Dry-type power transformer short circuit withstand capacity dynamic monitoring and evaluation system

Abstract

This invention relates to the field of electrical variable measurement technology, specifically disclosing a dynamic monitoring and evaluation system for the short-circuit withstand capability of dry-type power transformers. The system includes a high-speed data acquisition module, a short-circuit characteristic analysis module, an electrodynamic calculation module, a mechanical stress evaluation module, a dynamic evaluation decision module, and a result output and early warning module. By acquiring current and voltage signals in real time, extracting short-circuit characteristic parameters, calculating electrodynamic and mechanical stresses, and combining multi-level thresholds, historical data, and a lifespan decay model for comprehensive evaluation and early warning, it achieves accurate and dynamic monitoring of the transformer's short-circuit withstand capability and quantitative assessment of its health status, thereby improving the operational safety of power grid equipment.

Description

Technical Field

[0001] This invention belongs to the field of electrical variable measurement technology, specifically relating to a dynamic monitoring and evaluation system for the short-circuit withstand capacity of dry-type power transformers. Background Technology

[0002] In the field of power transmission and transformation, dry-type power transformers are key equipment, and their safe and stable operation is crucial to ensuring the reliability of the power grid. Among them, short-circuit withstand capability is a core indicator for measuring the mechanical strength and electrical performance of transformers, and it directly relates to the equipment's survivability and lifespan under sudden short-circuit faults.

[0003] Monitoring and assessing the short-circuit withstand capability of transformers is a crucial technical direction for ensuring their safe operation. This technology aims to dynamically evaluate the electrodynamic and mechanical stress state of transformer windings by real-time acquisition and analysis of key electrical parameters during short-circuit events, providing a basis for operation and maintenance decisions.

[0004] Existing technologies typically rely on relay protection devices for short-circuit fault detection. However, the function of such devices is limited to determining whether a short circuit has occurred, and they cannot accurately capture key waveform details such as the instantaneous peak value, asymmetry, and duration of the short-circuit current.

[0005] These missing details are precisely the core input parameters for calculating electrodynamics and assessing mechanical stress. Furthermore, traditional assessment methods rely on manual, post-hoc calculations, a complex and severely delayed process that cannot reflect the transformer's instantaneous withstand state under each short-circuit impact in real time, making it difficult to provide timely warnings of potential cumulative damage or sudden failure risks. Therefore, a technical solution capable of dynamically monitoring and accurately assessing short-circuit withstand capacity is urgently needed. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to overcome the technical defects of the existing technology in the monitoring and evaluation of the short-circuit withstand capacity of dry-type power transformers, such as incomplete detection parameters, lagging evaluation process and inability to reflect instantaneous mechanical stress state in real time, and to provide a dynamic monitoring and evaluation system for the short-circuit withstand capacity of dry-type power transformers.

[0007] To achieve the above objectives, the present invention adopts the following technical solution. The system includes a high-speed data acquisition module, a short-circuit characteristic analysis module, an electrodynamic calculation module, a mechanical stress assessment module, a dynamic assessment and decision-making module, and a result output and early warning module. The high-speed data acquisition module is used to synchronously acquire the instantaneous values ​​of the three-phase current and three-phase voltage on the high-voltage and low-voltage sides of the transformer at a sampling frequency higher than 100 kHz.

[0008] The short-circuit feature analysis module is connected to the output of the high-speed data acquisition module to perform real-time analysis of the acquired current and voltage signals, accurately extracting three core feature parameters: the peak amplitude of the short-circuit current, the asymmetry coefficient, and the short-circuit duration. The electrodynamic calculation module receives the feature parameters from the short-circuit feature analysis module and constructs a three-dimensional electrodynamic distribution model of the transformer windings under short-circuit conditions based on the principles of electromagnetic force physics. It then calculates the dynamic electrodynamic values ​​of each phase winding in both radial and axial dimensions.

[0009] The mechanical stress assessment module is connected to the electrodynamic calculation module. Based on the theory of mechanics of materials, it establishes the stress-strain constitutive relationship of the winding conductor. Combining the winding structural parameters and insulation material properties, it converts the electrodynamic values ​​into the real-time mechanical stress distribution and deformation displacement of the winding. The dynamic assessment decision module integrates preset multi-level assessment thresholds and a historical data comparison unit. This module receives the real-time stress data output by the mechanical stress assessment module and performs a comprehensive comparison and analysis with the transformer's factory withstand standards, historical short-circuit event records, and a lifespan degradation model based on the number of years of operation. It generates a quantitative assessment level of the current short-circuit event's impact on the transformer's health status. The result output and early warning module is connected to the dynamic assessment decision module. It converts the assessment level into specific human-computer interaction information and automatically triggers multi-level early warning signals when the assessment level exceeds the preset safety threshold.

