A harmonic collaborative compensation method and device suitable for a converter hybrid cluster
Patent Information
- Application Number
- CN202611371512.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-04
- Publication Date
- 2026-10-30
Smart Images
Figure CN123018873A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of power electronics and new energy grid-connected control technology, and in particular to a harmonic collaborative compensation method and device suitable for hybrid converter clusters. Background Technology
[0002] With the large-scale integration of distributed power sources into the grid, power quality issues in converter cluster grid-connected systems are becoming increasingly prominent, especially harmonic pollution and resonance problems caused by the nonlinear switching action of power electronic devices and multi-machine interaction.
[0003] Current technologies for harmonic mitigation typically employ globally unified compensation strategies or damping control based on fixed parameters. However, in complex operating conditions with multiple generators operating in parallel, resonant sources often exhibit time-varying and concealed characteristics. Existing detection methods struggle to quickly and accurately identify whether the resonance originates from a single generator unit, internal station interactions, or external grid disturbances. This lack of localization capability results in a lack of targeted mitigation strategies: applying excessively strong damping or compensation to non-resonant source units not only increases system control costs and losses but may also trigger new secondary resonances due to parameter mismatch; conversely, failing to identify the true resonant source renders oscillations ineffective.
[0004] Therefore, there is an urgent need to provide a harmonic collaborative compensation method and device suitable for hybrid converter clusters to solve the above-mentioned technical problems. Summary of the Invention
[0005] This invention provides a harmonic collaborative compensation method and apparatus suitable for converter hybrid clusters, which can suppress the oscillation of converter cluster grid-connected systems.
[0006] In a first aspect, embodiments of the present invention provide a harmonic collaborative compensation method applicable to hybrid converter clusters, comprising: Obtain the port impedance data of each generator unit in the converter cluster system; Based on the port impedance data, the impedance weight difference of each power generation unit is determined, and the resonant source is initially screened according to the impedance weight difference to obtain the initial screening result. In response to the initial screening result meeting the preset triggering condition, a small disturbance signal is injected into the control loop of the target power generation unit. The transfer function is extracted based on the output current response after the small disturbance signal is injected, and the resonant source type of the converter cluster system is determined according to the transfer function. Based on the type of resonant source, the local controller and cluster coordination controller in the converter cluster system are driven to perform differentiated harmonic collaborative compensation control.
[0007] Secondly, embodiments of the present invention also provide a harmonic collaborative compensation device suitable for a hybrid converter cluster, comprising: The acquisition module is used to acquire the port impedance data of each power generation unit in the converter cluster system; The initial screening module is used to determine the impedance weight difference of each of the power generation units based on the port impedance data, and to perform initial screening of the resonant source according to the impedance weight difference to obtain the initial screening result. The response module is used to inject a small disturbance signal into the control loop of the target power generation unit in response to the initial screening result meeting the preset triggering condition, extract the transfer function based on the output current response after the small disturbance signal is injected, and determine the resonant source type of the converter cluster system according to the transfer function. The processing module is used to drive the local controller and the cluster coordination controller in the converter cluster system to perform differentiated harmonic collaborative compensation control based on the type of the resonant source.
[0008] Thirdly, embodiments of the present invention also provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements the method described in any embodiment of the present invention.
[0009] Fourthly, embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to perform the methods described in any embodiment of the present invention.
