A method and system for constructing an impedance model of a new energy flexible low-frequency sending-out system

Abstract

This invention relates to the field of electrical signal processing technology, and more particularly to a method and system for constructing an impedance model for a new energy flexible low-frequency transmission system. The method includes the following steps: recording the real-time timing of voltage and current of the converter arm under different operating conditions; extracting the time interval between voltage peak and current peak values, the rate of change of voltage waveform slope, and the oscillation period of current waveform, and labeling them with corresponding operating condition tags to form a dynamic feature library for the converter arm; identifying the voltage and current waveforms of the converter arm, and simultaneously recording the line resistance and line reactance between converter stations. This invention improves the matching efficiency of the impedance model by integrating the dynamic features of the converter arm, constructing a dynamic feature library of the converter arm, and using high-frequency correction coordination to achieve adaptive optimization of the arm parameters and the construction of impedance models in high and low frequency bands.

Description

Technical Field

[0001] This invention relates to the field of electrical signal processing technology, and in particular to a method and system for constructing an impedance model for a flexible low-frequency transmission system for new energy sources. Background Technology

[0002] The core of the flexible low-frequency power transmission system for new energy consists of the multi-level matrix converter bridge arm and the line between the converter station. It is a key piece of equipment for grid-connected transmission of new energy power. Its impedance model, as the core foundation for system optimization design and stable control, needs to cover the dynamic characteristics of the bridge arm, line parameters, and the adaptation requirements of different operating conditions and high and low frequency bands. However, traditional impedance model construction methods have obvious shortcomings: only a few static features are extracted from the bridge arm, and a binding system of multi-dimensional dynamic features and operating condition labels is not formed, resulting in insufficient parameter matching efficiency and accuracy; the bridge arm parameters adopt fixed configurations, lacking real-time waveform preprocessing and dynamic calibration mechanisms, making it difficult to adapt to operating condition fluctuations; the determination of line connection relationships lacks accurate topology mapping, and resistance and reactance are mostly measured in a single measurement, resulting in poor accuracy of parameter access; moreover, the model only focuses on low-frequency characteristics, ignoring the influence of high-frequency oscillations, and does not set up a high-frequency correction link, failing to comprehensively characterize the impedance characteristics of the entire bridge arm-line link, thus restricting the completeness and accuracy of the impedance model. Summary of the Invention

[0003] Therefore, it is necessary to provide a method and system for constructing an impedance model of a new energy flexible low-frequency transmission system to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above objectives, a method for constructing an impedance model for a new energy flexible low-frequency transmission system is provided. The new energy flexible low-frequency transmission system includes a multi-level matrix converter bridge arm and inter-converter station lines. The method includes the following steps: Step S1: Record the real-time timing of voltage and current of the converter arm under different operating conditions; extract the time interval between voltage peak and current peak, the rate of change of voltage waveform slope and current waveform oscillation period, and label them with corresponding operating condition tags to form a dynamic feature library of the arm. Step S2: Identify the voltage and current waveforms of the converter bridge arm, and simultaneously record the line resistance and line reactance between converter stations; Step S3: Extract the voltage and current change features from the voltage and current waveforms, match the same-label features in the bridge arm dynamic feature library, and dynamically calibrate the bridge arm parameters in the preset impedance parameter framework. Step S4: Determine the connection relationship between the lines between converter stations, and connect the line resistance and line reactance to the corresponding connection nodes in the calibrated impedance parameter framework; Step S5: Analyze the high-frequency waveform timing of the output of the calibrated impedance parameter framework, identify the period of the repetitive oscillation signal, calculate the compensation coefficient accordingly, and perform high-frequency correction on the impedance parameter framework to obtain the impedance model of the new energy flexible low-frequency transmission system.

[0005] This invention also provides an impedance model construction system for a new energy flexible low-frequency transmission system, used to execute the above-described impedance model construction method for a new energy flexible low-frequency transmission system. The impedance model construction system for the new energy flexible low-frequency transmission system includes: The bridge arm feature library construction module is used to record the real-time timing of voltage and current of converter bridge arms under different operating conditions; extract the time interval between voltage peak and current peak, the rate of change of voltage waveform slope and current waveform oscillation period, and label them with corresponding operating condition tags to form a bridge arm dynamic feature library; The bridge arm and line parameter acquisition module is used to identify the voltage waveform and current waveform of the converter bridge arm, and simultaneously record the line resistance and line reactance between converter stations; The bridge arm parameter calibration module is used to extract the voltage and current change characteristics in the voltage waveform and current waveform, match the same label features in the bridge arm dynamic feature library, and dynamically calibrate the bridge arm parameters in the preset impedance parameter framework. The line parameter access module is used to determine the connection relationship of the lines between converter stations and to connect the line resistance and line reactance to the corresponding connection nodes of the calibrated impedance parameter framework. The high-frequency correction and model building module is used to analyze the high-frequency waveform timing of the output of the calibrated impedance parameter framework, identify the period of the repetitive oscillation signal, calculate the compensation coefficient accordingly, and perform high-frequency correction on the impedance parameter framework to obtain the impedance model of the new energy flexible low-frequency transmission system.

[0006] The beneficial effects of this invention are as follows: I. The multi-dimensional feature and operating condition label binding design of the bridge arm dynamic feature library enables accurate mapping between operating conditions and bridge arm features. The feature library covers three core dynamic features: the time interval between peak voltage and peak current, the rate of change of voltage waveform slope, and the oscillation period of current waveform. Each feature group is labeled with a corresponding operating condition label (such as load level and operating mode). This ensures a comprehensive representation of the bridge arm's operating status and allows the subsequent matching process to directly lock onto valid feature groups under the same operating condition, avoiding interference from invalid matching across operating conditions. This significantly improves the efficiency and initial accuracy of bridge arm parameter matching.