[0010] Furthermore, the high-speed data acquisition module includes a signal conditioning circuit and a synchronous sampling unit. The signal conditioning circuit uses isolation amplification and anti-aliasing filtering techniques to preprocess the raw current and voltage signals, ensuring that the signal amplitude matches the input range of the analog-to-digital converter and effectively suppressing high-frequency noise interference. The synchronous sampling unit uses the Global Positioning System clock signal as a synchronization reference to ensure that the acquisition times of the three-phase current and voltage signals are strictly aligned, with a time synchronization error of no more than 1 microsecond.

[0011] Furthermore, the short-circuit feature analysis module performs short-circuit feature extraction as follows: First, the module monitors the entire waveform of the input current signal. When the instantaneous value of any phase current exceeds twice its rated current value and the duration exceeds one power frequency cycle, a short-circuit event is detected, and this moment is recorded as time 0. Then, using time 0 as a reference, the module analyzes the current waveform data over two power frequency cycles, determining the peak amplitude of the short-circuit current using a peak-finding algorithm. The asymmetry coefficient is calculated by extracting the attenuation time constant of the DC component of the short-circuit current and the initial phase angle. The short-circuit duration is calculated from time 0 until the current value drops below 1.5 times the rated current and remains below this level for three power frequency cycles.

[0012] Furthermore, the three-dimensional electrodynamic distribution model used in the electrodynamic calculation module is constructed as follows: This model discretizes each phase winding of the transformer into 128 independent conductor units. For each conductor unit, the magnetic field force generated by the currents of all other conductor units under the current short-circuit current excitation is calculated according to Biosavart's law. The calculation fully considers the specific spatial geometry of the winding, the relative positional relationships between conductors, and the influence of the core magnetic circuit. Finally, the Lorentz force on each conductor unit is vector-synthesized to obtain the magnitude and point of application of the resultant force in the radial and axial directions for that phase winding.

[0013] Furthermore, the execution flow of the mechanical stress assessment module includes two stages: stress calculation and deformation analysis. In the stress calculation stage, the module uses the electrodynamic data provided by the electrodynamic calculation module, combined with the elastic modulus and yield strength of the winding conductors and the supporting stiffness parameters of the insulation material, to calculate the maximum equivalent stress value inside the winding using the finite element method. In the deformation analysis stage, based on the calculated stress distribution, the module solves the deformation control equation of the winding, outputs the deformation displacement at key locations, and ensures that this displacement does not exceed the maximum allowable deformation limit of the transformer design.

[0014] Furthermore, the dynamic evaluation and decision-making module includes three preset multi-level evaluation thresholds: an immediate warning threshold, a careful observation threshold, and a safe operation threshold. The immediate warning threshold corresponds to the winding mechanical stress reaching 80% of the material's yield strength or the deformation displacement reaching 90% of the design limit. The careful observation threshold corresponds to the stress reaching 50% of the yield strength or the deformation displacement reaching 60% of the design limit. The safe operation threshold corresponds to the stress being below 30% of the yield strength and the deformation displacement being below 40% of the design limit. The historical data comparison unit stores the evaluation results of previous short-circuit events of the transformer. The dynamic evaluation and decision-making module compares the current stress data with historical data to determine trends. If it finds that the stress increment caused by the current short circuit is 30% higher than the historical average, it automatically upgrades the evaluation level.

[0015] Furthermore, the dynamic assessment and decision-making module also integrates a lifespan degradation model. This model takes the transformer's cumulative operating years and the number of major short-circuit events as input, and dynamically corrects the transformer's remaining mechanical life using a pre-set insulation material aging equation and a mechanical fatigue cumulative damage model. The current assessment results will be compared with the withstand capability corrected by the lifespan degradation model, thus more accurately reflecting the transformer's actual health status.

[0016] Furthermore, the results output and early warning module supports multiple output interfaces and early warning methods. Output interfaces include an Ethernet communication interface, a 4G wireless communication module, and a local display unit, used to upload assessment levels, key parameters, and early warning information to the superior monitoring system or display them locally. Early warning methods are triggered based on the assessment level, including local audible and visual alarms, SMS push notifications to designated maintenance personnel, and sending emergency signals to the power grid dispatch system.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0018] This invention achieves real-time, accurate, and automated assessment of the short-circuit withstand capability of dry-type power transformers by constructing a complete technology chain from high-speed data acquisition to dynamic evaluation and decision-making. The system can fully capture the instantaneous detailed characteristics of the short-circuit current and calculate electrodynamic and mechanical stresses based on a rigorous physical model, significantly improving the accuracy and reliability of the assessment results.

[0019] The dynamic assessment and decision-making module combines multi-level thresholds, historical data, and lifespan models to make comprehensive judgments, so that the assessment conclusions not only reflect the severity of a single short-circuit impact, but also reflect the cumulative damage and health status trend of the transformer under long-term operation.