[0010] This invention provides a harmonic collaborative compensation method and apparatus suitable for hybrid converter clusters. After acquiring the port impedance data of each generator unit at the current steady-state operating point, the impedance weight value of each generator unit is calculated. This impedance weight value characterizes the contribution or sensitivity of the generator unit to the resonant energy of a specific frequency band of the system under the current operating condition. For example, the proportion of the negative damping component of the generator unit's port impedance near the resonant frequency can be calculated, or the influence of the unit's impedance change on the total impedance characteristics of the system can be quantified into a weight value based on impedance sensitivity analysis. The calculated real-time impedance weight value is then compared with a benchmark weight value. The benchmark weight value originates from historical statistical data or offline simulation models of the system under normal steady-state operating conditions, representing the impedance distribution characteristics that each unit should have under healthy conditions. The impedance weight difference essentially reflects the degree of deviation of each unit's current state from the healthy benchmark. By comparing the impedance weight differences of all generator units, the unit with the largest absolute difference is identified as a candidate resonant source. The physical basis for this selection is that when the system resonates, whether due to equipment parameter drift or control instability, the impedance characteristics of the fault source or the unit mainly involved in the oscillation often exhibit the most significant abnormal shift. By locking the unit with the largest deviation as the target power generation unit, a high-probability target point can be locked, effectively avoiding and suppressing the oscillation of the converter cluster grid-connected system. Attached Figure Description
[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention or related technologies, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0012] Figure 1 This is a flowchart of a harmonic collaborative compensation method for a hybrid converter cluster provided in an embodiment of the present invention; Figure 2 This is a hardware architecture diagram of the electronic device provided in an embodiment of the present invention; Figure 3 This is a structural diagram of a harmonic collaborative compensation device for a hybrid converter cluster provided in an embodiment of the present invention. Detailed Implementation
[0013] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.
[0014] Please refer to Figure 1 This invention provides a harmonic collaborative compensation method suitable for hybrid converter clusters, the method comprising: Step 100: Obtain the port impedance data of each power generation unit in the converter cluster system; Step 102: Determine the impedance weight difference of each power generation unit based on the port impedance data, and perform preliminary screening of resonant sources according to the impedance weight difference to obtain the preliminary screening results. Step 104: In response to the initial screening result meeting the preset triggering condition, inject a small disturbance signal into the control loop of the target power generation unit, extract the transfer function based on the output current response after the small disturbance signal is injected, and determine the resonant source type of the converter cluster system based on the transfer function. Step 106: Based on the type of resonant source, drive the local controller and cluster coordination controller in the converter cluster system to perform differentiated harmonic collaborative compensation control.
[0015] In this embodiment, after obtaining the port impedance data of each power generation unit at the current steady-state operating point, the impedance weight value of a single power generation unit is calculated. This impedance weight value characterizes the contribution or sensitivity of the power generation unit to the resonant energy of a specific frequency band of the system under the current operating condition. For example, it can be calculated by calculating the proportion of the negative damping component of the power generation unit's port impedance near the resonant frequency, or by quantifying the influence of the unit's impedance change on the total impedance characteristics of the system as a weight value based on impedance sensitivity analysis. The difference between the calculated real-time impedance weight value and the benchmark weight value is calculated. The benchmark weight value comes from historical statistical data or offline simulation models of the system under normal steady-state operating conditions, representing the impedance distribution characteristics that each unit should have under healthy conditions. The impedance weight difference essentially reflects the degree of deviation of the current state of each unit from the healthy benchmark. The impedance weight difference of all power generation units is compared, and the unit with the largest absolute difference is determined as the candidate resonant source. The physical basis for this selection is that when the system resonates, whether due to equipment parameter drift or control instability, the impedance characteristics of the fault source or the unit mainly involved in the oscillation often show the most significant abnormal deviation. By locking the unit with the largest deviation as the target power generation unit, a high-probability target point can be locked, effectively avoiding and suppressing the oscillation of the converter cluster grid-connected system.
[0016] In this embodiment, taking the frequency domain method as an example, the port impedance of the i-th power generation unit at frequency f can be expressed as: The formula for calculating the port impedance of the i-th generating unit at frequency f is given by: and These are the frequency domain representations of the port voltage and current, respectively.
[0017] in, and These are the frequency domain voltage and frequency domain current obtained after performing Discrete Fourier Transform on the voltage and current waveform data of the acquisition port. These impedance data not only contain the amplitude and phase information of the impedance, but also implicitly indicate the degree of influence of this unit on the system's resonance characteristics.