[0007] II. The dynamic calibration mechanism combines real-time waveform preprocessing with step-by-step parameter adjustment to achieve precise adaptive optimization of bridge arm parameters. Baseline correction eliminates overall waveform offset and interpolation replaces abnormal pulses, ensuring the reliability of pressure-current change feature extraction. After matching features, the adjustment trend of resistance and inductance is determined according to the direction of difference. The parameters are adjusted step-by-step at a fixed ratio and locked through waveform deviation verification (peak deviation ≤2%, period deviation ≤1ms). This ensures that the equivalent resistance and equivalent inductance always dynamically match the real-time operating state, solving the problem that traditional fixed parameters are difficult to adapt to system operating condition fluctuations.

[0008] III. A collaborative design for precise line parameter access and high-frequency correction constructs a complete impedance model covering the entire bridge arm-line link and taking into account both high and low frequency characteristics. Line connection nodes are clearly identified through terminal identification and topology table comparison. Line resistance and reactance, measured multiple times and averaged (measurement accuracy 0.5%FS), are accessed to ensure the accuracy of line parameter access. For high-frequency waveforms, repetitive oscillation periods are identified and compensation coefficients are calculated to correct the parameter framework. This overcomes the shortcomings of traditional models that only focus on low-frequency characteristics and ignore the impact of high-frequency oscillations. The final model accurately reflects the core impedance characteristics of the system's low-frequency operation and adapts to the dynamic response of the high-frequency band, comprehensively improving the completeness and accuracy of the impedance characteristic representation of the new energy flexible low-frequency transmission system. Attached Figure Description

[0009] Figure 1 A flowchart illustrating the steps involved in constructing an impedance model for a flexible low-frequency transmission system for new energy sources. Figure 2 This is a timing diagram of high-frequency voltage and current waveforms; Figure 3 Construct a system interface diagram for the impedance model of a flexible low-frequency transmission system for new energy. The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0010] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0011] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0012] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0013] To achieve the above objectives, please refer to Figures 1 to 3 A method for constructing an impedance model for a new energy flexible low-frequency transmission system, the new energy flexible low-frequency transmission system including a multi-level matrix converter bridge arm and inter-converter station lines, the method comprising the following steps: Preferably, step S1: record the real-time timing of voltage and current of converter bridge arms under different operating conditions; extract the time interval between voltage peak and current peak, the rate of change of voltage waveform slope and current waveform oscillation period, and label them with corresponding operating condition tags to form a bridge arm dynamic feature library; Optionally, the real-time timing of recording the voltage and current of the converter bridge arm in step S1 includes: For each working condition, the bridge arm is switched from the current working condition to the target working condition. After the output waveform has no obvious jump for three consecutive low-frequency cycles, timing recording is started. During the recording process, the stable operating condition within each low-frequency cycle is simultaneously marked. After recording is complete, the real-time timing data of the voltage and current of the converter bridge arm will be segmented according to the marked stable period, and only the timing data of the stable period will be retained.

[0014] In this embodiment, for three operating conditions—rated load output current of 1000A, 50% load output current of 500A, and light load output current of 200A—the operating condition switching module of the new energy flexible low frequency transmission system supports continuous adjustment of the trigger angle from 0° to 90° and linear adjustment of the modulation ratio from 0.1 to 1.0. This module switches the converter bridge arm from the current operating condition parameters to the target operating condition parameters.

[0015] In another embodiment, after the switching is completed, a Hall voltage sensor is used to acquire the real-time timing of the bridge arm voltage. This sensor has a measurement range of 0-3000V, an accuracy of 0.1%FS, and a bandwidth of 20kHz. A Rogowski coil current sensor is used to acquire the real-time timing of the bridge arm current. This sensor has a measurement range of 0-2000A, an accuracy of 0.2%FS, and a bandwidth of 50kHz. The sensor output signals are processed by a 16-bit resolution data acquisition card with a sampling rate of 100kHz and an input impedance of 1MΩ. After converting the signals into digital signals, they are transmitted to the real-time data recording unit, which has a storage delay of ≤1ms.

[0016] It should be noted that the low-frequency period is set to 20ms, corresponding to a low frequency of 50Hz. The waveform analysis module built into the data recording unit monitors the voltage peak and current peak within three consecutive low-frequency periods, i.e., 60ms, in real time. When the voltage peak deviation between adjacent periods is ≤2% of the rated voltage peak (3000V) and the current peak deviation is ≤2% of the rated current peak (1000A), it is determined that there is no obvious jump, and timing recording is started.

[0017] In another embodiment, the data recording unit generates a timestamp in μs based on the synchronization accuracy of the GPS timing module of 1 μs. After each low-frequency cycle of 20ms, a stable operating condition flag field is added to the data frame. This field is assigned a value of stable or unstable according to the real-time monitoring results. Stable corresponds to meeting the peak deviation condition, and unstable corresponds to not meeting the peak deviation condition.

[0018] After recording, the data is stored in a CSV file containing fields such as timestamp (μs), arm voltage (V), arm current (A), operating condition label, and stable state marker. This file is read using a time series segmentation tool, which supports batch processing based on field filtering. Data is filtered by the condition that the stable state marker equals stable, extracting consecutively marked stable time series segments and deleting those marked as unstable. The final retained stable period time series data must contain at least 5 consecutive stable markers corresponding to 100ms of time series information, serving as the basis for subsequent feature extraction.

[0019] Optionally, the formation of the bridge arm dynamic feature library in step S1 specifically involves: Extract the voltage peak value and current peak value within 5 consecutive low-frequency cycles, calculate the time interval between the two within each cycle, and take the average time interval of the 5 cycles as the time interval feature of the time series. Extract the slope change value of the voltage waveform at each rising edge, and take the average of the slope change values ​​of three consecutive rising edges as the slope change rate feature of the voltage waveform. Identify continuous complete oscillations in the current waveform, calculate the period values ​​of three adjacent oscillations, and take the average value as the oscillation period characteristic of the current waveform; The time interval features, voltage waveform slope change rate features, and current waveform oscillation period features are bound to the corresponding operating condition labels and stored in the bridge arm dynamic feature library according to the structure of operating condition label-feature group-original time sequence index.