[0020] The results output and early warning module provides diverse information interaction and alarm methods, ensuring that risk assessment conclusions can be communicated to operation and maintenance personnel in a timely and effective manner. This provides solid data support for preventive maintenance and rapid fault handling, fundamentally enhancing the safe operation guarantee capability of key power grid equipment. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the overall technical architecture of the dynamic monitoring and evaluation system for the short-circuit withstand capability of dry-type power transformers proposed in this invention.

[0022] Figure 2 This is a schematic diagram of the core principle framework for short-circuit feature analysis and electrodynamic calculation in this invention;

[0023] Figure 3 This is a logical flowchart of the mechanical stress assessment and dynamic assessment decision-making process in this invention;

[0024] Figure 4 This is a schematic diagram of the multi-level interaction relationship and data flow of high-speed data acquisition and multi-module collaboration in this invention;

[0025] Figure 5 This is a schematic diagram comparing the core principle of this invention with existing technologies in terms of evaluation accuracy. Detailed Implementation

[0026] Example 1: Please refer to the appendix Figures 1 to 5This embodiment details the specific implementation of the dynamic monitoring and evaluation system for the short-circuit withstand capability of dry-type power transformers. The system aims to achieve real-time, dynamic monitoring and quantitative evaluation of the electrodynamic forces and mechanical stresses experienced by dry-type power transformers during short-circuit events through a series of precise and collaborative hardware and software modules, and to generate early warning signals accordingly.

[0027] The core components of the system include a high-speed data acquisition module, a short-circuit characteristic analysis module, an electrodynamic calculation module, a mechanical stress assessment module, a dynamic assessment and decision-making module, and a result output and early warning module. These modules are connected and interact in a strict logical order and according to the data flow, forming a closed-loop monitoring and assessment chain.

[0028] The high-speed data acquisition module is the source of data sensing for the system. This module is directly connected to the secondary output terminals of the current transformers and voltage transformers on the high-voltage and low-voltage sides of the dry-type power transformer. Its core function is to synchronously acquire the instantaneous three-phase current and voltage signals on both the high-voltage and low-voltage sides of the transformer at a sampling frequency higher than 100 kHz. To achieve high-precision measurement, the high-speed data acquisition module integrates a signal conditioning circuit and a synchronous sampling unit. The signal conditioning circuit employs isolation amplification technology based on isolation operational amplifiers to effectively isolate common-mode interference that may be introduced from the high-voltage side, protecting the safety of the back-end processing circuitry.

[0029] Meanwhile, the circuit incorporates an anti-aliasing filter with a cutoff frequency of 50 kHz, employing a Butterworth low-pass filter topology. Its stopband attenuation is higher than 40 dB per decibel, ensuring effective suppression of high-frequency noise interference above the Nyquist frequency at a 100 kHz sampling frequency, preventing signal aliasing. The signal conditioning circuit also features a programmable gain amplifier with an adjustable gain range from 1 to 1000 times. Software configuration ensures that the amplitude of the raw signal from the transformer is precisely adjusted to the optimal input voltage range of the analog-to-digital converter, typically ±5V or ±10V, thereby fully utilizing the dynamic range of the analog-to-digital converter and improving the signal-to-noise ratio of the signal measurement. The synchronous sampling unit is crucial for achieving synchronous acquisition of multi-channel data.

[0030] This unit uses a high-precision Global Positioning System (GPS) receiver as an external clock reference, with the receiver's output pulse-per-second signal and Coordinated Universal Time (UTC) signal serving as synchronization sources. The synchronization sampling unit internally contains a highly stable temperature-compensated crystal oscillator with a frequency stability better than ±0.1 ppm. Upon receiving the GPS synchronization signal, the phase-locked loop (PLL) circuit within the synchronization sampling unit locks the local sampling clock with the GPS clock, ensuring strict alignment of the start-up times of the analog-to-digital converters (ADCs) for all channels.

[0031] After calibration, the time synchronization error of the synchronous sampling unit is no greater than 1 microsecond, which is crucial for subsequent analysis of the phase relationship of three-phase current and voltage, as well as calculation of instantaneous power and electromagnetic force. The acquired raw instantaneous current and voltage values ​​are encapsulated into data frames. Each data frame contains a timestamp, channel identifier, sampled value array, and cyclic redundancy check code, and is transmitted in real time to the short-circuit feature analysis module through a high-speed serial peripheral device interface or a gigabit Ethernet interface.