[0018] Subsequently, the system converts the real-time impedance data of each unit into impedance weight values. These weight values are not simply impedance magnitudes, but rather characterize the overall contribution of the generating unit to the system's resonant energy within the frequency band of interest. Specifically, the impedance weight value of the i-th generating unit can be calculated using the following formula: The impedance weighting value is calculated by weighted integration of the impedance magnitude within the frequency band of interest, where w(f) is the frequency weighting function. in,[ f 1 ,f [2] represents the resonant frequency band of interest. w(f) This is a frequency weighting function used to highlight frequency components with a higher risk of system resonance. In practical applications, w(f) The weighting factor can be set based on the system's historical resonant frequency distribution or impedance sensitivity analysis results, giving it a larger weighting factor in the high-frequency resonant band. Furthermore, the impedance weighting value can also be obtained based on impedance sensitivity analysis, which quantifies the impact of the unit impedance change on the system's total impedance characteristics as a weighting value. Its mathematical expression is: The weighted value calculated based on impedance sensitivity represents the partial derivative of the impedance change of the i-th unit with respect to the system's resonant frequency shift. in, This represents the sensitivity of the impedance change of the i-th generator unit to the system's resonant frequency. This represents the change in the unit impedance relative to the previous sampling period.
[0019] Next, the system calculates the difference between the calculated impedance weight value and the pre-stored reference weight value. Reference weight value Derived from historical statistical data or offline simulation models of the system under normal steady-state operating conditions, it represents the impedance distribution characteristics that each unit should have under healthy conditions. The impedance weight difference of the i-th generating unit is defined as: The impedance weight difference calculation formula reflects the degree of deviation of each unit's current state from the health baseline. Impedance weight difference Essentially, it reflects the degree to which each unit's current state deviates from a healthy baseline. The larger the difference, the further the unit's impedance characteristics deviate from normal operating conditions, and the higher the likelihood that it will participate in or cause resonance.
[0020] Finally, the system compares the impedance weight differences among all power generation units and identifies the unit with the largest absolute difference as the candidate resonant source. Assuming the converter cluster system has N power generation units, the index k of the candidate resonant source can be determined by the following formula: The target power generation unit is then used as the initial screening result. The physical basis for this selection is that when the system resonates, whether due to equipment parameter drift or control instability, the impedance characteristics of the fault source or the unit mainly involved in the oscillation often exhibit the most significant abnormal deviation. By locking the unit with the largest deviation as the target power generation unit, the computational redundancy and disturbance risks caused by traversing the entire system are effectively avoided.
[0021] In one embodiment of the present invention, the impedance weight difference of each power generation unit is determined based on port impedance data, and the resonant source is initially screened according to the impedance weight difference to obtain the initial screening result, including: Based on the port impedance data of each power generation unit at the current steady-state operating point, calculate the impedance weight value of a single power generation unit. The impedance weight difference is obtained by calculating the difference between the impedance weight value and the benchmark weight value corresponding to the system impedance network matrix under normal steady-state conditions. The generator unit with the largest impedance weight difference is identified as the candidate resonant source, and the candidate resonant source is used as the target generator unit to output the initial screening results.
[0022] In this embodiment, during actual execution, the system first collects port voltage and current waveform data in real time using voltage and current sensors deployed at the grid connection points of each power generation unit. Then, it calculates the port impedance data of each unit under the current operating conditions using impedance identification algorithms (such as the frequency domain method based on Fourier transform or the time domain method based on least squares). This impedance data not only includes the amplitude and phase information of the impedance but also implicitly indicates the degree of influence of the unit on the system's resonant characteristics. Subsequently, the system converts the real-time impedance data of each unit into impedance weight values. These weight values can be the weighted integral of the impedance magnitude over a specific frequency band or the contribution coefficient of the unit to the system's resonant frequency shift obtained based on impedance sensitivity analysis. Next, the system calculates the difference between the calculated impedance weight values and pre-stored benchmark weight values. The benchmark weight values are derived from historical statistical data or offline simulation models of the system under normal steady-state operating conditions, representing the impedance distribution characteristics that each unit should have under healthy conditions. The impedance weight difference essentially reflects the degree of deviation of each unit's current state from the healthy benchmark. Finally, the system compares the impedance weight differences of all power generation units, identifies the unit with the largest absolute difference as a candidate resonant source, and outputs the initial screening result as the target power generation unit.
