Excitation system dynamic characteristic analysis method and system

Abstract

The invention discloses an excitation system dynamic characteristic analysis method and system, and the method comprises the steps: reading wave recording file data of an excitation system, extracting original data containing analog quantity data and timestamp information, carrying out the classification of the original data, and generating a configuration file and a data file which meet the COMTRADE standard; performing dynamic characteristic parameter calculation on the classified and sorted original data, wherein dynamic characteristic parameters comprise quantitative calculation of overshoot, adjustment time, oscillation times, de-excitation time, excitation time and over-excitation inverse time limit; the method comprises the following steps: establishing a preset threshold value of a dynamic characteristic parameter, comparing a calculation result with the preset threshold value, generating a PID parameter adjustment instruction, an inversion failure alarm signal or an excitation control optimization strategy according to an over-limit parameter type, and comparing dynamic parameter calculation with the threshold value so as to identify a fault reason and generate an optimization instruction. The problem of low efficiency caused by the fact that traditional fault recognition depends on artificial experience is solved, and the method has the advantages of improving fault analysis efficiency and improving safety of a power system.

Description

Technical Field

[0001] The present invention relates to the technical field of power system relay protection, and in particular to a method and system for analyzing dynamic characteristics of an excitation system. Background Art

[0002] With the intelligent transformation of power systems and the continuous improvement of industrial automation, real-time monitoring of power equipment operating status and fault diagnosis have become core requirements for ensuring grid security. In particular, in the field of power safety incident analysis, there is an urgent need to develop technical means capable of millisecond-level event backtracking and multi-dimensional parameter correlation analysis. As the core device for voltage regulation and reactive power support for generator sets, the dynamic characteristics of the excitation system directly impact the transient stability and grid-source coordination performance of the power system.

[0003] Currently, power grid dispatching agencies and the Electric Power Research Institute have strict PSASP / BPA model calibration requirements for excitation systems. However, traditional operation and maintenance methods still have significant flaws: operators must manually analyze massive amounts of recorded data (including analog, switching, and SOE events) and estimate the regulation characteristic parameters using empirical formulas. This process not only takes hours but also makes it difficult to accurately identify hidden faults such as PID parameter mismatches and inverter logic conflicts. More seriously, when AVR misregulation causes unit oscillation or demagnetization failure, existing methods cannot achieve online calibration and adaptive adjustment of characteristic parameters, often requiring shutdown for manual testing. This can cause significant economic losses to large power plants with baseload missions. Summary of the Invention

[0004] In order to overcome the technical defects of the existing excitation system, such as the single dimension of dynamic characteristic parameter calculation and the difficulty in tracing the source of hidden faults, the present invention provides a method and system for analyzing the dynamic characteristics of the excitation system.

[0005] In order to solve the above problems, the present invention is implemented according to the following technical solutions:

[0006] In a first aspect, the present invention provides a method for analyzing the dynamic characteristics of an excitation system, comprising the following steps: S100: reading the recorded file data of the excitation system, extracting the original data containing analog data and timestamp information, classifying and organizing the original data, and generating configuration files and data files that comply with the COMTRADE standard; S200: calculating the dynamic characteristic parameters of the classified and organized original data, wherein the dynamic characteristic parameters include quantitative calculation of overshoot, adjustment time, number of oscillations, demagnetization time, excitation start time, and overexcitation reverse time limit; S300: establishing a preset threshold value of the dynamic characteristic parameter, comparing the calculation result with the preset threshold value, and generating a PID parameter adjustment instruction, an inverter failure alarm signal, or an excitation start control optimization strategy according to the type of the over-limit parameter.

[0007] In combination with the first aspect, the present invention provides a first specific implementation of the first aspect. Specifically, step S100 also includes: performing waveform visualization reproduction processing on the COMTRADE standard configuration file and data file, and establishing a dual cursor time and interval selection mechanism; wherein the dual cursor time and interval selection mechanism can obtain time axis synchronous positioning, arbitrary data interval interception and dynamic waveform zoom display.

[0008] In combination with the first aspect, the present invention provides a second specific implementation of the first aspect. Specifically, the calculation of the dynamic characteristic parameters is achieved through the following sub-steps: constructing a dynamic characteristic analysis database, the database receiving a time interval data array selected by a dual cursor, and executing to obtain the characteristic values in the time interval data array selected by the dual cursor; wherein the characteristic values include a maximum value, a starting value and a steady-state value; and calculating the overshoot, adjustment time and number of oscillations based on the characteristic values.

[0009] In combination with the first aspect, the present invention provides a third specific implementation of the first aspect. Specifically, the demagnetization time calculation formula is: T=T2-T1; wherein T1 is the start time of inversion, and T2 is the time when the voltage decays to 36.8% of the start time of inversion.

[0010] In combination with the first aspect, the present invention provides a fourth specific implementation of the first aspect. Specifically, the overexcitation inverse time calculation formula is: Where: I L is the measured excitation current; I FEL is the peak current multiple; T q Allow time for the top value; I LEL It is the long-term allowable current reference value.

[0011] In combination with the first aspect, the present invention provides a fifth specific implementation of the first aspect. Specifically, the optimization strategy is as follows: when in a step response state, an overshoot threshold, an oscillation number threshold and an adjustment time threshold are input; when the measured overshoot is greater than the overshoot threshold or the oscillation number threshold is greater than the oscillation number threshold or the adjustment time is greater than the adjustment time threshold, a PID parameter adjustment instruction is triggered; when in an inverter state, a five-level warning threshold is set for the demagnetization time, the first four levels trigger a graded warning signal, and the fifth level generates an inverter failure alarm and executes a shutdown instruction; when in an excitation state, a five-level warning threshold is set, the first four levels trigger a graded warning signal, and an optimization strategy for a soft start graded setting value and a given slope of the machine-end voltage is generated; when reaching the fifth level, if the machine-end voltage amplitude is lower than 10% of the rated machine-end voltage, and the excitation current amplitude is lower than 6% of the rated excitation current, an excitation failure alarm is issued and an excitation circuit self-test procedure is executed.

[0012] In the second aspect, the present invention also provides a dynamic characteristic analysis system of an excitation system, comprising: a data acquisition module for reading the recorded file data of the excitation system, extracting the original data containing analog data and timestamp information, classifying and organizing the original data, and generating configuration files and data files that comply with the COMTRADE standard; a data calculation module for calculating the dynamic characteristic parameters of the classified and organized original data; an analysis module for establishing a preset threshold value of the dynamic characteristic parameter, comparing the calculation result with the preset threshold value, and generating a PID parameter adjustment instruction, an inverter failure alarm signal, or an excitation control optimization strategy according to the type of the out-of-limit parameter.