[0032] The short-circuit characteristic analysis module receives real-time current and voltage data streams from the high-speed data acquisition module. The core task of this module is to accurately identify the onset of a short-circuit event and extract three key characteristic parameters from the current waveform: the peak amplitude of the short-circuit current, the asymmetry coefficient, and the short-circuit duration. Please refer to the appendix. Figure 2 The short-circuit feature analysis module internally runs a complex real-time signal processing algorithm. The module first performs continuous, point-by-point full-waveform monitoring of the six input current signals. It continuously calculates the root mean square value of the current within each power frequency cycle and compares it with the preset transformer rated current value.

[0033] When the instantaneous value of any phase current exceeds twice its rated current value, and this overcurrent condition lasts for more than one power frequency cycle (20 milliseconds), the module immediately determines that the first minor short-circuit event has started and accurately records this determination moment as the time 0 point of this short-circuit event. The recording accuracy of the time 0 point is synchronized with the timestamp of the high-speed data acquisition module, reaching the microsecond level. After confirming the start of the short-circuit event, the short-circuit feature analysis module uses the time 0 point as a reference and extracts complete current waveform data within at least two power frequency cycles for in-depth analysis.

[0034] For extracting the peak amplitude of the short-circuit current, the module employs an improved digital peak-finding algorithm. This algorithm first performs a 5.3-fold smoothing filter on the current data to suppress random noise. Then, it calculates the first-order difference sequence of the waveform and locates all possible extreme points by detecting zero-crossing points where the difference value changes from positive to negative. Next, the algorithm compares the amplitudes of these extreme points and selects the maximum positive extreme point and the minimum negative extreme point within the first power frequency cycle after time zero. The larger of their absolute values ​​is taken as the peak amplitude of the short-circuit current for that phase.

[0035] The calculation of the asymmetry coefficient is more complex; it quantifies the DC component content in the short-circuit current, which directly affects the magnitude of the electrodynamic force. The module first uses the Prony algorithm or a similar algorithm to fit the current waveform during the initial stage of the short circuit to extract the initial amplitude of its DC component and the decay time constant. The asymmetry coefficient is calculated using the core formula:

[0036] ;

[0037] in, It represents a specific moment starting from 0:00, usually taken as the peak moment of the first power frequency cycle after the short circuit occurs. The decay time constant representing the DC component is related to the equivalent resistance and inductance parameters of the transformer and is obtained through fitting. The calculated asymmetry coefficient reflects the degree of deviation of the short-circuit current waveform relative to a symmetrical sine wave, and its value is usually greater than 1.

[0038] To determine the short-circuit duration, the module continuously monitors the current waveform from time 0. The short-circuit event ends only when the root mean square value of the current in all phases falls below 1.5 times their rated current value, and this low-current state remains stable for at least three power frequency cycles (60 milliseconds). The short-circuit duration is the time interval from time 0 to the end of the event. These three extracted feature parameters, along with the corresponding phase information and timestamp, are encapsulated into a data structure and sent to the electrodynamic calculation module in real time.

[0039] The core responsibility of the electrodynamic calculation module is to convert the characteristic parameters provided by the short-circuit characteristic analysis module into specific electrodynamic forces acting on the transformer windings, based on electromagnetic principles. Please refer to the appendix. Figure 2 This module internally constructs a refined three-dimensional electrodynamic distribution model. This model first performs geometric discretization on each phase winding of the transformer. Specifically, each phase winding is uniformly divided into 32 line segments along the axial direction and into 4 sector regions along the circumference, resulting in a total of 128 independent conductor units for each phase winding.

[0040] Each conductor element is assigned attributes such as spatial coordinates, length, cross-sectional area, and current direction. For any conductor element, the electrodynamic calculation module needs to calculate the magnetic field force exerted on it by the currents of all other conductor elements under the current short-circuit current excitation. The calculation is based on the classical Biosavart law. The module first calculates the magnetic flux density vector at the location of the target conductor element based on the instantaneous value of the short-circuit current, the asymmetry coefficient, and the relative spatial positions of each conductor element.

[0041] This calculation process considers the specific spatial geometry of the windings, including the average diameter, height, and turn spacing, as well as the influence of the core window size and the permeability of the core material on the magnetic field distribution. The influence of the core is approximated by introducing the current image method or the finite element method to more accurately reflect the boundary conditions of the magnetic circuit. After obtaining the magnetic induction intensity at the target conductor unit, the module calculates the force on that conductor unit according to the Lorentz force formula.

[0042] This force is a vector, and its direction is perpendicular to the plane formed by the direction of the current and the direction of the magnetic field. After performing the above calculations on each conductor unit, the module performs vector synthesis of the Lorentz forces on all conductor units. The synthesis process consists of two steps: First, the radial component forces of all conductor units in the same phase winding are synthesized to obtain the total radial electromotive force of the phase winding, and the position of the resultant force application point along the winding height is determined.

