A transformer internal state on-line monitoring method and related device

Abstract

This invention belongs to the field of transformer monitoring technology and discloses an online monitoring method and related device for the internal condition of a transformer. By injecting an AC micro-current excitation signal into the low-voltage side of the transformer, voltage response, vibration response, and background power frequency current signals are simultaneously acquired. After frequency domain processing, the electromagnetic impedance spectrum and vibration transfer function spectrum are obtained. Based on the initial health state reference spectrum, the normalized electromagnetic-mechanical coupling coefficient across the entire frequency band is calculated, and a comprehensive health index is calculated by combining the winding mechanical resonance frequency, thus achieving condition monitoring. This method realizes real-time online non-destructive monitoring of transformers under normal operation, continuously tracking equipment status without power outages. Through the collaborative sensing of multiple physical quantities, the detection sensitivity for early minor winding deformation and core loosening is significantly improved, effectively capturing fault precursors and comprehensively enhancing the accuracy and reliability of condition assessment, meeting the high-precision and real-time requirements of the power grid for transformer operation and maintenance.

Description

Technical Field

[0001] This invention belongs to the field of transformer monitoring technology, and particularly relates to an online monitoring method and related device for the internal condition of a transformer. Background Technology

[0002] Transformers are indispensable core energy conversion and transmission devices in power systems. They undertake critical functions such as voltage transformation, power transmission, electrical isolation, and impedance matching, spanning the entire process of power generation, transmission, distribution, and consumption. Their operational reliability directly determines the safety, stability, and continuity of power supply in the power grid. During long-term operation, transformers are prone to various internal faults, such as winding deformation, inter-turn short circuits, multi-point grounding of the core, and insulation deterioration. These faults typically exhibit a gradual development pattern, with initial fault symptoms being relatively insidious. However, once they develop into serious faults, they can not only damage the transformer itself but also trigger large-scale power outages or even catastrophic power grid accidents. Therefore, effectively monitoring the internal operating status of transformers and promptly detecting early signs of faults has become a critical requirement for ensuring the safe and efficient operation of the power system.

[0003] Currently, there are still many technical bottlenecks in the industry for monitoring the internal condition of transformers, making it difficult to meet the power grid's requirements for high precision and real-time equipment operation and maintenance. Current monitoring methods are mainly offline detection. Traditional methods such as frequency response analysis (FRA) and short-circuit impedance testing require the transformer to be de-energized, making continuous online monitoring during normal operation impossible. This can easily lead to missing critical stages of fault development and causing the fault to escalate. Furthermore, existing monitoring methods often employ single physical quantity monitoring. While oil chromatography (DGA) is sensitive to electrical faults, it exhibits a significant lag in responding to mechanical deformation. Vibration monitoring is sensitive to mechanical loosening but is susceptible to external interference and its correlation with electrical status is unclear. Partial discharge monitoring can effectively identify insulation defects but is insensitive to early mechanical deformation. All these single monitoring methods have significant limitations. In addition, transformers have complex electromagnetic-mechanical-thermal multi-physics coupling effects. Existing single monitoring methods cannot fully reflect changes in equipment status under these coupling relationships, making it difficult to accurately assess the overall equipment status. For early mechanical faults such as minor winding deformation (<1%) and slight core loosening, the changes in traditional electrical parameters are extremely weak, making early and accurate detection difficult with current technology, and timely fault warnings impossible.

[0004] In summary, existing transformer internal condition monitoring technologies cannot achieve real-time, non-destructive, multi-dimensional collaborative monitoring under normal equipment operation, making it difficult to effectively capture early signs of faults and failing to meet the high requirements of the power grid for transformer operational reliability. Summary of the Invention

[0005] This invention provides a method and related device for online monitoring of the internal condition of a transformer. This method can effectively solve the problems of existing transformer internal condition monitoring technologies being unable to achieve real-time, non-destructive, multi-dimensional collaborative monitoring under normal equipment operation, and being difficult to effectively capture early signs of faults. It can meet the high requirements of the power grid for the reliability of transformer operation.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: A method for online monitoring of the internal condition of a transformer, comprising: An AC micro-current excitation signal is injected into the low-voltage side of the transformer, and the voltage response signal, the vibration response signal of the transformer tank surface, and the background power frequency current signal are collected simultaneously. The collected voltage response signal, vibration response signal, background power frequency current signal, and injected AC micro-current excitation signal were processed in the frequency domain to calculate the electromagnetic impedance spectrum and vibration transfer function spectrum of the transformer. Based on the preset electromagnetic impedance reference spectrum and vibration transfer function reference spectrum under the initial healthy state of the transformer, and combined with the electromagnetic impedance spectrum and vibration transfer function spectrum, the normalized electromagnetic-mechanical coupling coefficient of the entire frequency band is calculated. Based on the electromagnetic impedance spectrum, the normalized electromagnetic-mechanical coupling coefficient, and the mechanical resonance frequency of the transformer windings identified from the initial spectrum, the comprehensive health index of the transformer is calculated; the internal condition of the transformer is monitored based on the comprehensive health index.