[0023] In one embodiment of the present invention, determining the resonant source type of the converter cluster system based on the transfer function includes: Based on the difference between the transfer function and the converter's own transfer function, a transfer coefficient matrix is constructed, and the maximum value of the diagonal elements in the transfer coefficient matrix is extracted. If the maximum value of the diagonal element is greater than the first preset threshold and the maximum value of the off-diagonal element is less than the second preset threshold, then the resonant source type is determined to be a single power generation unit resonance. If the maximum value of the diagonal element is less than or equal to the first preset threshold, and the maximum value of the off-diagonal element is greater than or equal to the second preset threshold, then the type of resonance source is determined to be internal station interactive resonance. If the maximum value of the diagonal element is less than or equal to the first preset threshold, and the maximum value of the off-diagonal element is less than the second preset threshold, then the resonant source type is determined to be external power grid resonance.
[0024] In this embodiment, after obtaining the measured transfer function of the target power generation unit, it is necessary to further analyze its inherent coupling mechanism to determine the type of resonant source. First, the transfer coefficient matrix A is defined, where the element A ij Let A represent the transfer gain of the input disturbance of the j-th power generation unit to the output response of the i-th power generation unit. For the target power generation unit (let's say it's the k-th unit), its diagonal element A... kk It characterizes the coupling strength between its own input and its own output, i.e., the self-coupling property; off-diagonal element A kj(j / k) characterizes the mutual coupling properties of other elements to this element. In practical calculations, the difference between the measured transfer function and the theoretical ideal transfer function (or the reference transfer function under normal conditions) of the element can be calculated to eliminate the influence of inherent characteristics and highlight the abnormal coupling components. Subsequently, the maximum value M of the diagonal elements in the difference matrix is extracted. diag The maximum value M of the off-diagonal elements off The results are then compared with preset thresholds. The first and second preset thresholds are set based on the system's own coupling strength baseline and mutual coupling strength baseline, which can be obtained through offline simulation or statistical analysis of historical normal operation data.
[0025] In this embodiment, diagonal element A ii Characterizes the coupling strength between the input and output of the i-th power generation unit, i.e., its self-coupling characteristic; off-diagonal element A ij This characterizes the mutual coupling characteristics between the j-th power generation unit and the i-th power generation unit. The physical meaning of each element is: when a small disturbance signal is injected into the control loop of the j-th power generation unit... At that time, the output current response of the i-th generator unit The frequency domain ratio of it, that is: The transfer coefficient matrix elements are calculated as the frequency domain complex ratio of the injected disturbance to the output current response.
[0026] For the target power generation unit (let's call it the k-th unit), its diagonal element A kk It characterizes the coupling strength between its own input and its own output, i.e., the self-coupling property; off-diagonal element A kj This characterizes the mutual coupling characteristics of other units to that unit.
[0027] In practical calculations, to eliminate the influence of the inherent transmission characteristics of each power generation unit and highlight abnormal coupling components, it is necessary to subtract the measured transfer function from the theoretical ideal transfer function (or the reference transfer function under normal conditions) of the unit, and construct the difference matrix D: The formula for constructing the difference matrix eliminates the influence of inherent characteristics by subtracting the measured transfer function from the benchmark transfer function.
[0028] in, The elements of the reference transfer function matrix, measured under normal steady-state operating conditions, can be obtained through offline identification or statistical averaging of historical normal operating data. Elements in the difference matrix D. D ij It reflects the degree of deviation of the coupling relationship between units under the current operating conditions from the healthy state.