[0013] In combination with the second aspect, the present invention provides a first specific implementation of the second aspect. Specifically, the data acquisition module is further used to perform waveform reproduction processing on the configuration files and data files of the COMTRADE standard, and establish a dual cursor time and interval selection mechanism.

[0014] In combination with the second aspect, the present invention provides a second specific implementation of the second aspect. Specifically, the analysis module includes: a step response analysis submodule, which is used to generate a PID parameter adjustment instruction when any parameter among the measured overshoot, number of oscillations or adjustment time exceeds a preset threshold; an inversion analysis unit, which is used to execute the first four levels of graded warning signal output or the fifth level of inverter failure alarm and shutdown instruction based on the comparison result of the measured value of the demagnetization time with the five-level preset warning threshold; an excitation analysis unit, which is used to execute the first four levels of graded warning signal output, soft start graded set value optimization strategy and machine-end voltage given slope adjustment strategy, or the fifth level excitation failure alarm and excitation circuit self-test instruction based on the comparison result of the excitation process parameters with the five-level warning threshold; wherein the fifth level excitation failure judgment condition is: the machine-end voltage amplitude is lower than 10% of the rated machine-end voltage and the excitation current amplitude is lower than 6% of the rated excitation current.

[0015] In combination with the second aspect, the present invention provides a third specific implementation of the second aspect. Specifically, the data acquisition module includes: a file parsing unit that supports COMTRADE standard file parsing and is compatible with ASCII and binary formats; a cursor control unit that realizes dual-cursor linkage selection and dynamic calculation of interval parameters, and the dynamic calculation of interval parameters specifically involves intercepting data points within the interval for calculation.

[0016] Compared with the prior art, the present invention has the following beneficial effects:

[0017] 1. Quantify the dynamic characteristic parameters of the excitation system, including overshoot, adjustment time, and number of oscillations. These quantitative indicators can objectively and accurately reflect the dynamic performance of the excitation system, providing a quantifiable basis for evaluating system stability, response speed, and damping characteristics, changing the previous reliance on experience and qualitative judgment.

[0018] 2. By establishing preset thresholds for dynamic characteristic parameters and comparing them with calculated results, real-time monitoring of the excitation system's operating status and abnormality warnings are achieved. When parameters exceed the limit, PID parameter adjustment instructions, inverter failure alarm signals, or excitation control optimization strategies can be generated in a timely manner. This helps: provide early warnings when parameters approach or exceed the threshold, avoiding the escalation of faults. Based on the type of out-of-limit parameters, targeted control instructions or optimization strategies are automatically generated, such as adjusting PID parameters to improve system response, or optimizing excitation control logic to increase the excitation success rate, thereby improving the control accuracy and stability of the excitation system.

[0019] The entire analysis process, from data reading and processing to characteristic calculation, threshold comparison, and strategy generation, forms a closed loop. This reduces manual intervention and improves operation and maintenance efficiency. Real-time monitoring and dynamic adjustment also enhance the reliability and stability of the excitation system, ensuring the safe operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings, wherein:

[0021] Figure 1 It is a flow chart of a method for analyzing dynamic characteristics of an excitation system according to the present invention. DETAILED DESCRIPTION

[0022] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0023] In existing technology, power system maintenance personnel rely primarily on personal experience to analyze recorded waveform data when analyzing excitation system faults. The lack of a standardized processing process and the differences in data formats between equipment manufacturers lead to inefficient analysis. Maintenance personnel must repeatedly compare waveforms and manually calculate dynamic parameters, making it difficult to quickly identify problems such as abnormal overshoot or demagnetization time deviations. This often necessitates downtime testing to identify the fault point, resulting in system interruptions and financial losses.

[0024] To address these issues, the R&D team discovered that the bottlenecks of traditional approaches lie in the discretization of data processing and the lag in decision-making mechanisms. By studying power system fault diagnosis cases, they discovered a strong correlation between dynamic characteristic parameters and control system responses. Based on this, they proposed establishing an automated data processing pipeline to convert discrete waveform data into computable structured information, combining it with preset thresholds to enable rapid diagnosis. The key breakthrough lies in transforming empirical judgments into quantitative calculation models and eliminating device differences through standardized data interfaces.

[0025] Example 1

[0026] like Figure 1 As shown, the present invention provides a method for analyzing the dynamic characteristics of an excitation system, comprising the following steps:

[0027] S100: Read the recorded file data of the excitation system, extract the original data including analog data and time stamp information, classify and organize the original data and generate configuration files and data files that comply with the COMTRADE standard;

[0028] S200: Calculating dynamic characteristic parameters of the classified and organized raw data, wherein the dynamic characteristic parameters include quantitative calculation of overshoot, adjustment time, number of oscillations, demagnetization time, excitation start time, and overexcitation reverse time limit;

[0029] S300: Establishing preset thresholds for dynamic characteristic parameters, comparing calculation results with the preset thresholds, and generating PID parameter adjustment instructions, inverter failure alarm signals, or excitation control optimization strategies based on the type of over-limit parameters.

[0030] Specifically, the excitation system recording file data is read and the analog quantity and timestamp information is extracted. The original data is classified to generate COMTRADE standard files. Dynamic parameters such as overshoot and adjustment time are calculated. A threshold comparison mechanism is established and a technical solution is generated to generate corresponding control instructions.

[0031] Among them, waveform file data parsing refers to extracting analog quantities such as excitation voltage, excitation current, machine-end voltage, machine-end current, active power, reactive power, rotor temperature, and their corresponding time series information from the power system fault recording device. File reading algorithms can be used to achieve binary data conversion to ensure the integrity of the original data. COMTRADE standard file generation refers to defining metadata such as channel configuration and sampling rate in accordance with IEEE standards, achieving cross-platform compatibility through format conversion modules, and solving the problem of multi-source data integration. The calculation of dynamic characteristic parameters involves establishing a mathematical model for overshoot, obtaining the maximum deviation value through a peak detection algorithm, and calculating the overshoot percentage in combination with the steady-state value. The threshold comparison mechanism performs real-time matching through a preset parameter database. When it is detected that the demagnetization time exceeds the safety threshold, the protection logic is automatically triggered.

[0032] Specifically, the data standardization stage parses the binary stream of the original recording file, separates the timestamp and analog channel, and reconstructs it into a configuration file containing the sampling rate and channel name. The dynamic parameter calculation module uses a sliding window algorithm to traverse the time series data, locate the maximum value and steady-state value within the step response interval, and calculate the adjustment time based on the time difference. When the number of oscillations detected exceeds the preset value, the PID parameter optimization algorithm is automatically called to generate the adjustment coefficient. For the demagnetization process in the inverter state, an exponential decay model is used to calculate the time required for the voltage to drop to the set ratio. If it exceeds the allowable range, an alarm signal is immediately issued.