[0020] In one embodiment, a time sequence segment of five consecutive low-frequency cycles is extracted from the retained stable periodic time sequence data, with each cycle being 20ms and a total duration of 100ms. The peak detection module scans the time sequence segment cycle by cycle, capturing the maximum value of the voltage waveform within each cycle as the voltage peak value and the maximum value of the current waveform as the current peak value, and synchronously recording the timestamp (in μs) corresponding to each peak value, with a module detection delay ≤1ms. The difference between the voltage peak timestamp and the current peak timestamp within each cycle is calculated to obtain five time interval values ​​(in μs). These five values ​​are summed and divided by 5 to obtain the time interval feature value.

[0021] In another embodiment, the voltage waveform amplitude change is monitored. When the voltage gradually rises from 300V to 2700V, this interval is determined to be a rising edge. For each rising edge, the voltage values ​​(in V) at the start and end points and the corresponding time values ​​(in ms) are read. The ratio of the voltage difference to the time difference is calculated to obtain the slope change value of that rising edge (in V / ms). Three consecutive rising edges are selected, and their slope change values ​​are summed and divided by 3 to obtain the characteristic value of the voltage waveform slope change rate.

[0022] In another embodiment, the fluctuation of the current waveform is continuously monitored. When the amplitude of the current fluctuation exceeds 50A and exhibits a continuous pattern of two peaks and one trough, it is determined to be a complete oscillation. The start and end timestamps (in μs) of each complete oscillation are recorded, and the difference between the start timestamps of two adjacent oscillations is calculated as a single oscillation period value (in μs). Three consecutive adjacent oscillation period values ​​are selected, summed, and divided by 3 to obtain the characteristic value of the current waveform oscillation period.

[0023] The time interval feature value, voltage waveform slope change rate feature value, and current waveform oscillation period feature value are bound to the corresponding operating condition labels (rated load 1000A, 50% load 500A, light load 200A) to form feature groups. Through the data storage unit, the feature groups are stored in the bridge arm dynamic feature library according to the structure of "operating condition label - feature group - original time series index". The original time series index is the timestamp range of the time series data on which the feature extraction is based in the CSV file (format: "start timestamp μs - end timestamp μs").

[0024] Preferably, step S2: identify the voltage waveform and current waveform of the converter bridge arm, and simultaneously record the line resistance and line reactance between converter stations; In this embodiment, a Hall voltage sensor is used to acquire the analog voltage signal of the converter bridge arm. The sensor has a measurement range of 0-3000V and an accuracy of 0.1%FS. The output analog signal is filtered by a low-pass filter (cutoff frequency 1kHz, attenuation slope 40dB / decade) to remove high-frequency interference, and then connected to a 16-bit resolution data acquisition card. The sampling rate of the acquisition card is set to 100kHz and the input impedance is 1MΩ. The analog signal is converted into a digital signal and transmitted to the waveform recognition unit.

[0025] It should be noted that the waveform recognition unit monitors the amplitude fluctuation of the voltage digital signal in real time, and sets the effective voltage fluctuation range to 300V to 2700V. At the same time, the signal repetition period is counted by the period detection circuit. When the fluctuation range of 5 consecutive periods is stable within the effective range and the period deviation is ≤1ms (target low frequency period 20ms), the voltage waveform recognition is determined to be complete, and the voltage value (unit V) corresponding to each timestamp (unit μs) is recorded synchronously.

[0026] In one embodiment, a Rogowski coil current sensor is used to acquire the analog current signal of the converter arm. The sensor has a measurement range of 0-2000A and an accuracy of 0.2%FS. The output signal is transmitted to the signal conditioning module via a shielded cable. The module optimizes the signal amplitude through a differential amplifier circuit (amplification factor of 100 times, input offset voltage ≤5μV) and then connects to the same data acquisition card to convert it into a digital signal.

[0027] It should be noted that the waveform recognition unit sets the effective fluctuation range for the current digital signal to be 20A to 1800A. It verifies the periodicity of the signal through the same period detection logic as the voltage waveform. When five consecutive cycles meet the fluctuation range requirements and the period deviation is ≤1ms, the current waveform recognition is completed, and the current value (unit: A) corresponding to each timestamp (unit: μs) is recorded.

[0028] In another embodiment, the two ends of the line between converter stations are connected to the input terminals of a line impedance tester. The tester measures resistance from 0-10Ω and reactance from 0-100mH, with a measurement accuracy of 0.5%FS. A four-wire measurement method is used to eliminate the influence of wiring resistance. The tester has a built-in GPS timing module (synchronization accuracy 1μs), and clock calibration is achieved with the data acquisition card through a time synchronization protocol to ensure that the measurement timing sequence is consistent with the waveform acquisition timing sequence. The measurement interval is set to 10ms.

[0029] It should be noted that the tester collects the resistance (in Ω) and reactance (in mH) of the line in real time. After each measurement is completed, the measurement data and the corresponding timestamp (in μs) are packaged and transmitted to the data recording unit. The data recording unit aligns the voltage waveform data, current waveform data, line resistance data and line reactance data according to the timestamp to form a standardized dataset containing the fields of "timestamp μs, bridge arm voltage V, bridge arm current A, line resistance Ω, and line reactance mH".

[0030] Preferably, step S3: extract the voltage and current change features from the voltage waveform and current waveform, match the same label features in the bridge arm dynamic feature library, and dynamically calibrate the bridge arm parameters in the preset impedance parameter framework; Optionally, before extracting the voltage and current change features from the voltage and current waveforms in step S3, the following steps are included: Real-time acquisition of voltage and current waveforms, and baseline correction, using the average amplitude of the waveform as a reference to eliminate overall waveform offset; After correction, the amplitude of each local segment of the voltage waveform and current waveform is compared with the overall average amplitude. When the amplitude of a local segment exceeds a preset multiple of the overall average amplitude, the segment is determined to be an abnormal pulse. Replace the local waveform containing the abnormal pulse with the interpolation of the adjacent normal waveform, and then extract the voltage and current change characteristics from the voltage and current waveforms.