[0043] Radial electrodynamic forces primarily cause the windings to experience outward expansion or inward contraction. Secondly, the axial force components of all conductor units within the same phase winding are combined to obtain the total axial electrodynamic force for that phase winding, and the point of application of the resultant force is determined. Axial electrodynamic forces primarily cause the windings to experience compressive or tensile stress. Finally, the electrodynamic calculation module outputs the dynamic electrodynamic force values ​​for each phase winding in both radial and axial directions, including the magnitude, direction, and coordinates of the point of application of the force. These data form the basis for subsequent mechanical stress assessments.

[0044] The mechanical stress assessment module receives electrodynamic data from the electrodynamic calculation module. Its task is to convert these electromagnetic forces into actual mechanical stress and deformation within the winding conductors and insulation structure. Please refer to the appendix. Figure 3 The execution process of this module is strictly divided into two stages: stress calculation and deformation analysis.

[0045] During the stress calculation phase, the module first loads a pre-stored database of transformer structural parameters. This database contains material properties of the winding conductors, such as elastic modulus, Poisson's ratio, density, and yield strength; properties of the insulation material, such as the elastic modulus, compressive strength, and stiffness coefficient of the laminated insulation board; and geometric parameters of the winding, such as conductor cross-sectional dimensions, number of turns, number and distribution of support bars, and clamping force. Based on these parameters and combined with electrodynamic data, the module constructs a simplified finite element model of the winding. This model treats the winding as a structural system composed of multiple beam elements and spring elements, where beam elements represent conductors and spring elements represent insulating pads and support bars.

[0046] The module applies the calculated radial and axial electrodynamic forces as distributed or concentrated loads to the corresponding nodes of the finite element model. Subsequently, the module calls the solver to calculate the stress distribution of the model under the load. The calculation employs linear statics assumptions; for dynamic processes, transient dynamics analysis is used, discretizing the short-circuit duration into multiple time steps and calculating the stress response at each time step. The key stress index output by the module is the maximum equivalent stress value inside the winding, which typically occurs at conductor bends, strut contact points, or winding ends.

[0047] During the deformation analysis phase, the module further solves the deformation control equations of the winding based on the nodal forces and displacements obtained in the stress calculation phase. It outputs the deformation displacement at key locations, particularly the radial expansion, axial compression, and potential overall tilt displacement of the winding. The module internally presets the maximum allowable deformation limits for the transformer design, such as radial deformation not exceeding 2% of the winding inner diameter and axial deformation not exceeding 0.5% of the winding height. The final output data from the mechanical stress assessment module includes: maximum equivalent stress, stress distribution at key points, maximum deformation displacement, deformation distribution contour map, and the percentage of stress and deformation relative to their allowable limits. This data is transmitted in real-time to the dynamic evaluation decision module.

[0048] The dynamic evaluation and decision-making module is the intelligent central hub of the entire system, responsible for comprehensively analyzing and determining the level of real-time data sent by the mechanical stress evaluation module. Please refer to the appendix. Figure 3 This module integrates three core units: a multi-level evaluation threshold unit, a historical data comparison unit, and a lifespan degradation model unit. The multi-level evaluation threshold unit presets three distinct evaluation level thresholds: an immediate warning threshold, a careful observation threshold, and a safe operation threshold. The immediate warning threshold corresponds to the winding mechanical stress reaching 80% of the conductor material's yield strength, or the deformation displacement reaching 90% of the design's maximum allowable deformation limit.

[0049] Reaching this threshold indicates that the transformer is facing a serious risk of potentially permanent damage. Note that this threshold corresponds to stress reaching 50% of the yield strength or deformation reaching 60% of the design limit. This level indicates that the transformer is experiencing significant short-circuit stress and requires enhanced monitoring during subsequent operation. The safe operating threshold corresponds to stress below 30% of the yield strength and deformation below 40% of the design limit. At this level, the impact of a short-circuit event on the transformer is considered to be within safe limits.

[0050] The historical data comparison unit maintains a structured historical database, recording detailed information about each short-circuit event of the transformer, including the peak stress, maximum deformation, and final assessment level for each event. Upon receiving real-time stress data for the current short-circuit event, the dynamic assessment decision module immediately retrieves data from the historical database for the 10 most recent short-circuit events and calculates their average stress level. The module then compares the current stress data with this historical average.

[0051] If the stress increment caused by this short circuit is found to be 30% higher than the historical average, the module will automatically upgrade the initial assessment level, for example, from "observation" to "immediate warning," to reflect the unusual severity of this short circuit event. The lifespan degradation model unit is another important correction factor. This model uses the transformer's cumulative operating years and the number of major short circuit events it has experienced in the past as its main input parameters.