[0033] Traditional methods require manual capture of waveforms and manual measurement of parameter values. This system uses an automated data analysis process to calculate multiple parameters to address the problem of existing excitation systems relying on a single threshold to determine system failures. This solution uses a multi-dimensional parameter collaborative analysis mechanism, for example, combining overshoot and adjustment time to jointly determine system stability. Existing systems lack a closed-loop control mechanism. This dynamic characteristics analysis system directly converts the analysis results into PID adjustment instructions, forming a dynamic optimization loop. The excitation system has functions such as rapid fault location and automatic repair. The standardized data interface protocol eliminates equipment compatibility issues, the dynamic parameter modeling engine replaces manual measurement errors, and the intelligent threshold response mechanism shortens fault response time. In the scenario of sudden generator load increase, the system can quickly complete overshoot calculations and adjust the excitation controller parameters, avoiding the operational interruption caused by the traditional method of downtime testing.

[0034] In one embodiment, step S100 further includes: performing waveform visualization reproduction processing on the COMTRADE standard configuration file and data file, and establishing a dual cursor time and interval selection mechanism; wherein the dual cursor time and interval selection mechanism can obtain time axis synchronous positioning, arbitrary data interval interception and dynamic waveform zoom display.

[0035] Waveform visualization processing converts standard waveform data into an interactive graphical interface. This is achieved by invoking a graphics rendering engine to coordinate map and draw waveforms on the time series data in the COMTRADE file. This process transforms the raw data from discrete values into continuous waveform curves. The dual-cursor time and interval selection mechanism allows interactive manipulation of the timeline using two movable markers. Specifically, the graphical interface's event monitoring module captures cursor movement events and updates the associated data area in real time. This mechanism enables operators to precisely define the scope of analysis.

[0036] Specifically, during the waveform visualization and reproduction process, the time axis data and voltage and current parameters are synchronously mapped to a two-dimensional coordinate system to form a scalable time-amplitude waveform diagram. The dual-cursor linkage module calculates the projection position of the cursor on the time axis in real time by monitoring the displacement events of the mouse or touch device, and synchronously updates the time span value between the two cursors. When performing dynamic zoom operations, the coordinate transformation algorithm resamples and interpolates the current display interval to maintain the continuity of the waveform curve. The time axis synchronous positioning function ensures the consistency of the time base of multiple signals by establishing a master-slave channel association mechanism. When a cursor operation is performed on any channel, the other channels automatically and synchronously display the data status of the corresponding time point.

[0037] Traditional waveform analysis tools usually use a fixed window display mode, which cannot achieve dynamic observation of waveform details. Conventional single cursor solutions only support the positioning of isolated time points, while the dual cursor linkage mechanism adopted by this dynamic characteristic analysis system can directly obtain complete data characteristics within the time interval span by establishing a dynamic association between the starting point and the end point. The static display method in the existing technology requires repeated adjustment of the display ratio to observe waveforms of different time scales. This system realizes continuous and smooth waveform zooming function by integrating a multi-touch zoom algorithm, while also ensuring the precise positioning function of the time axis of the waveform data, solving the problem that traditional methods cannot freely select analysis intervals. The dynamic zoom display function breaks through the limitation of the fixed display ratio, allowing the operator to observe the overall trend and local details of the waveform at the same time. The dual cursor linkage mechanism improves the accuracy of parameter extraction and operational efficiency by establishing a direct association between the time interval and the data characteristics.

[0038] In one embodiment, the calculation of dynamic characteristic parameters is achieved through the following sub-steps: constructing a dynamic characteristic analysis database, which receives a time interval data array selected by a dual cursor, and executing to obtain characteristic values in the time interval data array selected by the dual cursor, wherein the characteristic values include a maximum value, a starting value, and a steady-state value; and calculating the overshoot, adjustment time, and number of oscillations based on the characteristic values.

[0039] The dynamic characteristics analysis database refers to a structured storage system used to store and manage dynamic process data. Specifically, it can be implemented using a relational database or a time series database. By establishing a data table structure to store timestamps, analog values, and feature tag fields, it supports the rapid retrieval and batch calculation of subsequent feature values, solving the problems of scattered storage and low query efficiency in traditional manual data recording. The dual-cursor selected time interval data array refers to the data set of all sampling points within the analysis period determined by two movable marking points in the graphical interface. Specifically, an array or linked list data structure can be used to store the start time, end time, and corresponding data sequence, achieving accurate interception of specific dynamic processes (such as the step response stage) and avoiding the time period selection bias caused by manual experience judgment. The maximum value refers to the peak value reached by the signal within the selected time interval, which can be obtained in real time using the traversal comparison algorithm; the starting value refers to the initial value of the signal corresponding to the trigger moment of the dynamic process, which can be determined by detecting the step instruction trigger time point or setting the threshold mutation point; the steady-state value refers to the final value of the signal that tends to be stable after the dynamic process ends, which can be calculated using the sliding window mean filtering algorithm. These three eigenvalues serve as the basic parameters for quantitative analysis and provide data support for the calculation of core indicators such as overshoot and adjustment time.

[0040] Specifically, after the excitation system recording data is imported, the dynamic characteristics analysis database receives the time interval data array determined by the dual cursor operation. For example, in step response analysis, by moving the dual cursors to accurately select the complete process data from the issuance of the step instruction to the system returning to stability, the database automatically stores all voltage and current sampling values within this interval. The eigenvalue extraction algorithm is then executed: the data array of this interval is traversed to obtain the maximum value V of the voltage waveform. max , the starting value V corresponding to the triggering moment of the positioning step instruction start , and use the sliding window algorithm to calculate the steady-state value V ss Based on these three characteristic values, the overshoot is obtained by calculating the relative deviation percentage between the maximum value and the steady-state value. The overshoot PO calculation formula is: The adjustment time is determined by counting the time required for the initial value to enter the steady-state value allowable deviation range for the first time, specifically counting the time required for the initial value to enter the steady-state value allowable deviation range for the first time. start First time entering the steady state value V ss The time required for the deviation to be within the range of ±5%; the number of oscillations is counted by detecting the frequency of the voltage waveform crossing above and below the steady-state value, specifically by the first time the controlled quantity reaches the starting value V start The absolute value is less than 2% of the steady-state value V ss The entire process is automated through the database's batch data processing capabilities, replacing the traditional inefficient operations of manually reading waveforms and measuring parameters.