[0031] In this embodiment, the data preprocessing unit reads voltage waveform data and current waveform data for three consecutive low-frequency cycles (total duration 60ms), calculates the average voltage amplitude and average current amplitude of all sampling points in the interval, subtracts the average voltage amplitude from the original voltage waveform value at each moment, and subtracts the average current amplitude from the original current waveform value at each moment to complete baseline correction and eliminate overall waveform offset.

[0032] In another embodiment, after correction, the voltage waveform and current waveform are divided into independent local segments with a duration of 1ms. The maximum amplitude of each segment is calculated and compared with the average amplitude of the corresponding waveform. When the maximum amplitude of a local segment exceeds 3 times the average amplitude of the corresponding waveform, it is determined to be an abnormal pulse, and the start and end timestamps of the segment are recorded synchronously.

[0033] It should be noted that, for the identified abnormal pulse segments, the last sampled value of the preceding 1ms normal segment and the first sampled value of the following 1ms normal segment are extracted. The interpolation nodes are equally divided based on the time interval of the abnormal pulse segments, and the amplitude data of each node is obtained through linear calculation. The original data of the abnormal pulse segments is completely replaced with the interpolated data to form voltage waveform data and current waveform data without offset and abnormal pulses, which provides a basis for subsequent extraction of voltage and current change features.

[0034] Optionally, the extraction of voltage-current variation features from the voltage and current waveforms in step S3 specifically involves: The real-time voltage and current waveforms are divided into continuous analysis units according to the low-frequency cycle, and each analysis unit contains one complete cycle. For each analysis unit, mark the voltage zero-crossing point and the current zero-crossing point. Starting from the zero-crossing point, extract the position and time of the voltage peak and the current peak respectively. Calculate the time difference between the peak voltage and peak current within the same analysis unit, and use it as the time interval characteristic of that unit; Calculate the slope change of the voltage and current waveforms during the rising phase, and take the average value of this change as the characteristic of voltage-current change.

[0035] In this embodiment, the data processing unit reads the preprocessed voltage and current waveform data and divides the waveform data into continuous analysis units with a low-frequency period of 20ms. Each analysis unit contains complete cycle data from one voltage zero-crossing point to the next voltage zero-crossing point in the same direction, synchronously associated with the current waveform data of the corresponding cycle. Within each analysis unit, the zero-crossing detection circuit monitors the voltage waveform amplitude change in real time. When the voltage value jumps from -5mV to +5mV, it is marked as a voltage zero-crossing point, and the timestamp of this moment (in μs) is recorded. For the current waveform, when the current value jumps from -10mA to +10mA, it is marked as a current zero-crossing point, and the corresponding timestamp (in μs) is recorded.

[0036] In one embodiment, starting from the voltage zero-crossing timestamp, the peak detection circuit scans subsequent voltage waveform data, captures the sampling point with the largest amplitude as the voltage peak value, and records the timestamp (in μs) corresponding to its position. Similarly, starting from the current zero-crossing timestamp, it scans subsequent current waveform data, captures the sampling point with the largest amplitude as the current peak value, and records its timestamp (in μs). The difference between the voltage peak timestamp and the current peak timestamp within the same analysis unit is calculated to obtain the time interval characteristic (in μs) of that unit.

[0037] In another embodiment, for the rising segment of the voltage waveform, the interval from the voltage zero-crossing point to the voltage peak is taken as the rising segment. The difference (in V) between the initial voltage value (zero-crossing voltage value 0V) and the voltage peak value within this interval is calculated and divided by the time difference (in ms) of this interval to obtain the voltage slope change (in V / ms). For the current waveform, the interval from the current zero-crossing point to the current peak value is taken as the rising segment. The difference (in A) between the initial current value (zero-crossing current value 0A) and the current peak value is calculated and divided by the time difference (in ms) of this interval to obtain the current slope change (in A / ms). The voltage slope change and the current slope change are summed and divided by 2 to obtain the voltage-current change characteristics of the analysis unit.

[0038] Optionally, in step S3, matching features with the same label in the bridge arm dynamic feature library includes: Retrieve all feature groups from the bridge boom dynamic feature library that are completely consistent with the current working condition label; For the called feature groups, they are compared in the order of time interval, slope change rate, and oscillation period, and feature groups in which each dimension is within the preset matching range are retained. From the retained feature groups, calculate the comprehensive deviation value between the current feature and each group of features, and select the feature with the smallest comprehensive deviation value as the matching result.

[0039] In this embodiment, the feature matching unit inputs the current operating condition label (such as rated load 1000A, 50% load 500A, light load 200A) through the query interface of the bridge arm dynamic feature library, and calls all feature groups in the library that are completely consistent with the label. Each feature group includes time interval feature value (unit μs), voltage waveform slope change rate feature value (unit V / ms), and current waveform oscillation period feature value (unit μs). The query response time is ≤100ms.

[0040] In one embodiment, the feature comparison circuit performs a dimension-by-dimensional comparison of the called feature groups in the order of time interval, slope change rate, and oscillation period. The preset matching ranges are time interval ±50μs, slope change rate ±0.5V / ms, and oscillation period ±10μs, respectively. Only feature groups whose feature values ​​in all three dimensions fall within the corresponding range are retained.

[0041] In another embodiment, the deviation calculation unit calculates the absolute difference between the current feature's time interval value and the time interval feature values ​​of each group, the absolute difference between the current voltage slope change rate value and the corresponding value of each group, and the absolute difference between the current current oscillation period value and the corresponding value of each group for the retained feature groups. The three absolute differences are divided by the upper limit of the preset matching range of the corresponding dimension (50μs, 0.5V / ms, 10μs) and then summed to obtain the comprehensive deviation value of each group. The feature group with the smallest comprehensive deviation value is selected as the matching result, and the original time series index of the feature group is recorded.