[0052] The model incorporates an aging equation for the insulation material, which considers both thermal and electrical aging effects, describing the decay of the insulation material's mechanical strength over operating time and under overload conditions. Simultaneously, the model employs a mechanical fatigue cumulative damage model based on Miner's linear cumulative damage theory to quantify the fatigue damage caused by multiple short-circuit impacts on the winding conductors. The life decay model integrates these factors and outputs a life decay factor, which is used to dynamically adjust the transformer's current withstand capability. For example, the actual stress threshold of a transformer with a long operating history and multiple short circuits may be lowered from the factory standard.

[0053] The dynamic assessment and decision-making module compares the real-time stress data output by the mechanical stress assessment module with the tolerance threshold corrected by the life decay model, thereby generating a quantitative assessment level that better reflects the actual health status of the transformer. This level is a comprehensive indicator; for example, a number 1 represents safe operation, 2 represents careful observation, and 3 represents immediate warning.

[0054] The results output and early warning module serves as the interface for interaction between the system and the external environment and maintenance personnel. This module receives the final assessment level and related detailed parameters from the dynamic evaluation and decision-making module. Please refer to the appendix. Figure 4 This module features diverse output interfaces. The wired communication interface adopts a standard industrial Ethernet interface, supporting communication protocols such as Modbus TCP or IEC 61850. It is used to package evaluation results, real-time stress data, deformation data, historical comparison conclusions, and other information into standard data frames and upload them to the substation's monitoring system or a remote power grid control center.

[0055] The wireless communication interface integrates a 4G or 5G wireless communication module, serving as a backup channel when Ethernet is unavailable, or for directly sending warning information to mobile maintenance terminals. The local display unit is typically a high-brightness LCD screen, installed on a control cabinet near the transformer, cyclically displaying the current assessment level, key parameters, and system status. Regarding the warning method, the result output and the warning module automatically trigger corresponding warning actions based on the received assessment level. When the assessment level is "Observe Carefully," the module may only trigger a color change in the local display unit, such as turning yellow, and displaying a prompt message on the screen.

[0056] When the assessment level is immediate warning, the module will simultaneously activate multiple warning levels: First, it will trigger a high-decibel buzzer and a red rotating warning light installed on the control cabinet for local audible and visual alarm. Second, it will send SMS alarms containing the specific transformer number, assessment level, exceeded parameters, and time of occurrence to the mobile phones of at least three pre-set maintenance personnel via the 4G wireless communication module. Finally, it will send an emergency signal conforming to the power system communication protocol to the power grid dispatching system via the Ethernet interface, notifying dispatchers that the transformer may face a failure risk and that adjustments to the operating mode should be considered. All output information and warning records are accurately timestamped and stored in the module's local non-volatile memory for subsequent querying and analysis.

[0057] Please refer to the attached document. Figure 5 This system, through its sophisticated and collaborative modular design, significantly improves the accuracy of short-circuit withstand capability assessment. Modules communicate via a high-speed data bus, ensuring real-time and consistent data flow. Upon power-up, each module performs a self-test, the high-speed data acquisition module begins continuous data acquisition, and the remaining modules remain in standby mode. Once the short-circuit feature analysis module detects a short-circuit event, the entire assessment chain is activated, completing the entire process from data acquisition to early warning output within hundreds of milliseconds, providing crucial real-time decision support for the safe operation of the power grid.

[0058] Example 2: This example provides another specific implementation of the dynamic monitoring and evaluation system for the short-circuit withstand capability of dry-type power transformers. Based on Example 1, it focuses on the alternative algorithm for calculating the asymmetry coefficient in the short-circuit characteristic analysis module, and the deepening of the historical data comparison strategy in the dynamic evaluation decision module.

[0059] In the short-circuit feature analysis module, in addition to the method for calculating the asymmetry coefficient based on formulas described in Example 1, this example introduces a pure numerical calculation method based on waveform decomposition and integration. This method does not rely on explicit fitting of the decay time constant. After determining the short-circuit event initiation and identifying time 0, the module first extracts the first complete power frequency cycle, i.e., 20 milliseconds of current sampling data, starting from time 0. Then, the module uses the Fast Fourier Transform algorithm to accurately calculate the amplitude and phase of the fundamental component of the current signal within this cycle. Finally, the module subtracts this calculated fundamental sinusoidal component from the original current sampling value to obtain the residual signal.

[0060] The residual signal mainly contains a DC component and higher harmonics. To extract the DC component, the module performs a moving average filter on the residual signal over the power frequency cycle. The output value of the moving average filter is an approximation of the DC component at that moment. The asymmetry coefficient is indirectly reflected by calculating the ratio of the peak value of the full-wave short-circuit current to the peak value of the fundamental current. Specifically, the module finds the absolute maximum value of the original current waveform within the power frequency cycle, denoted as the peak value of the full-wave.

[0061] Meanwhile, the peak value of the fundamental component is the fundamental amplitude obtained through Fast Fourier Transform. The asymmetry coefficient can be approximated as the total peak value divided by the fundamental peak value. This method has relatively controllable computational complexity and is robust to noise, providing another reliable feature extraction approach for the system.