[0041] Traditional methods rely on manual experience to visually select analysis intervals from waveform graphs, resulting in inaccurate time segment selection and large eigenvalue measurement errors. For example, the calculation of adjustment time requires manual judgment of the stabilization moment, which is susceptible to subjective factors. However, the dynamic characteristics analysis of the excitation system uses a dual-cursor linkage mechanism to programmatically select the analysis interval, combined with an automatic eigenvalue extraction algorithm from a database, eliminating human operational errors. Existing techniques typically use manual counting to calculate the number of oscillations, which can easily lead to omissions in high-frequency oscillations. However, the dynamic characteristics analysis of the excitation system achieves precise counting through programmed detection of waveform zero crossings. Furthermore, existing techniques lack a unified data storage structure, resulting in different personnel using inconsistent eigenvalue calculation methods when analyzing the same data. This dynamic characteristics analysis system, by constructing a standardized dynamic characteristics analysis database, ensures the repeatability of the calculation process and the comparability of the results. The dynamic characteristics analysis system's automated and precise calculation of the dynamic characteristics parameters of the excitation system addresses the inefficiency and inconsistent results associated with traditional methods relying on manual experience. By combining a dual cursor mechanism with an eigenvalue algorithm, key dynamic processes can be quickly located within complex waveforms and core parameters extracted, providing a reliable data foundation for subsequent threshold comparison and control strategy generation. For example, in step response overshoot analysis, the system automatically identifies the maximum voltage deviation and calculates the percentage. Compared to traditional manual measurement methods, calculation time is reduced to milliseconds and accuracy is improved by an order of magnitude. This dynamic characteristic analysis of the excitation system is particularly suitable for rapid analysis scenarios after power system faults, enabling operations and maintenance personnel to quickly obtain quantitative indicators and formulate adjustment strategies.

[0042] The traditional method relies on manual experience to visually select the analysis interval in the waveform diagram, which has the problems of inaccurate time period interception and large characteristic value measurement errors. For example, the calculation of the adjustment time requires manual judgment of the stable moment, which is easily affected by subjective factors. The dynamic characteristic analysis of the excitation system of the present application realizes the programmed interception of the analysis interval through a dual cursor linkage mechanism, combined with the automatic extraction algorithm of the characteristic value of the database, eliminating human operation errors. The statistics of the number of oscillations in the traditional method usually adopt a manual counting method, which is prone to omissions for high-frequency oscillations, while the dynamic characteristic analysis of the excitation system realizes accurate counting through programmed detection of the waveform zero crossing. In addition, the traditional method lacks a unified data storage structure, which may result in different personnel using inconsistent characteristic value calculation methods when analyzing the same data. The dynamic characteristic analysis of the excitation system ensures the repeatability of the calculation process and the comparability of the results by constructing a standardized dynamic characteristic analysis database. The excitation system solves the problems of low efficiency and poor result consistency caused by the traditional method's reliance on manual experience through the automated and precise calculation of dynamic characteristic parameters. By combining a dual cursor mechanism with an eigenvalue algorithm, key dynamic processes can be quickly located within complex waveforms and core parameters extracted, providing a reliable data foundation for subsequent threshold comparison and control strategy generation. For example, in step response overshoot analysis, the system automatically identifies the maximum voltage deviation and calculates the percentage. Compared to traditional manual measurement methods, calculation time is reduced to milliseconds and accuracy is improved by an order of magnitude. This dynamic characteristic analysis of the excitation system is particularly suitable for rapid analysis scenarios after power system faults, enabling operations and maintenance personnel to quickly obtain quantitative indicators and formulate adjustment strategies.

[0043] In one embodiment, the demagnetization time is calculated as follows: T=T2-T1; wherein the demagnetization time is equal to the time corresponding to the voltage decaying to 36.8% of the voltage value at the start of inversion minus the start of inversion time.

[0044] Among them, the inversion starting moment refers to the moment corresponding to the moment when the demagnetization control signal is triggered, which can be specifically implemented by adopting the trigger signal rising edge capture algorithm. The algorithm determines the starting reference of the demagnetization process by detecting the control signal level jump point. The moment when the voltage decays to 36.8% refers to the moment when the voltage value decays to the initial value during the demagnetization process and the natural exponential function decays to 1 / e (about 36.8%) of the initial value. Specifically, it can be implemented by fitting the voltage sampling data to the exponential decay curve, and the critical time point is determined by calculating the intersection of the curve and the 36.8% threshold.

[0045] Specifically, after the demagnetization process is triggered, the demagnetization start time is precisely locked by capturing the effective transition edge of the inverter control signal. Simultaneously, an exponential function is modeled for the voltage decay curve, and a mathematical analytical method is used to determine the time at which the voltage decays to 36.8% of its initial value. This proportional value, based on the decay characteristics of a natural exponential function, aligns with the physical laws of the demagnetization process in electrical systems. By calculating the difference between the two characteristic moments, the demagnetization duration can be directly calculated, thereby establishing a unified standard for defining the demagnetization process time interval.

[0046] Traditional methods typically rely on empirical values to set voltage decay thresholds or estimate demagnetization time through manual waveform observation, which is subject to strong subjectivity and inconsistent standards. This excitation system combines control signal trigger detection with mathematical modeling to transform the calculation of demagnetization time into a quantitative model with clear physical meaning, avoiding errors caused by manual intervention. By achieving objective calculation of demagnetization time, this excitation system overcomes the inefficiency of traditional methods caused by the lack of a unified time interval definition standard. It provides a verifiable time benchmark for determining whether the demagnetization process is normal, significantly improving the accuracy and consistency of excitation system fault analysis.

[0047] In a preferred embodiment, the overexcitation inverse time calculation formula is: Where: I L is the measured excitation current; I FEL is the peak current multiple; T q Allow time for the top value; I LEL It is the long-term allowable current reference value.

[0048] Among them, the measured excitation current refers to the current value passing through the excitation winding under the actual operating state. Specifically, Hall sensors or Rogowski coils can be used for real-time data acquisition to reflect the actual current level of the system's current working conditions. The peak current multiple refers to the ratio of the instantaneous maximum current that the excitation system can withstand to the rated current. It can be obtained through the equipment nameplate parameters or manufacturer's technical specifications and is used to characterize the equipment's overload resistance. The peak allowable time refers to the maximum time that the excitation system can operate sustainably under the peak current. It can be determined through equipment tolerance test data or historical operation records and is used to define the safety threshold under the equipment's extreme working conditions. The long-term allowable current reference value refers to the maximum allowable current of the excitation system under continuous operating conditions. Specifically, the benchmark parameters specified in international standards or industry specifications can be used to establish a unified overload assessment benchmark.