[0042] Optionally, the bridge arm parameters in the preset impedance parameter framework for dynamic calibration in step S3 include: Retrieve the reference parameters of the current bridge arm from the preset impedance parameter framework, including equivalent resistance and equivalent inductance; The adjustment trend of resistance and inductance is determined based on the direction of the difference between the matching characteristics and the preset reference characteristics. Adjust the equivalent resistance according to the difference ratio. Each time it is adjusted, compare the resistance correlation characteristic deviation between the analog waveform output by the parameter frame and the real-time acquired waveform. Once the resistance deviation meets the requirements, adjust the equivalent inductance using the same logic until the inductance-related characteristic deviation also meets the requirements.

[0043] In this embodiment, the parameter call interface retrieves the reference parameters of the current bridge arm from the preset impedance parameter framework, including the equivalent resistance reference value of 2Ω and the equivalent inductance reference value of 50mH, with a call response time of ≤50ms.

[0044] In one embodiment, the deviation analysis circuit extracts the resistance-related feature (voltage peak decay rate, in % / ms) from the matching features and the resistance-related feature (reference voltage peak decay rate 0.5% / ms) from the preset reference features, calculates the difference between the two, and if the matching feature value is greater than the reference feature value, it is determined that the equivalent resistance needs to be increased; otherwise, it needs to be decreased. At the same time, it extracts the inductance-related feature (current rise time, in ms) from the matching features and the inductance-related feature (reference current rise time 10ms) from the reference features. If the matching feature value is greater than the reference feature value, it is determined that the equivalent inductance needs to be increased; otherwise, it needs to be decreased.

[0045] It should be noted that the equivalent resistance is adjusted at an initial 5% difference ratio. After each adjustment, the waveform comparison unit aligns the analog voltage waveform output from the impedance parameter framework with the real-time acquired voltage waveform along the time axis, calculates the absolute deviation of their resistance correlation characteristics (voltage peak attenuation rate), and stops the equivalent resistance adjustment and locks the current value when the deviation is ≤3%. Subsequently, the equivalent inductance is adjusted according to the same logic, with an initial adjustment ratio of 5%. After each adjustment, the absolute deviation of the inductance correlation characteristics (current rise time) between the analog current waveform and the real-time acquired current waveform is compared until the deviation is ≤5%, at which point the current equivalent inductance value is locked, completing the dynamic calibration of the bridge arm parameters.

[0046] Most importantly, if there are no features in the same label feature group that match the preset range, perform the following operations: Gradually expand the matching range for each dimension, and re-select feature groups with the same label after each expansion; If no matching feature group is found even after expanding to the maximum allowable range, the feature with the smallest overall deviation value in the same label feature group is selected, and the match is marked as a low-fit match. The number of parameter verifications is increased in subsequent calibration processes.

[0047] In this embodiment, when there is no feature group in the same label feature group that meets the preset matching range (time interval ±50μs, slope change rate ±0.5V / ms, oscillation period ±10μs), the feature screening module starts the range expansion mechanism, expanding the matching range of each dimension by 20% each time. That is, after the first expansion, the time interval is ±60μs, the slope change rate is ±0.6V / ms, and the oscillation period is ±12μs. After expansion, the same label feature group is called again for dimension-by-dimensional comparison, and feature groups in all dimensions within the new range are retained. Each expansion and screening process takes ≤200ms.

[0048] In another embodiment, if no matching feature group is found even after continuously expanding to the maximum allowable range (time interval ±150μs, slope change rate ±1.5V / ms, oscillation period ±30μs), the deviation calculation unit recalculates the comprehensive deviation value for all feature groups with the same label (the calculation method is consistent with the original logic), selects the feature group with the smallest comprehensive deviation value as the matching result, and adds a "low fit matching" mark (marker field value is "1") to the feature matching result; when entering the subsequent calibration process, the parameter verification module increases the number of parameter verifications (3 times) to 5 times, and each verification is judged according to the original deviation threshold (resistance-related feature deviation ≤3%, inductance-related feature deviation ≤5%). The calibration parameters are locked only after all verifications are passed.

[0049] Optionally, verifying the effect of parameter adjustments during calibration includes: The analog waveform output from the preset impedance parameter framework is precisely aligned with the real-time acquired waveform along the time axis. Calculate the peak deviation, period deviation, and slope deviation of the aligned waveforms respectively; When all three types of deviations are within the allowable range of the corresponding dimension, the calibration is deemed qualified, parameter adjustment is stopped, and the current parameter is locked.

[0050] In this embodiment, the timestamps (in μs) of the simulated voltage waveform and simulated current waveform output based on the calibration preset impedance parameter framework and the timestamps of the voltage waveform and current waveform acquired in real time are used to control the time axis error between the two within the range of ≤10μs, thereby achieving precise alignment.

[0051] In one embodiment, the simulated voltage peak value (in V) and the real-time voltage peak value (in V) within each low-frequency cycle (20 ms) of the aligned waveform are extracted. The ratio of the absolute difference between the two values ​​to the real-time voltage peak value is calculated to obtain the voltage peak value deviation (in %). The same operation is performed on the current waveform to obtain the current peak value deviation (in %). Simultaneously, the absolute difference between the duration of each cycle of the simulated waveform (in ms) and the corresponding cycle duration of the real-time waveform is calculated as the cycle deviation (in ms).

[0052] In another embodiment, for the rising segment of the voltage waveform (from zero crossing to peak value), the ratio of the absolute difference between the simulated slope (in V / ms) and the real-time slope (in V / ms) to the real-time slope is calculated to obtain the voltage slope deviation (in %). The same operation is performed on the rising segment of the current waveform to obtain the current slope deviation (in %). The allowable range for peak deviation is set to ≤2%, the allowable range for period deviation is ≤1ms, and the allowable range for slope deviation is ≤5%. When the voltage peak deviation, current peak deviation, period deviation, voltage slope deviation, and current slope deviation are all within their respective allowable ranges, parameter adjustment is stopped and the current equivalent resistance value (in Ω) and equivalent inductance value (in mH) are stored.