[0062] In the historical data comparison unit of the dynamic evaluation decision module, this embodiment further enhances its analytical dimensions. In addition to the simple comparison of stress increments as described in Embodiment 1, this system also introduces fatigue damage accumulation comparison based on stress spectrum analysis. The historical data comparison unit not only stores the peak stress of each short-circuit event but also records the stress time history of each event, i.e., the curve of stress change over time. The unit integrates a rainflow counting algorithm to simplify the complex stress time history into a series of complete stress cycles. Each stress cycle includes a stress range and a mean.

[0063] The unit pre-sets corresponding stress-life curves for the transformer winding materials. For each historical short-circuit event, the unit calculates the resulting fatigue damage. When a new short-circuit event occurs, the module similarly calculates the fatigue damage for this event. The historical data comparison unit compares the fatigue damage of this event with the historical average fatigue damage of a single event. If the damage of this event exceeds 50% of the historical average, the dynamic assessment decision module considers that this short circuit significantly depletes the transformer's cumulative lifespan, even if the peak stress does not reach the immediate warning threshold. It may raise the assessment level to the observation level or trigger a dedicated cumulative damage warning signal. This deep comparison based on fatigue damage can more sensitively capture short-circuit events with low peak values ​​but frequent fluctuations that have a significant impact on material fatigue, thus providing a more comprehensive health status assessment.

[0064] Furthermore, in the results output and early warning module, this embodiment adds a function to generate a visual in-depth analysis report of the evaluation results. In addition to the basic level and parameter output, this module also has an embedded report generation engine. After a short-circuit event evaluation is completed, the engine automatically generates a comprehensive report that includes waveform diagrams, stress distribution cloud maps, deformation diagrams, historical trend comparison charts, and life decay impact analysis.

[0065] The report is saved in portable document format or Hypertext Markup Language format and can be automatically sent to a designated operation and maintenance management platform or email address via Ethernet or 4G wireless network. It provides operation and maintenance personnel with more intuitive and in-depth equipment status analysis data to assist them in making maintenance decisions and formulating condition-based maintenance plans.

Claims

1. A dynamic monitoring and evaluation system for short circuit withstand capability of dry-type power transformers, characterized in that, include: The high-speed data acquisition module is used to synchronously acquire the instantaneous values ​​of three-phase current and three-phase voltage on the high-voltage and low-voltage sides of the transformer at a sampling frequency higher than 100 kHz. The short-circuit feature analysis module is connected to the output of the high-speed data acquisition module. It is used to analyze the acquired current and voltage signals in real time and extract three core feature parameters: peak amplitude of short-circuit current, asymmetry coefficient, and short-circuit duration. The electrodynamic calculation module receives feature parameters from the short-circuit feature analysis module, constructs a three-dimensional electrodynamic distribution model of the transformer winding under short-circuit conditions based on the physical principle of electromagnetic force, and calculates the dynamic electrodynamic values ​​of each phase winding in the radial and axial dimensions. The specific construction method of the three-dimensional electrodynamic distribution model used in the electrodynamic calculation module is as follows: The electrodynamic calculation module discretizes each phase winding of the transformer into 128 independent conductor units; The electrodynamic calculation module calculates the magnetic flux density vector at the location of the target conductor unit based on the instantaneous value of the short-circuit current, the asymmetry coefficient, and the relative spatial position of each conductor unit. For each conductor element, the magnetic force exerted by the currents of all other conductor elements under the current short-circuit current excitation is calculated according to Biosavart's law. The calculation fully considers the specific spatial geometry of the winding, the relative positional relationship between the conductors, and the influence of the core magnetic circuit. Finally, the Lorentz force on each conductor unit is vector-combined to obtain the magnitude and position of the resultant force of the phase winding in the radial and axial directions. The mechanical stress assessment module is connected to the electrodynamic calculation module. Based on the theory of mechanics of materials, it establishes the stress-strain constitutive relationship of the winding conductor. Combining the winding structure parameters and the characteristics of the insulation material, it converts the electrodynamic values ​​into the real-time mechanical stress distribution and deformation displacement of the winding. The dynamic assessment and decision-making module integrates preset multi-level assessment thresholds and historical data comparison units. It receives real-time stress data output by the mechanical stress assessment module, compares and analyzes it with the transformer's factory withstand standards, historical short-circuit event records, and life decay models based on years of operation, and generates a quantitative assessment level of the current short-circuit event on the transformer's health status. The results output and early warning module is connected to the dynamic evaluation and decision-making module. It is used to convert the evaluation level into specific human-computer interaction information and automatically trigger multi-level early warning signals when the evaluation level exceeds the preset safety threshold.