[0049] Traditional methods rely on empirical values or fixed thresholds to judge the overexcitation state, without considering the relationship between the dynamic changes in current and the tolerance characteristics of the equipment. The excitation system of this application uses mathematical formulas to couple the equipment limit parameters, real-time current data and standardized reference values to establish a corresponding relationship between current intensity and allowable time, avoiding the subjective errors of empirical judgment, and solving the problem that the protection time cannot be dynamically adjusted under different working conditions. The excitation system can dynamically calculate the allowable operating time based on the real-time excitation current changes, accurately determine the timing of overexcitation protection action, and provide an accurate overload protection strategy for the excitation system. This calculation method can directly guide the optimization of the action logic of the protection device, avoid equipment damage or system failure caused by premature or delayed triggering of protection, and effectively improve the safety and control reliability of the excitation system under overload conditions.

[0050] In a specific implementation process, the waveform analysis step, when in a step response state, inputs an overshoot threshold, an oscillation number threshold, and an adjustment time threshold; when the measured overshoot is greater than the overshoot threshold, or the oscillation number is greater than the oscillation number threshold, or the adjustment time is greater than the adjustment time threshold, generates a PID parameter adjustment instruction; when in an inverter state, a five-level warning threshold is set for the demagnetization time, the first four levels trigger a graded warning signal, and the fifth level generates an inverter failure alarm and executes a shutdown instruction; when in an excitation state, a five-level warning threshold is set, the first four levels trigger a graded warning signal, and an optimization strategy for the soft start graded setting value and the given slope of the generator-end voltage is generated; when reaching the fifth level, if the generator-end voltage amplitude is lower than 10% of the rated generator-end voltage and the excitation current amplitude is lower than 6% of the rated excitation current, a starting failure alarm is issued and an excitation circuit self-test program is executed, and starting the excitation circuit self-test program includes: detecting the output voltage of the excitation transformer; checking the thyristor trigger pulse of the power cabinet; and verifying the open and close positions of the demagnetization switch.

[0051] The step response state refers to the dynamic response of the excitation system when stimulated by a step input signal. This can be determined by monitoring sudden voltage or current changes and used to identify the system's transient regulation needs. Overshoot refers to the maximum deviation of the output from the steady-state value during the system's response. This can be determined by calculating the percentage difference between the peak value and the steady-state value and is used to assess system stability. PID adjustment instructions are control optimization instructions based on proportional, integral, and differential parameters. They can be generated from a preset rule base or algorithm model and used to dynamically adjust controller parameters. The inversion state refers to the operating phase of the excitation system during deexcitation. This can be determined by detecting the inversion trigger signal or voltage polarity change and is used to monitor the safety of the deexcitation process. The deexcitation time refers to the interval from the start of inversion to the time the voltage decays to a specified threshold. This can be calculated by recording the difference between the start of inversion and the time the voltage decays to 36.8% and is used to assess deexcitation efficiency. The starting state refers to the process by which the excitation system builds voltage from its initial state. This can be identified by monitoring the generator-side voltage rise curve and is used to optimize the control strategy during the startup phase.

[0052] In a preferred embodiment, when it is detected that the system is in a step response state, the following steps are performed: receiving dynamic threshold inputs for overshoot, number of oscillations and adjustment time through a human-computer interaction interface; collecting the overshoot, number of oscillations and adjustment time parameter values of the control process in real time, and when any parameter exceeds the corresponding threshold, triggering the PID parameter adjustment process; receiving the proportional coefficient, integral coefficient and differential coefficient values of the current PID controller, and generating an adjustment instruction based on the fuzzy control rule library, the adjustment instruction including: when the overshoot is greater than a preset value, giving priority to adjusting the differential coefficient to suppress the overshoot; when the number of oscillations is greater than a preset value, synchronously adjusting the proportional coefficient and the integral coefficient to increase the system damping; when the adjustment time is greater than a preset value, optimizing the proportional coefficient to speed up the response speed.

[0053] In a preferred embodiment, when it is detected that the system is in the inverter state, the following steps are performed: a five-level threshold classification of the demagnetization time T is set: T1∈[6,7)s, T2∈[7,8)s, T3∈[8,9)s, T4∈[9,10)s, T5≥10s; when T reaches Ti (i=1-4), the i-th level warning signal is displayed through the monitoring interface; when T≥T5, the inverter failure protection action is triggered, including: cutting off the inverter circuit power supply; recording the waveform data at the time of the fault; and generating a fault code containing a timestamp.

[0054] In a preferred embodiment, when it is detected that the system is in the excitation state, the following steps are performed: a five-level threshold classification of the excitation time Tq is set: Tq1∈[15,16)s, Tq2∈[16,17)s, Tq3∈[17,18)s, Tq4∈[18,19)s, Tq5≥20s; when Tq reaches Tqi (i=1-4), the i-th level alarm processing is performed, including: adjusting the soft start classification setting value, the adjustment range is 10%-20% of the previous value; modifying the given slope of the generator-end voltage; when Tq≥Tq5, performing the following dual judgment: detecting whether the generator-end voltage amplitude is lower than 10% of the rated generator-end voltage and the excitation current amplitude is lower than 6% of the rated excitation current at the same time; if so, triggering the excitation failure protection, executing: locking the excitation circuit contactor; generating a fault log containing the actual measured values of the generator-end voltage amplitude and the excitation current amplitude; uploading the fault signal to the monitoring system through the communication interface.

[0055] In a preferred embodiment, when the rated generator voltage starting mode is preset, in addition to satisfying the aforementioned starting state optimization conditions, the following additional constraints are enforced: voltage overshoot is strictly limited to no more than 5% of the rated voltage, the number of oscillations is strictly limited to no more than 3, and the adjustment time is strictly limited to no more than 15 seconds. If any parameter is detected as out of limit, the following actions are performed: optimizing the soft-start tiered setting strategy, dynamically modifying PID parameters, and correcting the voltage reference rate. The optimized tiered setting strategy includes an adjustment step size of no less than 5% of the rated generator voltage for each tier; dynamically modifying PID parameters with a proportional coefficient adjustment range of [0.8, 1.2] Kp0 and an integral coefficient adjustment range of [0.5, 1.5] Ki0 (Kp0 and Ki0 are the initial setting values of the proportional coefficient and the integral coefficient, respectively); and adjusting the voltage reference rate of rise within a range of [0.5, 1.0] V0 / s (V0 is the rated voltage).

[0056] Traditional methods only trigger a unified alarm when parameters exceed their limits, failing to take differentiated measures based on the operating phase. For example, existing technologies fail to automatically adjust PID parameters during the step response phase, leading to repeated system oscillations. Alternatively, when the demagnetization time exceeds its limit, data is recorded but no timely alarm is issued, increasing the risk of failure. This excitation system addresses the disconnect between parameter overlimit processing and the operating state through a mechanism for classifying and differentially processing operating states. Furthermore, in different operating phases of the excitation system, corresponding control instructions, alarm signals, or optimization strategies can be automatically triggered based on the type of overlimit parameter. For example, oscillations can be quickly suppressed during the step response phase to reduce adjustment time; timely alarms can be issued during the inversion phase to prevent equipment damage; and the startup process can be optimized during the excitation phase to avoid voltage overshoots. This improves fault handling efficiency, reduces manual analysis costs, and enhances system operational stability.