[0053] Of particular importance is that after completing the bridge arm parameter calibration in step S3, the following operations are performed: The calibrated parameters, matching features, and corresponding real-time waveform segments are packaged into a calibration record. Calibration records are stored in order of bridge arm number and calibration time. Each time a calibration is completed, the historical calibration log of the parameter framework is automatically updated. The log includes a comparison of the values ​​before and after parameter adjustment and the calibration pass / fail result.

[0054] In this embodiment, after the bridge arm parameter calibration is completed, the calibrated equivalent resistance value (unit Ω), equivalent inductance value (unit mH), matched time interval characteristic value (unit μs), voltage waveform slope change rate characteristic value (unit V / ms), current waveform oscillation period characteristic value (unit μs), and the corresponding real-time waveform segment of the calibration process (duration 60ms, containing 3 complete low-frequency cycles, timestamp range from calibration start timetamp to calibration end timetamp, unit μs) are packaged into a calibration record. The record file is named "bridge arm number (01-10) - calibration timestamp (accurate to μs based on GPS timing, format YYYY-MM-DDHH:MM:SS.μs)" and stored in the storage medium according to the bridge arm number and calibration time.

[0055] It should be noted that after each calibration, the system automatically extracts the equivalent resistance reference value (2Ω) and equivalent inductance reference value (50mH) before parameter adjustment and the adjusted calibration value, generates a numerical comparison table, and records the bridge arm number, calibration time, and calibration pass / fail judgment result (pass / fail, based on the judgment result of peak deviation ≤2%, period deviation ≤1ms, and slope deviation ≤5%). These are integrated into a calibration log entry and written to the historical calibration log of the impedance parameter framework. The logs are arranged in reverse order of calibration time, and each entry is assigned a unique 10-digit numerical identifier and stored in a log file in a specified path. The log file is named according to the date (YYYY-MM-DD), and a new file is generated daily while retaining historical files for archiving.

[0056] Preferably, step S4: determine the connection relationship of the lines between converter stations, and connect the line resistance and line reactance to the corresponding connection nodes of the calibrated impedance parameter framework; In this embodiment, the physical terminal identifiers (such as etched numbers or label information) at both ends of the line between converter stations are identified. The terminal identifiers are compared with a preset line topology table (including the mapping relationship between terminal identifiers and impedance parameter frame nodes) to confirm the corresponding nodes at both ends of the line in the impedance parameter frame. For example, if the terminal identifier at one end of the line is "T-01" and the other end is "T-02", and "T-01" corresponds to node "N1" and "T-02" corresponds to node "N2" in the topology table, then the line connection relationship is determined to be the connection link between nodes "N1" and "N2" in the frame.

[0057] In one embodiment, the line resistance and reactance are measured using a line impedance tester. The tester uses a four-wire connection (the current terminal is connected to both ends of the line to inject test current, and the voltage terminal is connected in parallel to both ends of the line to collect voltage). The test current is set to a low-frequency sine wave (for example, frequency 50Hz, amplitude 5A), and the measurement interval is set to a fixed value (for example, 10ms). Multiple measurements are taken continuously (for example, 3 times). The average value of the multiple measurements is calculated as the real-time value of the line resistance (for example, 3 measurements are 0.58Ω, 0.62Ω, and 0.60Ω, with an average of 0.60Ω) and the real-time value of the line reactance (for example, 3 measurements are 41mH, 43mH, and 42mH, with an average of 42mH).

[0058] In another embodiment, in the calibrated impedance parameter framework, the connection links of nodes "N1" and "N2" are located according to the determined connection relationship. The calculated line resistance value and line reactance value are written into the resistance parameter bit and reactance parameter bit corresponding to the link. After writing, the consistency between the parameter value and the measured mean is confirmed by the frame's built-in verification function (the deviation is within a preset range, for example, ≤0.5%). After confirming that there are no errors, the line connection relationship, the connected resistance value, reactance value and the original measurement data are stored in the frame's history record module in a fixed format (including timestamp and line identifier).

[0059] Please see Figure 2 Step S5: Analyze the high-frequency waveform timing of the output of the calibrated impedance parameter framework, identify the period of the repetitive oscillation signal, calculate the compensation coefficient accordingly, and perform high-frequency correction on the impedance parameter framework to obtain the impedance model of the new energy flexible low-frequency transmission system.

[0060] In this embodiment, high-frequency waveform timing data is output from the calibrated impedance parameter framework, including high-frequency voltage waveform timing (frequency range 2kHz-20kHz, unit V) and high-frequency current waveform timing (frequency range 2kHz-20kHz, unit A), with corresponding timestamps (unit μs) recorded synchronously. The high-frequency waveform timing is divided into analysis windows with a fixed duration of 1ms. Each window contains 1ms of time-continuous data (e.g., from the start timestamp t0 to t0+1000μs), ensuring that there are at least 5 complete potential oscillation cycles within each window (calculated based on a minimum high-frequency oscillation cycle of 200μs).

[0061] It should be noted that for each window, the oscillation signal is identified through a peak detection circuit: the peak determination threshold is set to 1.5 times the average amplitude of the waveform within the window. When the amplitude of a point on the waveform exceeds this threshold and the amplitudes of the two adjacent points are both less than that point, it is marked as a peak and its timestamp (in μs) is recorded. The difference in timestamps between adjacent peaks is calculated to obtain a single oscillation period value (in μs), and the frequency of occurrence of all period values ​​within the window is counted (e.g., 220μs appears 4 times and 230μs appears 6 times in a certain window). The frequency of period values ​​in all windows is summarized, and the period value with the highest total frequency is selected as the period of the repetitive oscillation signal (e.g., 230μs has a total frequency of 72 times).