[0007] Furthermore, the frequency domain processing of the acquired voltage response signal, vibration response signal, background power frequency current signal, and injected AC micro-current excitation signal to calculate the electromagnetic impedance spectrum and vibration transfer function spectrum of the transformer includes: Based on the voltage response signal, background power frequency current signal and injected AC micro-current excitation signal, the excitation response is separated by frequency domain subspace decomposition method, and the electromagnetic impedance spectrum of the transformer is calculated. Based on the vibration response signal and the injected AC microcurrent excitation signal, the vibration transfer function spectrum, which characterizes the coupling efficiency from electromagnetic force to mechanical vibration, is calculated.

[0008] Furthermore, based on the voltage response signal, background power frequency current signal, and injected AC micro-current excitation signal, the excitation response is separated using the frequency domain subspace decomposition method, and the electromagnetic impedance spectrum of the transformer is calculated. The specific formula is as follows:

[0009] In the formula, For frequency Electromagnetic impedance at the location; Indicates the signal at frequency Fourier transform coefficients at the location; It is a voltage response signal; Background power frequency current signal; Pass function for background; To inject an excitation current signal.

[0010] Furthermore, based on the vibration response signal and the injected AC micro-current excitation signal, the vibration transfer function spectrum, which characterizes the coupling efficiency from electromagnetic force to mechanical vibration, is calculated. The specific formula is as follows:

[0011] In the formula, For frequency Vibration transfer function at the location; Indicates the signal at frequency Fourier transform coefficients at the location; It is a vibration response signal; To inject an excitation current signal; The sensitivity of the accelerometer; Install coupling coefficients for the sensor.

[0012] Furthermore, the normalized electromagnetic-mechanical coupling coefficient across the entire frequency band is calculated by combining the electromagnetic impedance spectrum and the vibration transfer function spectrum based on the preset electromagnetic impedance spectrum and vibration transfer function spectrum under the initial healthy state of the transformer, including: Calculate the frequency separately Calculate the ratio between the amplitude ratios of the vibration transfer function spectrum and the vibration transfer function reference spectrum, and the amplitude ratio of the electromagnetic impedance spectrum and the electromagnetic impedance reference spectrum. Multiply the ratio between the two amplitude ratios by the ratio of the current frequency to the reference frequency, and by the Gaussian window function value based on the mechanical resonance frequency of the transformer winding, to obtain the normalized electromagnetic-mechanical coupling coefficient at the frequency. The specific formula is as follows:

[0013] In the formula, For frequency The normalized electromagnetic-mechanical coupling coefficient at the location; This is the vibration transfer function of the transformer measured under reference conditions; This is the electromagnetic impedance of the transformer measured under reference conditions; For frequency Electromagnetic impedance at the location; For frequency Vibration transfer function at the location; The current frequency point; For reference frequency; This is the mechanical resonant frequency of the transformer winding; The resonant bandwidth; This represents the Gaussian window function.

[0014] Furthermore, the comprehensive health index of the transformer, calculated based on the electromagnetic impedance spectrum, the normalized electromagnetic-mechanical coupling coefficient, and the mechanical resonant frequency of the transformer winding identified from the initial spectrum, includes: A comprehensive health index for transformers, reflecting their internal condition, is constructed using the following formula:

[0015] In the formula, This indicates the overall health index of the transformer; This represents the normalized change in electrical state; This represents the normalized change in the coupling coefficient; This represents the normalized offset of the resonant frequency; , , These are the corresponding weight coefficients; The normalized change in electrical state is calculated based on the electromagnetic impedance spectrum, and the specific formula is as follows:

[0016] In the formula, Number of frequency points; Frequency point index; For frequency The magnitude of the electromagnetic impedance at that point; This represents the amplitude of the electromagnetic impedance measured by the transformer under reference conditions. This represents the standard deviation of the electromagnetic impedance amplitude measured multiple times under reference conditions. The normalized change in the coupling coefficient is calculated based on the normalized electromagnetic-mechanical coupling coefficient, and the specific formula is as follows:

[0017] In the formula, is the number of major mechanical resonance frequency points; m is the index of the major mechanical resonance frequency points; For the first The center frequency of each resonance peak; The coupling coefficient at the resonant frequency in the current state; The coupling coefficient is the reference state. The normalized offset of the resonance frequency is calculated based on the mechanical resonance frequency of the transformer winding, and the specific formula is as follows:

[0018] In the formula, Let be the mechanical resonant frequency of the transformer winding, representing the th The center frequency of each resonance peak in a healthy state; This represents the full width at half maximum (FWHM) of the corresponding resonance peak under the baseline condition.

[0019] Furthermore, the monitoring of the transformer's internal condition based on a comprehensive health index includes: The current internal condition of the transformer is determined based on a comprehensive health index, as follows: If the overall health index is 0, then the current internal state of the transformer is judged to be healthy. If the overall health index is within the first preset range, the current internal state of the transformer is judged to be slightly abnormal. If the overall health index is within the second preset range, the current internal condition of the transformer is judged to be moderately abnormal. If the overall health index is in the third preset range, the current internal state of the transformer is judged to be seriously abnormal.