[0029] Subsequently, the maximum value of the diagonal elements and the maximum value of the off-diagonal elements in the difference matrix are extracted, and the calculation methods are as follows: If M diag Significantly higher than the first preset threshold and M off If the value is below the second preset threshold, it indicates that the resonant energy is mainly concentrated inside the target unit, caused by its own parameter mismatch or control abnormality, and is judged as a single power generation unit resonance; if M diag Not exceeding the standard but M off If the value is significantly higher than the second preset threshold, it indicates that the resonance is excited by strong interactions between multiple power generation units, and the target unit is merely a node in the interactive network, thus it is determined to be an internal interactive resonance within the power station. If neither exceeds its respective threshold, it indicates that the target unit is neither faulty nor interactively driven, and the resonance energy is more likely to originate from background harmonics or impedance matching issues in the external power grid, thus it is determined to be an external power grid resonance. This classification criterion based on the amplitude relationship of matrix elements transforms the abstract characteristics of the transfer function into clear physical mechanism labels.
[0030] In one embodiment of the present invention, based on the type of resonant source, the local controller and the cluster coordination controller in the driver converter cluster system perform differentiated harmonic collaborative compensation control, including: If the resonant source type is a single generating unit resonance, the local controller of the target generating unit is driven to switch to transient virtual impedance control mode or adjust the phase-locked loop bandwidth to provide active damping suppression. If the type of resonance source is internal resonant resonance within the power station, the drive cluster coordination controller dynamically allocates harmonic compensation weights and reactive power compensation weights according to the impedance characteristics of each power generation unit in order to collaboratively suppress resonant resonance.
[0031] In this embodiment, if the resonant source is an internal interconnected resonance within the power station, the drive cluster coordination controller dynamically allocates harmonic compensation weights and reactive power compensation weights based on the impedance characteristics of each power generation unit to collaboratively suppress the interconnected resonance. Specifically, internal interconnected resonance within the power station is usually a collective oscillation mode caused by impedance coupling or control parameter mismatch between multiple converters. In this case, independent adjustment of any single unit may disrupt the original coupling balance or even exacerbate the oscillation. Therefore, the cluster coordination controller must take a global perspective and coordinate the allocation of compensation tasks. Unlike the traditional average allocation based on capacity ratio, this application proposes a differentiated weight allocation strategy based on impedance characteristics, aiming to avoid the instability risk caused by multiple units competing for control.
[0032] In one embodiment of the present invention, the local controller driving the target power generation unit switches to transient virtual impedance control mode or adjusts the phase-locked loop bandwidth, including: A virtual impedance component that is positively correlated with the target frequency corresponding to the type of resonant source is superimposed in the outer voltage loop or inner current loop of the target power generation unit. Alternatively, the bandwidth of the phase-locked loop can be adaptively reduced based on the grid voltage drop depth to weaken the voltage outer loop tracking characteristics and improve the current inner loop accuracy. The drive cluster coordination controller dynamically allocates harmonic compensation weights and reactive power compensation weights based on the impedance characteristics of each generation unit, including: Calculate the negative impedance amplitude of each power generation unit at the target frequency, and use the reciprocal of the negative impedance amplitude as the compensation weight allocation factor for the corresponding power generation unit. Based on the compensation weight allocation factor, generate the harmonic compensation current reference value and reactive power compensation current reference value for each power generation unit.
[0033] In this embodiment, the negative impedance amplitude characterizes the active gain characteristics or instability tendency of the power generation unit at a specific frequency. At the resonant frequency, the larger the negative impedance amplitude, the more negative damping energy the unit injects into the system, the lower its stability margin, and the more likely it is to become a source of oscillation or an amplifier. If such "fragile" units are given more compensation tasks according to conventional thinking, it is very easy to cause their control to saturate or further deviate from the stable operating point, thereby inducing more serious secondary resonances. Conversely, this application uses the reciprocal of the negative impedance amplitude as a weighting factor, which means that units with larger negative impedance amplitudes (more unstable) are assigned smaller compensation weights; while units with smaller negative impedance amplitudes (more stable, better damping characteristics) undertake the main compensation tasks.