[0062] In another embodiment, the compensation coefficients are calculated based on the repetitive oscillation signal period: the resistance compensation coefficient is (repetitive oscillation period ÷ reference oscillation period) × 0.8 (the reference oscillation period is preset to 200μs), and the inductance compensation coefficient is (repetitive oscillation period ÷ reference oscillation period) × 0.7. The resistance compensation coefficient is multiplied by the current equivalent resistance value in the frame to obtain the corrected equivalent resistance value (unit Ω), and the inductance compensation coefficient is multiplied by the current equivalent inductance value to obtain the corrected equivalent inductance value (unit mH). The corrected equivalent resistance value, equivalent inductance value, line resistance, and line reactance are re-entered into the impedance parameter frame. The frame is then started to output the corrected high-frequency waveform timing. The oscillation period deviation (corrected deviation ≤ 5μs) and peak deviation (corrected deviation ≤ 3%) of the waveform before and after correction are compared. After confirming that the deviation meets the requirements, the frame is the impedance model of the new energy flexible low-frequency transmission system.

[0063] Optionally, the timing sequence of the high-frequency waveform output from the impedance parameter framework analyzed and calibrated in step S5 is as follows: The high-frequency waveform is divided into multiple analysis windows with a fixed duration, and each window contains at least 5 complete potential oscillation cycles; For each window, identify the period value of all oscillation signals and count their frequency of occurrence; Summarize the statistical results of all windows and take the period value with the highest frequency as the period of the repetitive oscillation signal.

[0064] In this embodiment, high-frequency waveform timing data (frequency range 2kHz-20kHz) is retrieved from the calibrated impedance parameter framework, including high-frequency voltage waveform timing data (unit V) and high-frequency current waveform timing data (unit A), synchronously associated with corresponding timestamps (unit μs). A waveform segmentation tool is used to divide the high-frequency waveform timing data into multiple analysis windows with a fixed duration of 1ms. The timestamp range of each window is a continuous 1ms (e.g., from start timestamp t1 to end timestamp t1+1000μs), ensuring that each window contains at least 5 complete potential oscillation cycles (based on a minimum high-frequency oscillation cycle of 200μs, a 1ms window can accommodate 5 cycles). For each analysis window, the peak position of the oscillation signal is identified by a peak detection circuit: the peak determination threshold is set to 1.5 times the average amplitude of the waveform within the window. When the waveform amplitude exceeds this threshold and the amplitudes of the adjacent points are all less than this point, it is marked as a peak, and the timestamp (in μs) corresponding to each peak is recorded. The difference in timestamps between two adjacent peaks is calculated to obtain the period value (in μs) of a single oscillation signal. The same operation is performed on all peaks within the window to obtain multiple period values, and the frequency of occurrence of each period value is counted (e.g., 180μs occurs 3 times and 200μs occurs 5 times in a certain window). The frequency statistics of period values ​​of all analysis windows (for example, a total of 20 windows) are summarized to form a summary table containing "period value μs - total frequency of occurrence". The period value with the highest total frequency of occurrence in the summary table (e.g., 200μs, total frequency of occurrence 68 times) is selected as the period of the repetitive oscillation signal.

[0065] Please see Figure 3 The present invention also provides an impedance model construction system for a new energy flexible low-frequency transmission system, used to execute the above-described impedance model construction method for a new energy flexible low-frequency transmission system. The impedance model construction system for the new energy flexible low-frequency transmission system includes: The bridge arm feature library construction module is used to record the real-time timing of voltage and current of converter bridge arms under different operating conditions; extract the time interval between voltage peak and current peak, the rate of change of voltage waveform slope and current waveform oscillation period, and label them with corresponding operating condition tags to form a bridge arm dynamic feature library; The bridge arm and line parameter acquisition module is used to identify the voltage waveform and current waveform of the converter bridge arm, and simultaneously record the line resistance and line reactance between converter stations; The bridge arm parameter calibration module is used to extract the voltage and current change characteristics in the voltage waveform and current waveform, match the same label features in the bridge arm dynamic feature library, and dynamically calibrate the bridge arm parameters in the preset impedance parameter framework. The line parameter access module is used to determine the connection relationship of the lines between converter stations and to connect the line resistance and line reactance to the corresponding connection nodes of the calibrated impedance parameter framework. The high-frequency correction and model building module is used to analyze the high-frequency waveform timing of the output of the calibrated impedance parameter framework, identify the period of the repetitive oscillation signal, calculate the compensation coefficient accordingly, and perform high-frequency correction on the impedance parameter framework to obtain the impedance model of the new energy flexible low-frequency transmission system.

[0066] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0067] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for constructing an impedance model for a flexible low-frequency transmission system for new energy sources, characterized in that, The new energy flexible low-frequency transmission system includes a multi-level matrix converter bridge arm and an inter-converter station line. The method includes the following steps: Step S1: Record the real-time timing of voltage and current of the converter arm under different operating conditions; extract the time interval between voltage peak and current peak, the rate of change of voltage waveform slope and current waveform oscillation period, and label them with corresponding operating condition tags to form a dynamic feature library of the arm. Step S2: Identify the voltage and current waveforms of the converter bridge arm, and simultaneously record the line resistance and line reactance between converter stations; Step S3: Extract the voltage and current change features from the voltage and current waveforms, match the same-label features in the bridge arm dynamic feature library, and dynamically calibrate the bridge arm parameters in the preset impedance parameter framework. In step S3, the bridge arm parameters in the preset impedance parameter framework for dynamic calibration include: Retrieve the reference parameters of the current bridge arm from the preset impedance parameter framework, including equivalent resistance and equivalent inductance; The adjustment trend of resistance and inductance is determined based on the direction of the difference between the matching characteristics and the preset reference characteristics. Adjust the equivalent resistance according to the difference ratio. Each time it is adjusted, compare the resistance correlation characteristic deviation between the analog waveform output by the parameter frame and the real-time acquired waveform. Once the resistance deviation meets the requirements, adjust the equivalent inductance using the same logic until the inductance-related characteristic deviation also meets the requirements. Step S4: Determine the connection relationship between the lines between converter stations, and connect the line resistance and line reactance to the corresponding connection nodes in the calibrated impedance parameter framework; Step S5: Analyze the high-frequency waveform timing of the output of the calibrated impedance parameter framework, identify the period of the repetitive oscillation signal, calculate the compensation coefficient accordingly, and perform high-frequency correction on the impedance parameter framework to obtain the impedance model of the new energy flexible low-frequency transmission system.