2. The dynamic monitoring and evaluation system for short-circuit withstand capability of dry-type power transformers according to claim 1, characterized in that, The high-speed data acquisition module includes a signal conditioning circuit and a synchronous sampling unit; The signal conditioning circuit uses isolation amplification and anti-aliasing filtering technology to preprocess the original current and voltage signals, ensuring that the signal amplitude is adapted to the input range of the analog-to-digital converter and effectively suppressing high-frequency noise interference. The synchronous sampling unit uses the GPS clock signal as the synchronization reference to ensure that the acquisition time of the three-phase current and voltage signals is strictly aligned, and the time synchronization error is no more than 1 microsecond.

3. The dynamic monitoring and evaluation system for short-circuit withstand capability of dry-type power transformers according to claim 1, characterized in that, The short-circuit feature parsing module performs the short-circuit feature extraction process as follows: The module first performs full waveform monitoring on the input current signal. When the instantaneous value of any phase current exceeds twice its rated current value and the duration exceeds one power frequency cycle, it is determined that a short circuit event has started and the moment is recorded as time 0. The module then uses time 0 as a reference to analyze the current waveform data within two power frequency cycles, and determines the peak amplitude of the short-circuit current through a peak-finding algorithm. The asymmetry coefficient is calculated by extracting the decay time constant of the DC component of the short-circuit current and the initial phase angle; The short-circuit duration is calculated from time 0 until the current value drops to less than 1.5 times the rated current and remains there for 3 power frequency cycles.

4. The dynamic monitoring and evaluation system for short-circuit withstand capability of dry-type power transformers according to claim 1, characterized in that, The execution process of the mechanical stress assessment module includes two stages: stress calculation and deformation analysis. During the stress calculation stage, the module uses the finite element method to calculate the maximum equivalent stress value inside the winding based on the electrodynamic data provided by the electrodynamic calculation module, combined with the elastic modulus, yield strength, and support stiffness parameters of the winding conductor and the insulation material. During the deformation analysis phase, the module solves the deformation control equation of the winding based on the calculated stress distribution, outputs the deformation displacement at key locations, and ensures that the displacement does not exceed the maximum allowable deformation limit of the transformer design.

5. The dynamic monitoring and evaluation system for short-circuit withstand capability of dry-type power transformers according to claim 1, characterized in that, The dynamic evaluation and decision-making module has three preset levels of evaluation thresholds: immediate warning threshold, attention observation threshold, and safe operation threshold. The immediate warning threshold corresponds to the winding mechanical stress reaching 80% of the material's yield strength or the deformation displacement reaching 90% of the design limit; Note that the threshold corresponds to a stress reaching 50% of the yield strength or a deformation displacement reaching 60% of the design limit; The safe operating threshold corresponds to stress below 30% of the yield strength and deformation displacement below 40% of the design limit.

6. The dynamic monitoring and evaluation system for short-circuit withstand capability of dry-type power transformers according to claim 5, characterized in that, The historical data comparison unit stores the evaluation results of all short-circuit events of the transformer. The dynamic assessment and decision-making module compares the current stress data with historical data to determine trends. If it finds that the stress increment caused by the current short circuit is 30% higher than the historical average, it will automatically upgrade the assessment level.

7. The dynamic monitoring and evaluation system for short-circuit withstand capability of dry-type power transformers according to claim 1, characterized in that, The dynamic evaluation and decision-making module also integrates a lifespan decay model. The model takes the transformer’s cumulative years of operation and the number of major short-circuit events as input, and dynamically corrects the transformer’s remaining mechanical life through a preset insulation material aging equation and mechanical fatigue cumulative damage model. The current assessment results will be compared with the withstand capability corrected by the life decay model, thus more accurately reflecting the actual health status of the transformer.

8. The dynamic monitoring and evaluation system for short-circuit withstand capability of dry-type power transformers according to claim 1, characterized in that, The result output and early warning module supports multiple output interfaces and early warning methods; The output interfaces include an Ethernet communication interface, a 4G wireless communication module, and a local display unit, which are used to upload the assessment level, key parameters, and early warning information to the superior monitoring system or display them locally. The early warning methods are triggered according to the assessment level, including local audible and visual alarms, SMS push notifications to designated maintenance personnel, and sending emergency signals to the power grid dispatch system.

9. The dynamic monitoring and evaluation system for short-circuit withstand capability of dry-type power transformers according to claim 1, characterized in that, The system operates within a multi-timescale hierarchical framework; The hierarchical framework includes a planning layer, a tactical layer, and an operational layer; The planning layer is used to set long-term supply and demand targets and game theory frameworks on a monthly timescale. The tactical layer is used to plan major logistics routes and inventory strategies on a weekly timescale. The operation layer is used to perform evaluation analysis output by the dynamic evaluation decision module and early warning signals output by the result output and early warning module on a daily time scale.

Images