[0034] For example, assuming three converters in a cluster participate in collaborative compensation, and their negative impedance amplitudes measured at the resonant frequency are Z1, Z2, and Z3 respectively, then their compensation weight allocation factors k1, k2, and k3 are proportional to 1 / Z1, 1 / Z2, and 1 / Z3, respectively. After normalization, the converter with the best stability will receive the largest compensation current reference value command, while the converter on the verge of instability will only undertake auxiliary fine-tuning tasks. This "supporting the strong and suppressing the weak" allocation mechanism physically avoids the risk of applying excessive control pressure to unstable nodes, ensuring the stability of the collaborative compensation process itself. Compared with global unified compensation, this strategy not only significantly reduces the overall control cost and equipment losses of the system, but more importantly, it eliminates the hidden danger of new oscillation modes caused by blind compensation, demonstrating the superiority of targeted hierarchical collaborative control in complex multi-machine interaction scenarios. It should be understood that the above weight allocation based on the reciprocal of negative impedance is only a preferred implementation method. In other embodiments, the weight factors can also be comprehensively corrected by combining multiple dimensions such as the remaining capacity and thermal state of each unit.
[0035] In one embodiment of the present invention, before acquiring the port impedance data of each power generation unit in the converter cluster system, the method further includes: The power resonance signal in the region where forced resonance occurs is obtained, and the forced resonance signal and random noise signal in the power resonance signal are separated based on higher-order statistical analysis to obtain the forced resonance center frequency; The phase information corresponding to the center frequency of forced resonance is extracted based on time-frequency analysis, and the transient energy flow direction of the system is calculated based on the phase information to determine whether the forced resonance disturbance source is located in the converter cluster system.
[0036] In this embodiment, specifically, in actual grid-connected operation environments, the acquired power signals are often mixed with a large amount of background noise and random disturbances. Directly using the original signal for impedance measurement or resonant source localization may lead to misjudgment. This embodiment introduces a preprocessing mechanism combining higher-order statistics and time-frequency analysis to extract pure forced resonance components from the complex mixed signal, providing a high signal-to-noise ratio input basis for subsequent accurate localization. This preprocessing step can be automatically triggered when the system detects abnormal power quality or harmonic exceedances, or it can be performed as a periodic inspection task. It should be understood that although this embodiment describes it as a pre-processing step, under certain ideal operating conditions with high signal-to-noise ratios, this step can be omitted or simplified, directly proceeding to the impedance screening process of the embodiment, which does not depart from the protection scope of this application.
[0037] In one embodiment of the present invention, the forced resonance signal and random noise signal in the power resonance signal are separated based on higher-order statistical analysis to obtain the forced resonance center frequency, including: Calculate the third-order cumulant of the power resonant signal, and construct a bispectral diagonal slice based on the third-order cumulant. The frequency point with the largest amplitude in the bispectral diagonal slice is determined as the forced resonant center frequency. Phase information corresponding to the forced resonance center frequency is extracted based on time-frequency analysis, including: Morlet wavelet is selected as the basis wavelet, and the wavelet parameters are set so that the wavelet center frequency is equal to the forced resonance center frequency. The Morlet wavelet is used to perform continuous wavelet transform on the power resonant signal to extract the wavelet coefficients near the forced resonant center frequency. Phase information is obtained based on wavelet coefficients.
[0038] In this embodiment, specifically, after determining the forced resonance center frequency, it is necessary to further obtain the phase trajectory of this frequency component over time to analyze the energy flow direction. Traditional Fourier transforms cannot simultaneously achieve both time and frequency resolution, while wavelet transforms possess excellent time-frequency localization characteristics. This application preferably uses the Morlet wavelet as the basis wavelet function because the Morlet wavelet exhibits excellent clustering properties similar to Gaussian windows in both the time and frequency domains, making it particularly suitable for extracting the instantaneous phase of oscillating signals. The key lies in the adaptive setting of the wavelet parameters: by adjusting the scaling factor and translation factor, the center frequency of the selected Morlet wavelet is precisely aligned with the forced resonance center frequency identified in the aforementioned steps. When the wavelet center frequency matches the resonant frequency in the signal, the modulus of the wavelet coefficients at the corresponding scale will reach its maximum value, and other components far from this frequency will be effectively filtered out due to orthogonality or rapid attenuation. At this point, the complex wavelet coefficients at this specific scale are extracted, and their argument represents the instantaneous phase of the forced resonance signal at that moment. Compared to filter banks with fixed parameters, this method of dynamically adjusting wavelet parameters based on the measured center frequency can adapt to the slight drift that may occur in the resonant frequency, ensuring the accuracy and robustness of phase extraction.