2. The impedance model construction method for a new energy flexible low-frequency transmission system according to claim 1, characterized in that, Step S1, which records the real-time timing of the voltage and current of the converter bridge arm, includes: For each working condition, the bridge arm is switched from the current working condition to the target working condition. After the output waveform has no obvious jump for three consecutive low-frequency cycles, timing recording is started. During the recording process, the stable operating condition within each low-frequency cycle is simultaneously marked. After recording is complete, the real-time timing data of the voltage and current of the converter bridge arm will be segmented according to the marked stable period, and only the timing data of the stable period will be retained.

3. The impedance model construction method for a new energy flexible low-frequency transmission system according to claim 1, characterized in that, The specific steps in step S1 for forming the bridge arm dynamic feature library are as follows: Extract the voltage peak value and current peak value within 5 consecutive low-frequency cycles, calculate the time interval between the two within each cycle, and take the average time interval of the 5 cycles as the time interval feature of the time series. Extract the slope change value of the voltage waveform at each rising edge, and take the average of the slope change values ​​of three consecutive rising edges as the slope change rate feature of the voltage waveform. Identify continuous complete oscillations in the current waveform, calculate the period values ​​of three adjacent oscillations, and take the average value as the oscillation period characteristic of the current waveform; The time interval features, voltage waveform slope change rate features, and current waveform oscillation period features are bound to the corresponding operating condition labels and stored in the bridge arm dynamic feature library according to the structure of operating condition label-feature group-original time sequence index.

4. The impedance model construction method for a new energy flexible low-frequency transmission system according to claim 1, characterized in that, Before extracting the voltage and current change features from the voltage and current waveforms in step S3, the following steps are included: Real-time acquisition of voltage and current waveforms, and baseline correction, using the average amplitude of the waveform as a reference to eliminate overall waveform offset; After correction, the amplitude of each local segment of the voltage waveform and current waveform is compared with the overall average amplitude. When the amplitude of a local segment exceeds a preset multiple of the overall average amplitude, the segment is determined to be an abnormal pulse. Replace the local waveform containing the abnormal pulse with the interpolation of the adjacent normal waveform, and then extract the voltage and current change characteristics from the voltage and current waveforms.

5. The impedance model construction method for a new energy flexible low-frequency transmission system according to claim 1, characterized in that, The specific steps for extracting voltage and current variation features from voltage and current waveforms in step S3 are as follows: The real-time voltage and current waveforms are divided into continuous analysis units according to the low-frequency cycle, and each analysis unit contains one complete cycle. For each analysis unit, mark the voltage zero-crossing point and the current zero-crossing point. Starting from the zero-crossing point, extract the position and time of the voltage peak and the current peak respectively. Calculate the time difference between the peak voltage and peak current within the same analysis unit, and use it as the time interval characteristic of that unit; Calculate the slope change of the voltage and current waveforms during the rising phase, and take the average value of this change as the characteristic of voltage-current change.

6. The impedance model construction method for a new energy flexible low-frequency transmission system according to claim 1, characterized in that, In step S3, the features with the same label in the bridge arm dynamic feature library include: Retrieve all feature groups from the bridge boom dynamic feature library that are completely consistent with the current working condition label; For the called feature groups, they are compared in the order of time interval, slope change rate, and oscillation period, and feature groups in which each dimension is within the preset matching range are retained. From the retained feature groups, calculate the comprehensive deviation value between the current feature and each group of features, and select the feature with the smallest comprehensive deviation value as the matching result.

7. The impedance model construction method for a new energy flexible low-frequency transmission system according to claim 6, characterized in that, Verification of parameter adjustment effectiveness during calibration includes: The analog waveform output from the preset impedance parameter framework is precisely aligned with the real-time acquired waveform along the time axis. Calculate the peak deviation, period deviation, and slope deviation of the aligned waveforms respectively; When the peak deviation, period deviation, and slope deviation are all within the allowable range of the corresponding dimension, the calibration is deemed qualified, parameter adjustment is stopped, and the current parameter is locked.

8. The impedance model construction method for a new energy flexible low-frequency transmission system according to claim 1, characterized in that, The timing sequence of the high-frequency waveform output from the impedance parameter framework analyzed and calibrated in step S5 is as follows: The high-frequency waveform is divided into multiple analysis windows with a fixed duration, and each window contains at least 5 complete potential oscillation cycles; For each window, identify the period value of all oscillation signals and count their frequency of occurrence; Summarize the statistical results of all windows and take the period value with the highest frequency as the period of the repetitive oscillation signal.

9. An impedance model construction system for a new energy flexible low-frequency transmission system, characterized in that, The impedance model construction system for the new energy flexible low-frequency transmission system as described in claim 1 includes: The bridge arm feature library construction module is used to record the real-time timing of voltage and current of converter bridge arms under different operating conditions; extract the time interval between voltage peak and current peak, the rate of change of voltage waveform slope and current waveform oscillation period, and label them with corresponding operating condition tags to form a bridge arm dynamic feature library; The bridge arm and line parameter acquisition module is used to identify the voltage waveform and current waveform of the converter bridge arm, and simultaneously record the line resistance and line reactance between converter stations; The bridge arm parameter calibration module is used to extract the voltage and current change characteristics in the voltage waveform and current waveform, match the same label features in the bridge arm dynamic feature library, and dynamically calibrate the bridge arm parameters in the preset impedance parameter framework. The line parameter access module is used to determine the connection relationship of the lines between converter stations and to connect the line resistance and line reactance to the corresponding connection nodes of the calibrated impedance parameter framework. The high-frequency correction and model building module is used to analyze the high-frequency waveform timing of the output of the calibrated impedance parameter framework, identify the period of the repetitive oscillation signal, calculate the compensation coefficient accordingly, and perform high-frequency correction on the impedance parameter framework to obtain the impedance model of the new energy flexible low-frequency transmission system.

Images