[0020] An online monitoring system for the internal condition of a transformer, comprising: The data acquisition module is used to inject AC micro-current excitation signals into the low-voltage side of the transformer and simultaneously acquire voltage response signals, vibration response signals on the surface of the transformer tank, and background power frequency current signals. The frequency domain processing module is used to process the acquired voltage response signal, vibration response signal, background power frequency current signal and injected AC micro-current excitation signal in the frequency domain, and calculate the electromagnetic impedance spectrum and vibration transfer function spectrum of the transformer respectively. The first calculation module is used to calculate the normalized electromagnetic-mechanical coupling coefficient across the entire frequency band based on the preset electromagnetic impedance reference spectrum and vibration transfer function reference spectrum under the initial healthy state of the transformer, combined with the electromagnetic impedance spectrum and vibration transfer function spectrum. The second calculation module is used to calculate the transformer's comprehensive health index based on the electromagnetic impedance spectrum, the normalized electromagnetic-mechanical coupling coefficient, and the transformer winding mechanical resonance frequency identified from the initial spectrum; and to monitor the internal condition of the transformer based on the comprehensive health index.

[0021] An online monitoring device for the internal condition of a transformer, comprising: Memory, used to store computer programs; A processor is used to implement the above-described online monitoring method for the internal state of a transformer when executing the computer program.

[0022] A computer-readable storage medium storing a computer program, which, when executed by a processor, is used for the above-described online monitoring method for the internal condition of a transformer.

[0023] Compared with the prior art, the present invention has the following beneficial effects: This invention provides an online monitoring method for the internal condition of a transformer. By injecting an AC micro-current excitation signal into the low-voltage side of the transformer, voltage response, vibration response, and background power frequency current signals are simultaneously acquired. After frequency domain processing, the electromagnetic impedance spectrum and vibration transfer function spectrum are obtained. Based on the initial health state reference spectrum, the normalized electromagnetic-mechanical coupling coefficient across the entire frequency band is calculated. Combined with the winding mechanical resonance frequency, a comprehensive health index is calculated to achieve condition monitoring. The micro-current excitation stimulates the electromagnetic-mechanical coupling effect inside the transformer. Through simultaneous acquisition of multiple signals and frequency domain analysis, the integrated characteristics reflecting both electrical and mechanical conditions are extracted. The normalized coupling coefficient quantifies the dynamic changes in electromagnetic and mechanical interactions, while the comprehensive health index integrates multi-dimensional information, thus overcoming the limitations of monitoring a single physical quantity. This method achieves real-time online non-destructive monitoring of the transformer under normal operation, continuously tracking equipment status without power outages. Through the collaborative sensing of multiple physical quantities, the detection sensitivity for early minor winding deformation and core loosening is significantly improved, effectively capturing fault precursors and comprehensively enhancing the accuracy and reliability of condition assessment, meeting the high-precision and real-time requirements of the power grid for transformer operation and maintenance. Attached Figure Description

[0024] Figure 1 This is a core flowchart of an online monitoring method for the internal condition of a transformer provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of an online monitoring system for the internal condition of a transformer, provided in an embodiment of the present invention. Detailed Implementation

[0025] To further understand the content of this invention, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments are merely illustrative and not limiting of the invention.

[0026] The technical terms involved in this invention are explained below: FRA stands for Frequency Response Analysis.

[0027] DGA stands for Dissolved Gas Analysis, which refers to dissolved gas analysis, also known as oil chromatography.

[0028] IEC 61850 is a communication standard for power system automation developed by the International Electrotechnical Commission (IEC).

[0029] Modbus is a serial communication protocol.

[0030] FRA stands for Frequency Response Analysis.

[0031] As described in the background section, existing monitoring methods for the internal condition of transformers have the following technical shortcomings: First, some methods rely primarily on offline detection. For example, frequency response analysis (FRA) and short-circuit impedance testing require power outages, making continuous online monitoring impossible and potentially missing critical periods of fault development. Second, oil chromatography (DGA) is sensitive to electrical faults but lags behind in responding to mechanical deformation; vibration monitoring is sensitive to mechanical loosening but susceptible to external interference and its correlation with electrical condition is unclear; partial discharge monitoring is sensitive to insulation defects but insensitive to early mechanical deformation. All of these methods have the limitation of monitoring only a single physical quantity. Third, there is a lack of multi-physics coupling analysis: electromagnetic, mechanical, and thermal multi-physics coupling exists inside transformers, and existing single monitoring methods cannot fully reflect state changes under this coupling relationship. Fourth, early fault detection is difficult: early mechanical faults such as minor winding deformation (<1%) and slight core loosening do not show significant changes in traditional electrical parameters, making them difficult to detect in a timely manner.

[0032] To address the aforementioned issues, this embodiment provides an online monitoring method for the internal state of a transformer. This monitoring method can monitor the coordinated changes in the internal mechanical and electrical states of a transformer in real time and without damage during normal operation, thereby enabling early warning of potential faults.

[0033] For example, such as Figure 1 As shown, this embodiment provides a method for online monitoring of the internal condition of a transformer, including: An AC micro-current excitation signal is injected into the low-voltage side of the transformer, and the voltage response signal, the vibration response signal of the transformer tank surface, and the background power frequency current signal are collected simultaneously. The collected voltage response signal, vibration response signal, background power frequency current signal, and injected AC micro-current excitation signal were processed in the frequency domain to calculate the electromagnetic impedance spectrum and vibration transfer function spectrum of the transformer. Based on the preset electromagnetic impedance reference spectrum and vibration transfer function reference spectrum under the initial healthy state of the transformer, and combined with the electromagnetic impedance spectrum and vibration transfer function spectrum, the normalized electromagnetic-mechanical coupling coefficient of the entire frequency band is calculated. Based on the electromagnetic impedance spectrum, the normalized electromagnetic-mechanical coupling coefficient, and the mechanical resonance frequency of the transformer windings identified from the initial spectrum, the comprehensive health index of the transformer is calculated; the internal condition of the transformer is monitored based on the comprehensive health index.