[0057] Example 2

[0058] The present invention proposes a dynamic characteristic analysis system for an excitation system, comprising: a data acquisition module for reading recorded file data of the excitation system, extracting raw data containing analog data and timestamp information, classifying and arranging the raw data, and generating configuration files and data files that comply with the COMTRADE standard; a data calculation module for calculating dynamic characteristic parameters of the classified and arranged raw data; and an analysis module for establishing preset thresholds for the dynamic characteristic parameters, comparing the calculation results with the preset thresholds, and generating PID parameter adjustment instructions, inverter failure alarm signals, or excitation control optimization strategies according to the types of the over-limit parameters.

[0059] The data acquisition module is the unit that implements data format standardization, which can be achieved by configuring a parser for COMTRADE standard files. This module converts recorded data of different formats into standardized, analyzable data, resolving the compatibility issues of the original data. The data calculation module is the unit that performs dynamic parameter quantification calculations, which can be achieved through built-in algorithm models such as overshoot and demagnetization time. This module converts complex system characteristics into comparable numerical indicators. The analysis module is the unit that performs threshold comparison and strategy generation, which can be achieved through a preset threshold library and logical judgment rules. This module automatically matches the corresponding control strategy based on the type of parameter exceeding the limit.

[0060] Specifically, the data acquisition module standardizes the raw recorded data, and the data calculation module extracts characteristic values within the time interval to calculate dynamic parameters. When the analysis module detects that the overshoot in the step response phase exceeds the threshold, it automatically triggers PID parameter adjustment instructions. If the inverter monitor detects an abnormal demagnetization time, it immediately generates an inverter failure alarm signal. For parameter out-of-limit conditions during the excitation process, a control strategy including voltage ramp rate optimization is generated. These three modules form a closed-loop processing flow, enabling automated operation from data analysis to control decision-making.

[0061] Traditional methods rely on manual experience to analyze recorded data, and multiple tests are required to locate the cause of the fault. However, this solution uses a standardized data interface and an automated analysis module to directly output the fault type and processing strategy, avoiding the loss of downtime caused by repeated manual testing. The existing technology lacks a threshold judgment mechanism for different working conditions. The excitation system can match multiple operating states such as step, inversion, and excitation through the set dynamic parameter threshold library to achieve accurate fault classification and identification, thus solving the problem of downtime losses caused by low efficiency of manual analysis. The standardized data interface eliminates the obstacles to multi-format data parsing, the dynamic parameter calculation module realizes the quantitative evaluation of complex system characteristics, and the threshold comparison mechanism can quickly identify the type of over-limit fault and generate corresponding control instructions, which shortens the fault handling cycle of the excitation system and effectively improves the operational reliability.

[0062] In a preferred embodiment, the data acquisition module includes a file parsing unit and a cursor control unit. The file parsing unit supports COMTRADE standard file parsing and is compatible with ASCII and binary formats; the cursor control unit implements dual cursor linkage selection and dynamic calculation of interval parameters.

[0063] The file parsing unit refers to a processing unit capable of recognizing the waveform file structure defined by the COMTRADE international standard. This can be implemented using a hybrid parsing engine with a dual-channel processing architecture consisting of an ASCII text decoder and a binary stream parser. By setting an automatic format recognition algorithm, the parsing mode can be dynamically switched based on the file header identifier, resolving multi-format compatibility issues. The cursor control unit refers to a waveform manipulation module with a dual-position marking function. This can be implemented using an event-driven coordinate synchronization mechanism. The main and secondary cursors are dynamically associated through a timeline mapping relationship. When either cursor position changes, the interval parameters are recalculated, enabling instant analysis of any waveform segment.

[0064] Specifically, during the operation phase of the file parsing unit, the hybrid parsing engine simultaneously loads the COMTRADE configuration file (.cfg) and the data file (.dat), and automatically identifies the ASCII character set or binary encoding features through the format detection module. For composite format files, a segmented parsing strategy is used to process different data blocks separately. During the operation phase of the cursor control unit, after the operation interface generates a visual waveform, the user drags the double cursors to form an analysis interval, triggering the dynamic calculation thread to automatically extract the voltage and current characteristic values within the interval. The dual cursor linkage mechanism maintains the accuracy of the interval span through a timestamp synchronization algorithm, ensuring that the parameter calculation results strictly correspond to the selected waveform segment.

[0065] Traditional waveform analysis tools only support a single file format and require manual conversion of data formats, while this excitation system achieves automatic compatibility with multiple formats through a dual-channel parsing engine. Existing waveform analysis systems use a single-point cursor positioning method and cannot directly obtain the complete time interval corresponding to the characteristic parameters. This solution uses a dual-cursor dynamic linkage mechanism to achieve accurate interception and parameter calculation of any interval while maintaining the complete display of the waveform, thereby achieving seamless analysis of excitation waveform data in different formats and eliminating the manual format conversion operation link. The dual-cursor linkage mechanism enables engineers to directly select any fault characteristic interval on the complete waveform, and the system automatically calculates the dynamic characteristic parameters within the interval, significantly shortening the processing chain from data parsing to characteristic analysis, and improving the accuracy of fault location and parameter calculation.

[0066] In a preferred embodiment, the analysis module includes a step response analysis submodule, which is used to generate a PID parameter adjustment instruction when any parameter among the measured overshoot, oscillation number or adjustment time exceeds a preset threshold; an inversion analysis unit, which is used to execute the first four levels of graded warning signal output or the fifth level of inverter failure alarm and shutdown instruction based on the comparison result of the measured value of the demagnetization time with the five-level preset warning threshold; an excitation analysis unit, which is used to execute the first four levels of graded warning signal output, soft start graded setting optimization strategy and machine-end voltage given slope adjustment strategy, or the fifth level of excitation failure alarm and excitation circuit self-test instruction based on the comparison result of the excitation process parameters with the five-level warning threshold; wherein the fifth level excitation failure judgment condition is: the machine-end voltage amplitude is lower than 10% of the rated machine-end voltage and the excitation current amplitude is lower than 6% of the rated excitation current.

[0067] Specifically, the step response analysis submodule continuously receives real-time sampling values of voltage and current from the data acquisition module. When it detects that the overshoot exceeds the preset threshold during the step response process, it immediately calls the built-in PID parameter tuning algorithm to generate a new proportional coefficient and integral time constant, and sends it to the excitation regulator through the communication interface. The inverter analysis unit starts the timing function after the demagnetization command is issued, and synchronously monitors the exponential decay curve of the generator-end voltage. When the time required for the voltage value to drop to 36.8% of the initial value exceeds the safety limit, it sends an inverter failure code and an audible and visual alarm signal to the central control unit. The excitation analysis unit collects the generator-end voltage rise rate once per second during the system startup phase. When it detects that the excitation time is extended by more than a set percentage compared to the standard operating conditions, it automatically generates an optimization strategy file containing suggestions for adjusting the given slope of the generator-end voltage.