[0039] Furthermore, after obtaining the phase information corresponding to the forced resonance center frequency, the transient energy flow direction of the system can be calculated based on the phase information to determine whether the forced resonance disturbance source is located in the converter cluster system. Specifically, the transient energy flow direction is a physical quantity characterizing the resonant energy transmission path. Using the instantaneous phase difference between the port voltage and current of each generator unit at the forced resonance center frequency extracted above, the instantaneous active power oscillation component of each branch and its integral form of transient energy can be calculated. If the transient energy of a generator unit or a feeder continuously flows out (i.e., the energy slope is positive or the cumulative value is positive), it indicates that the node is the source of resonant energy injection; conversely, if the energy continuously flows in, it is the sink of resonant energy absorption. By comparing the sign and magnitude of the transient energy flow direction of each node in the cluster, it is possible to quickly determine whether the disturbance source is located inside the converter cluster (such as the control instability of a certain converter) or comes from the external power grid (such as background harmonic excitation from the power grid). This judgment result can serve as an important auxiliary basis for determining the type of resonant source in Example 1. Especially when the initial screening result of impedance weight difference is in a critical state, the energy flow information can provide an additional physical verification dimension, further improving the reliability and accuracy of the overall positioning strategy.
[0040] In one embodiment of the present invention, the forced resonance signal and random noise signal in the power resonance signal are separated based on higher-order statistical analysis to obtain the forced resonance center frequency, including: Calculate the third-order cumulant of the power resonant signal, and construct a bispectral diagonal slice based on the third-order cumulant. The frequency point with the largest amplitude in the bispectral diagonal slice is determined as the forced resonant center frequency.
[0041] In this embodiment, higher-order statistics have a unique advantage over traditional second-order statistics (such as power spectral density) in that they are naturally immune to Gaussian noise. Since background noise in power systems typically approximates a Gaussian distribution, its third-order and higher-order cumulants are theoretically zero. However, forced resonant signals often exhibit significant nonlinear coupling characteristics, and their higher-order cumulants are not zero. Therefore, by calculating the third-order cumulants of the power resonant signal, interference from Gaussian white noise and colored noise can be effectively suppressed, highlighting the nonlinear characteristics of the resonant signal. Here, and represent two independent time delay variables, and represents the input power resonant signal sequence. Based on this, to reduce the computational dimensionality and retain the most important frequency coupling information, this application further constructs a bispectral diagonal slice. The bispectrum is a two-dimensional Fourier transform of the third-order cumulant, while the diagonal slice is a one-dimensional special section with time , and its mathematical expression essentially reflects the distribution of the signal's self-coupling strength along the frequency axis. In the diagonal slice spectrum, the frequency point where the amplitude peaks corresponds to the frequency component with the strongest nonlinear interaction in the system, and this is determined as the forced resonant center frequency. This processing method not only significantly reduces the amount of data processing, but also avoids the interference of redundant information caused by full bispectral calculation, enabling the resonant frequency to be locked robustly even in low signal-to-noise ratio environments.
[0042] like Figure 2 , Figure 3 As shown, this embodiment of the invention provides a harmonic collaborative compensation device suitable for hybrid converter clusters. The device embodiment can be implemented through software, hardware, or a combination of both. From a hardware perspective, as... Figure 2 The diagram shown is a hardware architecture diagram of an electronic device for a harmonic collaborative compensation device suitable for a hybrid converter cluster, provided by an embodiment of the present invention. Except for... Figure 2 In addition to the processor, memory, network interface, and non-volatile memory shown, the electronic device in the embodiment may also include other hardware, such as a forwarding chip responsible for processing packets. Taking software implementation as an example, such as... Figure 3 As shown, a device in a logical sense is formed by the CPU of the electronic device in which it is located reading the corresponding computer program from the non-volatile memory into the memory for execution.