[0034] The monitoring method provided in this embodiment will be further explained below: For example, this embodiment provides an online monitoring method for the internal state of a transformer. This method employs an online monitoring approach for the internal state of a transformer based on the fusion of electromagnetic, acoustic, and vibration multi-physics fields. It can solve the problem that existing monitoring methods cannot simultaneously monitor the coordinated changes in the mechanical and electrical states inside the transformer. This method injects a micro-current excitation of a specific spectrum into the transformer, simultaneously collects the electrical response and the vibration response of the tank surface, constructs a fusion feature of multi-physics field coupling, and extracts a health index that is highly sensitive to changes in the internal state. The specific steps are as follows: Step 1: Synchronous acquisition of controllable electromagnetic excitation and multi-physics response: A set of AC microcurrent excitation signals with a frequency range of 10 Hz to 10 kHz is injected on the low-voltage side of the transformer (through the bushing end screen or a dedicated injection interface). Its amplitude is controlled between 0.1% and 1% of the transformer's rated current to avoid affecting the transformer's normal operation. Simultaneously, the following response signals are collected: Voltage response signal , measured at the injection point, with the dimension of volt (V); Vibration response signal The output is a voltage signal measured by a triaxial accelerometer installed on the surface of the transformer tank, with the dimension of volt (V). Background power frequency current signal The current is measured by the current transformer on the high-voltage side of the transformer, and its dimension is ampere (A).

[0035] All signals are synchronized at the microsecond level via GPS or a hardware-synchronized clock.

[0036] Step 2: Multiphysics Response Separation and Feature Extraction 1. Electromagnetic impedance spectrum calculation: The input electromagnetic impedance spectrum of the transformer is calculated from the injected excitation current and the measured voltage response. Considering the strong power frequency background interference, the excitation response is separated using the frequency domain subspace decomposition method, as shown in Equation 1 below:

[0037] In the formula, For frequency The electromagnetic impedance at the point is in ohms (Ω). Indicates the signal at frequency Fourier transform coefficients at the location; It is a voltage response signal, with the dimension of volts (V). The background power frequency current signal is expressed in amperes (A). The background transfer function represents the transfer relationship from the background current to the measured voltage. Its dimension is volts per ampere (V / A), and its value is obtained by measuring the frequency domain ratio of the background voltage and the background current when the excitation is off. The injected excitation current signal is in the ampere (A); where, for a time signal... Its Fourier transform ,coefficient Dimensions of the original signal same.

[0038] It can be seen that the above formula, by subtracting the background interference term, extracts the impedance response generated only by microcurrent excitation, which significantly improves the signal-to-noise ratio of impedance measurement of in-operation transformers.

[0039] 2. Calculation of vibration transfer function spectrum: The transfer function between vibration acceleration response and excitation current is defined to characterize the coupling efficiency from electromagnetic force to mechanical vibration, as shown in Equation 2 below:

[0040] In the formula, For frequency The vibration transfer function at point , with dimensions of meters per square second per ampere (m / (s)). 2 ·A)); Indicates the signal at frequency Fourier transform coefficients at the location; The vibration response signal is the voltage signal output by the accelerometer (without sensitivity conversion), and its dimension is volt (V). To inject an excitation current signal; The sensitivity of the accelerometer is measured in volts per meter per square second (V / (m / s²)). 2 ), representing unit acceleration (1 m / s²). 2 The output voltage generated. This value is the sensor's factory calibration parameter; The coupling coefficient for the sensor is dimensionless and ranges from 0 to 1, representing the mechanical coupling efficiency between the sensor and the surface of the fuel tank. This value is obtained through impact testing or finite element simulation calibration.

[0041] It can be seen that the above transfer function is directly related to the injected current (generating electromagnetic force) and the tank vibration (mechanical response), and its changes can sensitively reflect mechanical states such as changes in winding clamping force and loosening of iron core clamps.

[0042] Step 3: Calculation of normalized electromagnetic-mechanical coupling coefficient: To quantify the degree of coupling between electromagnetic and mechanical states and eliminate the influence of dimensions, a normalized electromagnetic-mechanical coupling coefficient is defined, as shown in Formula 3 below:

[0043] In the formula, For frequency The normalized electromagnetic-mechanical coupling coefficient at the point is dimensionless. The vibration transfer function of the transformer measured under reference conditions has dimensions in m / (s). 2 ·A); The electromagnetic impedance of the transformer is measured under reference conditions, with dimensions in Ω. For frequency Electromagnetic impedance at the location; For frequency Vibration transfer function at the location; The current frequency is expressed in Hertz (Hz). The reference frequency is in Hertz (Hz). The mechanical resonant frequency of the transformer winding is expressed in Hertz (Hz). The resonant bandwidth, with dimensions in Hertz (Hz), is typically taken as... 10%; This represents the Gaussian window function. The reference state of the transformer is also called the initial healthy state.

[0044] This demonstrates that the normalized electromagnetic-mechanical coupling coefficient is particularly sensitive to changes in coupling strength near the transformer's mechanical resonant frequency. When the windings become loose or deformed, not only the resonant frequency... It will shift, coupling coefficient The amplitude and distribution will also undergo characteristic changes.