[0068] In a preferred embodiment, the step response analysis submodule is integrated into the core control unit of the excitation regulator and performs the following steps: real-time acquisition of the generator output voltage / current waveform data, and extraction of the step response characteristic parameters through a digital filtering algorithm; comparison of the measured overshoot, number of oscillations and adjustment time with a preset threshold library: when any one exceeds the threshold, the PID parameter tuning process is triggered; a PID adjustment instruction is generated based on the fuzzy control algorithm: when the overshoot exceeds the limit, the differential coefficient is adjusted first to suppress the overshoot; when the number of oscillations exceeds the limit, the integral coefficient and the differential coefficient are corrected synchronously; when the adjustment time exceeds the limit, the proportional coefficient is optimized to speed up the response speed.

[0069] In a preferred embodiment, the inverter analysis unit implements a five-level warning mechanism through the demagnetization circuit monitoring interface: calculates the demagnetization time; and sets the five-level warning thresholds of the demagnetization time T: T1∈[6,7)s, T2∈[7,8)s, T3∈[8,9)s, T4∈[9,10)s, T5≥10s; when T reaches Ti (i=1-4), the i-th level warning signal is displayed through the monitoring interface; when T≥T5, the inverter failure protection action is triggered, and the shutdown command is immediately output and the demagnetization resistor is driven to be forced into operation.

[0070] In a preferred embodiment, the excitation analysis unit implements excitation process protection through voltage / current dual closed-loop detection: the real-time amplitudes of the generator terminal voltage and the excitation current in the excitation stage are sampled, and a five-level judgment threshold for the excitation time Tq is set: Tq1∈[15,16)s, Tq2∈[16,17)s, Tq3∈[17,18)s, Tq4∈[18,19)s, Tq5≥20s; when Tq reaches Tqi (i=1-4), the i-th level alarm processing is executed, including starting the soft start grade setting optimization and implementing the generator terminal voltage given slope adjustment; when Tq≥Tq5, the following dual judgment is performed: whether the generator terminal voltage amplitude is lower than 10% of the rated generator terminal voltage and the excitation current amplitude is lower than 6% of the rated excitation current at the same time; if so, the excitation failure protection is triggered and the following operations are executed: locking the excitation circuit contactor; generating a fault log containing the actual measured values of the generator terminal voltage amplitude and the excitation current amplitude; and uploading the fault signal to the monitoring system through the communication interface.

[0071] In a preferred embodiment, when the rated generator terminal voltage starting mode is preset, in addition to meeting the aforementioned starting state optimization conditions, the following additional constraints are enforced: voltage overshoot is strictly limited to no more than 5% of the rated voltage, the number of oscillations is strictly limited to no more than 3, and the settling time is strictly limited to no more than 15 seconds. If any parameter is detected as out of limit, the following actions are executed: optimizing the soft start's hierarchical setting strategy, dynamically modifying PID parameters, and correcting the generator terminal voltage reference rate. Among them, the hierarchical setting strategy of the soft start is optimized, and the preset values are divided into three levels: Preset30∈[0.1, Preset90], Preset90∈[Preset30, Preset100], and Preset100∈[Preset90, 1.1]. The adjustment step of each level is not less than 5% of the rated generator-end voltage. In the interval [Preset30, Preset90], the generator-end voltage given rise rate setting value parameter Vset is called as the soft start given rise rate, and in the interval [Preset90, Preset100], Vset / 2 is called as the soft start given rise rate; the PID parameters are dynamically corrected, where the proportional coefficient adjustment range is [0.8, 1.2] Kp0, and the integral coefficient adjustment range is [0.5, 1.5] Ki0 (Kp0 and Ki0 are the initial setting values of the proportional coefficient and the initial setting values of the integral coefficient, respectively); the rising rate of the voltage set value is adjusted, and the rate adjustment range is Vset∈[0.5, 1.0] V0 / s (V0 is the rated generator-end voltage).

[0072] Traditional excitation system fault diagnosis uses a single analysis module to handle multiple operating states, resulting in inadequate fault feature extraction and delayed response measures. In existing technologies, operating and maintenance personnel must manually switch between different analysis modes, making it impossible to implement state-adaptive fault handling. This excitation system constructs dedicated analysis units with functional decoupling, forming independent and parallel processing channels for monitoring and handling three typical operating conditions: step response, inverter demagnetization, and system excitation. This effectively eliminates processing delays caused by multi-state mixed analysis and achieves precise matching of fault diagnosis and response measures under different operating states. When the system is in the step response state, PID parameter adjustment instructions are generated and issued in a very short time. In the event of inverter demagnetization anomalies, the alarm signal triggering time is significantly shortened compared to traditional methods. The generation accuracy of optimization strategies for the excitation process is improved. Each analysis unit allocates independent computing resources, avoiding resource competition during multi-tasking, ensuring extremely high system reliability indicators.

[0073] In a preferred embodiment, the data acquisition module includes a file parsing unit and a cursor control unit. The file parsing unit supports COMTRADE standard file parsing and is compatible with both ASCII and binary formats. The cursor control unit implements dual cursor linkage selection and dynamic interval parameter calculation, specifically performing calculations on data points within an interval.

[0074] Among them, COMTRADE standard file parsing refers to the parsing method of the power system transient data exchange format in accordance with the IEEE standard. It can be implemented based on file header information identification and data block decoding, and by establishing a format mapping table to match different field definitions. Compatibility with ASCII and binary formats refers to the ability to read text and binary recording files. It can be implemented by branch processing logic, performing character stream parsing for ASCII files and byte stream parsing for binary files. Dual cursor linkage selection refers to the time axis synchronization control mechanism of two independent marking points. It can be implemented by graphical interface coordinate mapping and data index bidirectional binding to automatically update the corresponding data segment index when the cursor moves. Dynamic calculation of interval parameters refers to real-time calculation based on data points in the selected time interval. It can be implemented by using a sliding window algorithm combined with cache optimization technology to trigger feature value extraction and parameter update when the cursor position changes.

[0075] Specifically, the file parsing unit determines the type of recording file through the built-in format recognition module and calls the corresponding decoder to extract data. ASCII format files are read line by line and then split into fields, while binary files intercept valid data based on predefined byte offsets. The cursor control unit establishes two draggable markers in the visual interface and converts the screen coordinates into data index positions on the time axis. When any cursor moves, the data buffer loads the corresponding waveform segment according to the current index range, and at the same time triggers the eigenvalue calculation module to perform real-time statistics on the maximum value and starting value in the interval. The interval parameter dynamic calculation module continuously updates dynamic characteristic parameters such as overshoot and adjustment time based on the latest acquired eigenvalues, and pushes the results to the display terminal.

