current response sequence include: The first time spectrum is obtained by performing a short-time Fourier transform on the polarization current response sequence, and the second time spectrum is obtained by performing a short-time Fourier transform on the depolarization current response sequence. Based on the second time spectrum, the relaxation time constant and the weighting coefficients corresponding to each relaxation time constant are extracted; Based on the first time spectrum, the first energy distribution ratio of the polarization current response sequence within a preset frequency band is calculated to obtain the polarization energy characteristics; wherein, the preset frequency band includes a first frequency band and a second frequency band; Based on the second time spectrum, the second energy distribution ratio of the depolarization current response sequence within a preset frequency band is calculated to obtain the depolarization energy characteristics; Within a preset time window, the integral difference between the first time spectrum and the second time spectrum is calculated to obtain the integral charge characteristics.

4. The method according to claim 1, characterized in that, The resin-impregnated fiber sleeve sample is of a predetermined size and has undergone vacuum drying treatment.

5. The method according to claim 1, characterized in that, The determination of the insulation status assessment result of the adhesive-impregnated fiber optic sleeve sample based on the characteristic parameters includes: The feature parameters are input into a pre-trained insulation condition assessment model to obtain the insulation condition assessment results of the impregnated fiber sheath sample; wherein, the insulation condition assessment model is constructed based on a machine learning algorithm and trained using historical samples with known aging conditions; the insulation condition assessment results include insulation condition category, aging degree level and confidence information.

6. The method according to claim 1, characterized in that, The moisture content of the impregnated fiber sleeve sample was lower than the preset moisture content threshold.

7. A device for assessing the insulation condition of resin-impregnated fiber optic tubing, characterized in that, The device includes: The instruction issuing module is used to issue test configuration instructions to the test terminal; wherein, the test terminal includes a high-frequency square wave power supply, a high-voltage relay, a current measurement unit and a data processing unit; the test configuration instructions include instructing the application of a bipolar square wave voltage excitation with preset parameters to the adhesive-impregnated fiber sleeve sample, acquiring the polarization current response sequence during the polarization stage, and acquiring the depolarization current response sequence during the depolarization stage. The data receiving module is used to receive test data uploaded from the test terminal; wherein the test data includes a polarization current response sequence collected during the polarization phase and a depolarization current response sequence collected during the depolarization phase. The analysis module is used to perform joint analysis on the polarization current response sequence and the depolarization current response sequence to obtain characteristic parameters; An evaluation module is used to determine the insulation status evaluation result of the adhesive-impregnated fiber sheath sample based on the characteristic parameters.

8. A computing device, characterized in that, Including memory and processor; The memory stores one or more computer programs, the one or more computer programs including instructions; when the instructions are executed by the processor, the computing device performs the method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program for performing the method as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, The computer program product includes one or more computer instructions, which, when executed by a computer, perform the method as described in any one of claims 1 to 6.

Images