[0045] Step 4: Calculation of Transformer Comprehensive Health Index: Key indicators are extracted from the above characteristics to construct a comprehensive transformer health index that reflects the internal condition of the transformer, as shown in Formula 4 below:

[0046] In the formula, This indicates the overall health index of the transformer; This represents the normalized change in electrical state; This represents the normalized change in the coupling coefficient; This represents the normalized offset of the resonant frequency; , , These are the corresponding weight coefficients; The components are defined as follows: Normalized change in electrical state (Dimensionless):

[0047] In the formula, The number of frequency points being analyzed. This represents the amplitude of the electromagnetic impedance being measured, with dimensions in Ω.

[0048] The electromagnetic impedance amplitude is given in the reference state, with dimensions in Ω.

[0049] Ω represents the standard deviation of the electromagnetic impedance amplitude measured multiple times under reference conditions.

[0050] Normalized variation of coupling coefficients (Dimensionless):

[0051] In the formula, The number of main mechanical resonance frequency points (usually 2 to 3).

[0052] For the first The center frequency of each resonance peak has the dimension of Hz.

[0053] The coupling coefficient at the resonant frequency in the current state is dimensionless.

[0054] The coupling coefficient is a dimensionless coefficient under healthy baseline conditions.

[0055] Specifically, the normalized offset of the resonant frequency (Dimensionless):

[0056] In the formula, For the first The center frequency of each resonance peak in a healthy state, with dimensions in Hz.

[0057] is the full width at half maximum (FWHM) of the corresponding resonance peak under the reference condition, with the dimension Hz.

[0058] Among the weighting coefficients mentioned above, , , All are dimensionless and satisfy the following conditions: The weighting coefficients can be adjusted based on the transformer type (such as reactors, power transformers) and fault history.

[0059] The Transformer Comprehensive Health Index (THI) can be further explained as follows: THI≈0: Healthy condition (consistent with the condition of a new transformer or after a major overhaul).

[0060] 0 < THI ≤ 0.3: Minor anomaly, it is recommended to strengthen monitoring.

[0061] 0.3 < THI ≤ 0.6: Moderate anomaly, it is recommended to carry out planned maintenance.

[0062] THI > 0.6: Severe anomaly, immediately power off for inspection.

[0063] Exemplarily, this embodiment also provides an implementation process for on-line monitoring of the internal state of the transformer, which is as follows: To implement the above monitoring method, this embodiment provides the following hardware facilities: Incentive injection unit, including: Programmable power amplifier: frequency range 10 Hz - 10 kHz, maximum output current 10 A (adjustable according to the transformer capacity), coupled to the low-voltage side of the transformer through an isolation transformer.

[0064] Injection interface: special junction box, connected to the end screen of the transformer bushing or the neutral point of the low-voltage winding.

[0065] Signal acquisition unit, including: Voltage sensor: differential input, bandwidth DC - 20 kHz, accuracy 0.1%.

[0066] Three-axis acceleration sensor: frequency range 0.5 Hz - 10 kHz, sensitivity 100 mV / g, installed at key positions on the side and top of the oil tank (at least 3 measurement points).

[0067] Current transformer: used to monitor the power frequency background current.

[0068] Synchronous data acquisition card: 24-bit resolution, synchronous sampling rate ≥ 50 kHz, number of channels not less than 8.

[0069] Data processing unit, including: Industrial control computer or embedded processor, running algorithm software.

[0070] Communication module: supports protocols such as IEC61850, MODBUS, etc., and accesses the substation monitoring system.

[0071] In this embodiment, the specific steps of the implementation process are as follows: 1. Install all sensors when the transformer is in a power-off state (such as after new commissioning or overhaul).

[0072] 2. Conduct the first comprehensive test to obtain the healthy benchmark spectrum: Inject incentives at different frequency points and measure and .

[0073] Identify the mechanical resonance frequency and coupling coefficient .

[0074] Repeat the measurement 5 times and calculate the standard deviation at each frequency point. .

[0075] Step S2: Online monitoring started: 1. After the transformer is put into operation, the system enters online monitoring mode.

[0076] 2. Automatically execute a stimulus-response test every 24 hours (can also be triggered manually): Initiate a 10-minute test procedure during a period when the transformer load is relatively stable (such as midnight).

[0077] Sinusoidal currents at 10 characteristic frequency points (covering 10 Hz-10 kHz) are injected sequentially.

[0078] All response signals are collected synchronously.

[0079] Step S3: Data Processing and Feature Extraction 1. Preprocess the collected data: remove power frequency interference, remove trends, and add windows.

[0080] 2. Execute formulas 1 and 2 to calculate the current state. and .

[0081] 3. Identify the current mechanical resonance frequency using a spectral peak search algorithm. .

[0082] 4. Calculate the normalized electromagnetic-mechanical coupling coefficient using Formula 3. .

[0083] Step S4: Health Status Assessment 1. Compare the current characteristics with the health benchmark and substitute them into Formula 4 to calculate the THI index.

[0084] 2. Fault type identification based on THI value and characteristic change patterns: Axial deformation of windings: mainly manifested in the high-frequency range (>1 kHz). The change is significant, with the resonant frequency shifting downwards.

[0085] Winding radial deformation: low frequency range (<500 Hz) The changes are obvious, with the resonance peaks widening.

[0086] Loose iron core: all frequency bands The overall frequency increases, and the resonant frequency shifts slightly upward.