[0076] Traditional waveform analysis tools only support a single file format and require manual conversion of the data format before importing, resulting in a cumbersome analysis process. Existing waveform analysis interfaces mostly use a single cursor positioning method, which cannot accurately capture the key intervals of the dynamic process and relies on manual estimation of characteristic time points. This excitation system eliminates the data preprocessing link through dual-format compatible analysis, and the dual-cursor linkage mechanism achieves millisecond-level precision interval capture, combined with real-time parameter calculation function to replace manual characteristic value measurement. This enables direct analysis of multi-format waveform files and reduces efficiency losses caused by manual conversion operations. The dual-cursor linkage mechanism ensures the precise positioning of the time interval required for dynamic characteristics analysis and eliminates subjective errors in manual selection. The interval parameter dynamic calculation function outputs characteristic values in real time during cursor movement, shortening the time delay from waveform selection to parameter acquisition.

[0077] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Therefore, any modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of the technical solution of the present invention.

Claims

1. A method for analyzing dynamic characteristics of an excitation system, characterized in that: The following steps are involved: S100: Read the recorded file data of the excitation system, extract the original data including analog data and time stamp information, classify and organize the original data and generate configuration files and data files that comply with the COMTRADE standard; S200: Calculating dynamic characteristic parameters of the classified and organized raw data, wherein the dynamic characteristic parameters include quantitative calculation of overshoot, adjustment time, number of oscillations, demagnetization time, excitation start time, and overexcitation reverse time limit; S300: Establishing preset thresholds for dynamic characteristic parameters, comparing calculation results with the preset thresholds, and generating PID parameter adjustment instructions, inverter failure alarm signals, or excitation control optimization strategies based on the type of over-limit parameters.

2. The method for analyzing dynamic characteristics of an excitation system according to claim 1, characterized in that: Step S100 also includes, specifically: The configuration files and data files of the COMTRADE standard are processed for waveform visualization and a dual cursor time and interval selection mechanism is established; The dual cursor time and interval selection mechanism can achieve time axis synchronous positioning, arbitrary data interval interception and dynamic waveform zoom display.

3. A method for analyzing dynamic characteristics of an excitation system according to claim 1 or 2, characterized in that: The dynamic characteristic parameter calculation is achieved through the following sub-steps: Constructing a dynamic characteristic analysis database, the database receiving a time interval data array selected by a double cursor and executing acquisition of characteristic values in the time interval data array selected by the double cursor; Wherein, the characteristic value includes a maximum value, an initial value and a steady-state value; The overshoot, adjustment time and oscillation number are calculated based on the characteristic value.

4. The method for analyzing dynamic characteristics of an excitation system according to claim 3, characterized in that: The demagnetization time calculation formula is: T = T2 - T1; Among them, T1 is the time when the inverter starts, and T2 is the time when the voltage decays to 36.8% of the inverter start.

5. The method for analyzing dynamic characteristics of an excitation system according to claim 3, characterized in that: The calculation formula for the overexcitation inverse time limit is: Where: I L is the measured excitation current; I FEL is the peak current multiple; T q Allow time for the top value; I LEL It is the long-term allowable current reference value.

6. The method for analyzing dynamic characteristics of an excitation system according to claim 1, characterized in that: The optimization strategy is specifically as follows: When in the step response state, the overshoot threshold, oscillation number threshold and adjustment time threshold are input. When the measured overshoot is greater than the overshoot threshold, the oscillation number is greater than the oscillation number threshold, or the adjustment time is greater than the adjustment time threshold, a PID parameter adjustment instruction is generated. When in the inverter state, the demagnetization time sets five warning thresholds. The first four levels trigger graded warning signals, and the fifth level generates an inverter failure alarm and executes a shutdown command. When in the excitation starting state, five levels of warning thresholds are set. The first four levels trigger graded warning signals, generating an optimization strategy for the soft start graded set value and the given slope of the machine-end voltage. When reaching the fifth level, if the machine-end voltage amplitude is lower than 10% of the rated machine-end voltage and the excitation current amplitude is lower than 6% of the rated excitation current, an excitation failure alarm is issued and the excitation circuit self-check procedure is executed.

7. A dynamic characteristics analysis system for an excitation system, characterized in that: The method for analyzing dynamic characteristics of an excitation system according to any one of claims 1 to 6, characterized in that it comprises: The data acquisition module is used to read the recorded file data of the excitation system, extract the raw data containing analog data and timestamp information, classify and organize the raw data, and generate configuration files and data files that comply with the COMTRADE standard; The data calculation module is used to calculate the dynamic characteristic parameters of the classified and sorted raw data; The analysis module is used to establish preset thresholds for dynamic characteristic parameters, compare the calculation results with the preset thresholds, and generate PID parameter adjustment instructions, inverter failure alarm signals, or excitation control optimization strategies based on the type of out-of-limit parameters.

8. The excitation system dynamic characteristics analysis system according to claim 7, characterized in that: The data acquisition module is also used to perform waveform reproduction processing on the configuration files and data files of the COMTRADE standard, and establish a dual cursor time and interval selection mechanism.

9. The excitation system dynamic characteristics analysis system according to claim 7, characterized in that: The analysis module includes: The step response analysis submodule is used to generate PID parameter adjustment instructions when any of the measured overshoot, oscillation number or adjustment time exceeds the preset threshold; The inverter analysis unit is used to execute the output of the first four levels of graded warning signals or the fifth level of inverter failure alarm and shutdown command based on the comparison results of the actual demagnetization time value and the five-level preset warning threshold; The excitation analysis unit is used to execute the output of the first four levels of graded warning signals, the adjustment strategy of the soft starting grade setting value and the given slope of the terminal voltage, or the fifth level excitation failure alarm and the excitation circuit self-check instruction according to the comparison results of the excitation process parameters and the five-level warning threshold; The fifth-level excitation failure judgment condition is: the amplitude of the terminal voltage is lower than 10% of the rated terminal voltage and the amplitude of the excitation current is lower than 6% of the rated excitation current for a continuous period.

10. The excitation system dynamic characteristics analysis system according to claim 7, characterized in that: The data acquisition module includes: File parsing unit, supports COMTRADE standard file parsing, compatible with ASCII and binary formats; The cursor control unit realizes the dual cursor linkage selection and the dynamic calculation of the interval parameters. The dynamic calculation of the interval parameters is specifically to intercept the data points within the interval for calculation.

Images