[0087] Insulation degradation: The phase angle decreases significantly in the mid-frequency range (100 Hz-1 kHz).

[0088] 3. Generate a diagnostic report, including THI value, probability of failure, and maintenance recommendations.

[0089] Step S5: Alarm and Data Management 1. Set three alarm thresholds: Warning: THI > 0.3 or any characteristic change exceeds twice the standard deviation of the healthy baseline.

[0090] Alarm: THI>0.5 or resonant frequency offset>5%.

[0091] Emergency alert: THI > 0.7 or a new resonance peak appears.

[0092] 2. All test data, feature spectra, and historical THI trends are stored in a database that supports remote network access.

[0093] Parameter determination and verification: Excitation frequency selection: 10-100 Hz: Sensitive to the condition of the iron core.

[0094] 100-1000 Hz: Sensitive to the overall mechanical condition of the winding.

[0095] 1-10 kHz: Sensitive to localized winding deformation and insulation condition.

[0096] Weight coefficient optimization: Based on historical fault data (including FRA and core inspection results) from 100 transformers of different capacities and models, logistic regression analysis was used to determine the method that achieves the highest fault classification accuracy. combination.

[0097] On-site verification results: A prototype system was deployed on 15 transformers ranging from 110 kV to 500 kV in 5 substations, with a monitoring period of 12-24 months.

[0098] Successful alert: Early winding deformation of three transformers (THI continued to rise to 0.4-0.5), confirmed by the power outage FRA.

[0099] The core clamps of one transformer were loose (THI suddenly increased to 0.6), which was discovered during a core inspection.

[0100] No missed reports, false alarm rate <2% (mainly caused by extreme temperature changes).

[0101] For example, such as Figure 2As shown, this embodiment also provides an online monitoring system for the internal condition of a transformer, including: a data acquisition module, used to inject an AC micro-current excitation signal into the low-voltage side of the transformer, and simultaneously acquire voltage response signals, vibration response signals on the surface of the transformer tank, and background power frequency current signals; The frequency domain processing module is used to process the acquired voltage response signal, vibration response signal, background power frequency current signal and injected AC micro-current excitation signal in the frequency domain, and calculate the electromagnetic impedance spectrum and vibration transfer function spectrum of the transformer respectively. The first calculation module is used to calculate the normalized electromagnetic-mechanical coupling coefficient across the entire frequency band based on the preset electromagnetic impedance reference spectrum and vibration transfer function reference spectrum under the initial healthy state of the transformer, combined with the electromagnetic impedance spectrum and vibration transfer function spectrum. The second calculation module is used to calculate the transformer's comprehensive health index based on the electromagnetic impedance spectrum, the normalized electromagnetic-mechanical coupling coefficient, and the transformer winding mechanical resonance frequency identified from the initial spectrum; and to monitor the internal condition of the transformer based on the comprehensive health index.

[0102] The present invention also provides an online monitoring device for the internal condition of a transformer, comprising: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of the online monitoring method for the internal condition of the transformer.

[0103] The present invention also provides a computer program product, including a computer program / instructions, which, when executed by a processor, implement the steps of the online monitoring method for the internal state of the transformer.

[0104] When the processor executes the computer program, it implements the above-mentioned steps for online monitoring of the internal state of the transformer, such as: injecting an AC micro-current excitation signal into the low-voltage side of the transformer, and simultaneously acquiring voltage response signals, vibration response signals on the surface of the transformer tank, and background power frequency current signals. The collected voltage response signal, vibration response signal, background power frequency current signal, and injected AC micro-current excitation signal were processed in the frequency domain to calculate the electromagnetic impedance spectrum and vibration transfer function spectrum of the transformer. Based on the preset electromagnetic impedance reference spectrum and vibration transfer function reference spectrum under the initial healthy state of the transformer, and combined with the electromagnetic impedance spectrum and vibration transfer function spectrum, the normalized electromagnetic-mechanical coupling coefficient of the entire frequency band is calculated. Based on the electromagnetic impedance spectrum, the normalized electromagnetic-mechanical coupling coefficient, and the mechanical resonance frequency of the transformer windings identified from the initial spectrum, the comprehensive health index of the transformer is calculated; the internal condition of the transformer is monitored based on the comprehensive health index.

[0105] For example, the computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing preset functions, wherein the instruction segments describe the execution process of the computer program in the online monitoring device for the internal condition of the transformer. For example, the computer program can be divided into a data acquisition module, a frequency domain processing module, a first calculation module, and a second calculation module; the specific functions are as follows: The data acquisition module is used to inject an AC micro-current excitation signal into the low-voltage side of the transformer, and simultaneously acquire the voltage response signal, the vibration response signal on the surface of the transformer tank, and the background power frequency current signal; The frequency domain processing module is used to perform frequency domain processing on the acquired voltage response signal, vibration response signal, background power frequency current signal, and injected AC micro-current excitation signal, and calculate the electromagnetic impedance spectrum and vibration transfer function spectrum of the transformer respectively; The first calculation module is used to calculate the normalized electromagnetic-mechanical coupling coefficient across the entire frequency band based on the preset electromagnetic impedance reference spectrum and vibration transfer function reference spectrum under the initial health state of the transformer, combined with the electromagnetic impedance spectrum and vibration transfer function spectrum; The second calculation module is used to calculate the comprehensive health index of the transformer based on the electromagnetic impedance spectrum, the normalized electromagnetic-mechanical coupling coefficient, and the mechanical resonance frequency of the transformer winding identified from the initial spectrum; and the internal state of the transformer is monitored based on the comprehensive health index.

[0106] The online monitoring device for the internal condition of the transformer can be a desktop computer, laptop, handheld computer, or cloud server, etc. The online monitoring device for the internal condition of the transformer may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above are examples of online monitoring devices for the internal condition of transformers and do not constitute a limitation on such devices. The device may include more components than described above, or combine certain components, or use different components. For example, the online monitoring device for the internal condition of the transformer may also include input / output devices, network access devices, buses, etc.

[0107] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or any conventional processor. The processor is the control center for the online monitoring of the transformer's internal condition, connecting various parts of the online monitoring equipment for the entire transformer's internal condition via various interfaces and lines.

[0108] The memory can be used to store the computer program and / or modules. The processor realizes various functions of the online monitoring device for the internal condition of the transformer by running or executing the computer program and / or modules stored in the memory and calling the data stored in the memory.

[0109] The memory may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a function (such as sound playback, image playback, etc.). The data storage area may store data created based on the use of the mobile phone (such as audio data, phonebook, etc.). Furthermore, the memory may include high-speed random access memory and non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart media cards (SMC), secure digital cards (SD cards), flash cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0110] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the online monitoring method for the internal state of a transformer.

[0111] If the modules / units integrated in the online monitoring system for the internal condition of the transformer are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.

[0112] Based on this understanding, the present invention can implement all or part of the processes in the above-described online monitoring method for the internal state of a transformer, or it can be accomplished by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the above-described online monitoring method for the internal state of a transformer. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or a preset intermediate form, etc.

[0113] The computer-readable storage medium may include: any entity or device capable of carrying the computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0114] It should be noted that the content contained in the computer-readable storage medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable storage medium does not include electrical carrier signals and telecommunication signals.

[0115] Compared with existing monitoring methods, this invention provides an online monitoring method and related device for the internal condition of transformers, which has the following advantages: First, early warning: It can detect early mechanical faults with winding deformation of less than 1%, weeks to months earlier than traditional methods.

[0116] Second, it covers multiple fault types: it monitors mechanical and electrical conditions simultaneously and can distinguish between different types of faults such as winding deformation, core loosening, and insulation deterioration.

[0117] Third, online and non-destructive: The excitation current is small, which does not affect the normal operation of the transformer and enables continuous online monitoring.

[0118] Fourth, quantitative assessment: The transformer comprehensive health index provides intuitive quantitative status indicators, supporting trend analysis and threshold alarms.

[0119] Fifth, strong anti-interference capability: Through background interference elimination and normalization processing, the influence of changes in operating conditions such as temperature and load is effectively suppressed.

[0120] The above embodiments are merely one of the implementation methods for achieving the technical solution of the present invention. The scope of protection claimed by the present invention is not limited to this embodiment, but also includes any variations, substitutions and other implementation methods that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention.

[0121] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for online monitoring of the internal condition of a transformer, characterized in that, include: An AC micro-current excitation signal is injected into the low-voltage side of the transformer, and the voltage response signal, the vibration response signal of the transformer tank surface, and the background power frequency current signal are collected simultaneously. The collected voltage response signal, vibration response signal, background power frequency current signal, and injected AC micro-current excitation signal were processed in the frequency domain to calculate the electromagnetic impedance spectrum and vibration transfer function spectrum of the transformer. Based on the preset electromagnetic impedance reference spectrum and vibration transfer function reference spectrum under the initial healthy state of the transformer, and combined with the electromagnetic impedance spectrum and vibration transfer function spectrum, the normalized electromagnetic-mechanical coupling coefficient of the entire frequency band is calculated. Based on the electromagnetic impedance spectrum, the normalized electromagnetic-mechanical coupling coefficient, and the mechanical resonance frequency of the transformer windings identified from the initial spectrum, the comprehensive health index of the transformer is calculated; the internal condition of the transformer is monitored based on the comprehensive health index.

2. The method for online monitoring of the internal condition of a transformer according to claim 1, characterized in that, The frequency domain processing of the acquired voltage response signal, vibration response signal, background power frequency current signal, and injected AC micro-current excitation signal, respectively, yields the electromagnetic impedance spectrum and vibration transfer function spectrum of the transformer, including: Based on the voltage response signal, background power frequency current signal and injected AC micro-current excitation signal, the excitation response is separated by frequency domain subspace decomposition method, and the electromagnetic impedance spectrum of the transformer is calculated. Based on the vibration response signal and the injected AC microcurrent excitation signal, the vibration transfer function spectrum, which characterizes the coupling efficiency from electromagnetic force to mechanical vibration, is calculated.

3. The online monitoring method for the internal condition of a transformer according to claim 2, characterized in that, The electromagnetic impedance spectrum of the transformer is calculated by separating the excitation response using the voltage response signal, background power frequency current signal, and injected AC micro-current excitation signal, and the frequency domain subspace decomposition method is employed. The specific formula is as follows: In the formula, For frequency Electromagnetic impedance at the location; Indicates the signal at frequency Fourier transform coefficients at the location; It is a voltage response signal; Background power frequency current signal; Pass function for background; To inject an excitation current signal.

4. The online monitoring method for the internal condition of a transformer according to claim 2, characterized in that, Based on the vibration response signal and the injected AC micro-current excitation signal, the vibration transfer function spectrum, which characterizes the coupling efficiency from electromagnetic force to mechanical vibration, is calculated. The specific formula is as follows: In the formula, For frequency Vibration transfer function at the location; Indicates the signal at frequency Fourier transform coefficients at the location; It is a vibration response signal; To inject an excitation current signal; The sensitivity of the accelerometer; Install coupling coefficients for the sensor.

5. The online monitoring method for the internal condition of a transformer according to claim 1, characterized in that, The normalized electromagnetic-mechanical coupling coefficient across the entire frequency band is calculated by combining the electromagnetic impedance spectrum and vibration transfer function spectrum based on the preset electromagnetic impedance spectrum and vibration transfer function spectrum under the initial healthy state of the transformer, including: Calculate the frequency separately Calculate the ratio between the amplitude ratios of the vibration transfer function spectrum and the vibration transfer function reference spectrum, and the amplitude ratio of the electromagnetic impedance spectrum and the electromagnetic impedance reference spectrum. Multiply the ratio between the two amplitude ratios by the ratio of the current frequency to the reference frequency, and by the Gaussian window function value based on the mechanical resonance frequency of the transformer winding, to obtain the normalized electromagnetic-mechanical coupling coefficient at the frequency. The specific formula is as follows: In the formula, For frequency The normalized electromagnetic-mechanical coupling coefficient at the location; This is the vibration transfer function of the transformer measured under reference conditions; This is the electromagnetic impedance of the transformer measured under reference conditions; For frequency Electromagnetic impedance at the location; For frequency Vibration transfer function at the location; The current frequency point; For reference frequency; This is the mechanical resonant frequency of the transformer winding; The resonant bandwidth; This represents the Gaussian window function.

6. The method for online monitoring of the internal condition of a transformer according to claim 1, characterized in that, The comprehensive health index of the transformer is calculated based on the electromagnetic impedance spectrum, the normalized electromagnetic-mechanical coupling coefficient, and the mechanical resonant frequency of the transformer winding identified from the initial spectrum. This includes: A comprehensive health index for transformers, reflecting their internal condition, is constructed using the following formula: In the formula, This indicates the overall health index of the transformer; This represents the normalized change in electrical state; This represents the normalized change in the coupling coefficient; This represents the normalized offset of the resonant frequency; , , These are the corresponding weight coefficients; The normalized change in electrical state is calculated based on the electromagnetic impedance spectrum, and the specific formula is as follows: In the formula, Number of frequency points; Frequency point index; For frequency The magnitude of the electromagnetic impedance at that point; This represents the amplitude of the electromagnetic impedance measured by the transformer under reference conditions. This represents the standard deviation of the electromagnetic impedance amplitude measured multiple times under reference conditions. The normalized change in the coupling coefficient is calculated based on the normalized electromagnetic-mechanical coupling coefficient, and the specific formula is as follows: In the formula, is the number of major mechanical resonance frequency points; m is the index of the major mechanical resonance frequency points; For the first The center frequency of each resonance peak; The coupling coefficient at the resonant frequency in the current state; The coupling coefficient is the reference state. The normalized offset of the resonance frequency is calculated based on the mechanical resonance frequency of the transformer winding, and the specific formula is as follows: In the formula, Let be the mechanical resonant frequency of the transformer winding, representing the th The center frequency of each resonance peak in a healthy state; This represents the full width at half maximum (FWHM) of the corresponding resonance peak under the baseline condition.

7. The method for online monitoring of the internal condition of a transformer according to claim 1, characterized in that, The monitoring of the transformer's internal condition based on a comprehensive health index includes: The current internal condition of the transformer is determined based on a comprehensive health index, as follows: If the overall health index is 0, then the current internal state of the transformer is judged to be healthy. If the overall health index is within the first preset range, the current internal state of the transformer is judged to be slightly abnormal. If the overall health index is within the second preset range, the current internal condition of the transformer is judged to be moderately abnormal. If the overall health index is in the third preset range, the current internal state of the transformer is judged to be seriously abnormal.

8. An online monitoring system for the internal condition of a transformer, characterized in that, include: The data acquisition module is used to inject AC micro-current excitation signals into the low-voltage side of the transformer and simultaneously acquire voltage response signals, vibration response signals on the surface of the transformer tank, and background power frequency current signals. The frequency domain processing module is used to process the acquired voltage response signal, vibration response signal, background power frequency current signal and injected AC micro-current excitation signal in the frequency domain, and calculate the electromagnetic impedance spectrum and vibration transfer function spectrum of the transformer respectively. The first calculation module is used to calculate the normalized electromagnetic-mechanical coupling coefficient across the entire frequency band based on the preset electromagnetic impedance reference spectrum and vibration transfer function reference spectrum under the initial healthy state of the transformer, combined with the electromagnetic impedance spectrum and vibration transfer function spectrum. The second calculation module is used to calculate the transformer's comprehensive health index based on the electromagnetic impedance spectrum, the normalized electromagnetic-mechanical coupling coefficient, and the transformer winding mechanical resonance frequency identified from the initial spectrum; and to monitor the internal condition of the transformer based on the comprehensive health index.

9. An online monitoring device for the internal condition of a transformer, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the online monitoring method for the internal state of a transformer according to any one of claims 1-7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it is used to implement the online monitoring method for the internal state of the transformer according to any one of claims 1-7.

Images