[0043] This embodiment provides a harmonic collaborative compensation device suitable for hybrid converter clusters, comprising: The acquisition module 300 is used to acquire the port impedance data of each power generation unit in the converter cluster system; The initial screening module 302 is used to determine the impedance weight difference of each of the power generation units based on the port impedance data, and to perform initial screening of the resonant source according to the impedance weight difference to obtain the initial screening result. The response module 304 is used to inject a small disturbance signal into the control loop of the target power generation unit in response to the initial screening result meeting the preset triggering condition, extract the transfer function based on the output current response after the small disturbance signal is injected, and determine the resonant source type of the converter cluster system according to the transfer function. The processing module 306 is used to drive the local controller and the cluster coordination controller in the converter cluster system to perform differentiated harmonic collaborative compensation control based on the type of the resonant source.
[0044] In this embodiment of the invention, the acquisition module 300 can be used to execute step 100 in the above method embodiment, the initial screening module 302 can be used to execute step 102 in the above method embodiment, the response module 304 can be used to execute step 104 in the above method embodiment, and the processing module 306 can be used to execute step 106 in the above method embodiment.
[0045] It is understood that the structures illustrated in the embodiments of the present invention do not constitute a specific limitation on a harmonic co-compensation device suitable for a hybrid converter cluster. In other embodiments of the present invention, a harmonic co-compensation device suitable for a hybrid converter cluster may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.
[0046] The information interaction and execution process between the modules in the above-mentioned device are based on the same concept as the method embodiment of the present invention, and the specific details can be found in the description of the method embodiment of the present invention, and will not be repeated here.
[0047] This invention also provides an electronic device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements a harmonic collaborative compensation method for a converter hybrid cluster according to any embodiment of this invention.
[0048] This invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform a harmonic collaborative compensation method applicable to a converter hybrid cluster according to any embodiment of this invention.
[0049] Specifically, a system or apparatus equipped with a storage medium may be provided, on which software program code implementing the functions of any of the embodiments described above is stored, and the computer (or CPU or MPU) of the system or apparatus may read and execute the program code stored in the storage medium.
[0050] In this case, the program code read from the storage medium can itself implement the function of any of the above embodiments, and therefore the program code and the storage medium storing the program code constitute part of the present invention.
[0051] Storage media embodiments for providing program code include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), magnetic tapes, non-volatile memory cards, and ROMs. Alternatively, program code can be downloaded from a server computer via a communication network.
[0052] Furthermore, it should be clear that not only can the program code read by the computer be executed, but also the operating system or other components operating on the computer can be instructed based on the program code to perform some or all of the actual operations, thereby realizing the function of any of the embodiments described above.
[0053] Furthermore, it is understood that the program code read from the storage medium is written to the memory set in the expansion board inserted into the computer or to the memory set in the expansion module connected to the computer. Then, based on the instructions of the program code, the CPU or other components installed on the expansion board or expansion module execute some and all of the actual operations, thereby realizing the function of any of the above embodiments.
[0054] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or electronic device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or electronic device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or electronic device that includes said element.
[0055] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various storage media that can store program code, such as ROM, RAM, magnetic disk, or optical disk.
[0056] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A harmonic collaborative compensation method applicable to hybrid converter clusters, characterized in that, include: Obtain the port impedance data of each generator unit in the converter cluster system; Based on the port impedance data, the impedance weight difference of each power generation unit is determined, and the resonant source is initially screened according to the impedance weight difference to obtain the initial screening result. In response to the initial screening result meeting the preset triggering condition, a small disturbance signal is injected into the control loop of the target power generation unit. The transfer function is extracted based on the output current response after the small disturbance signal is injected, and the resonant source type of the converter cluster system is determined according to the transfer function. Based on the type of resonant source, the local controller and cluster coordination controller in the converter cluster system are driven to perform differentiated harmonic collaborative compensation control.
2. The method according to claim 1, characterized in that, The process of determining the impedance weight difference of each power generation unit based on the port impedance data, and performing initial screening of resonant sources based on the impedance weight difference to obtain the initial screening results includes: Based on the port impedance data of each power generation unit at the current steady-state operating point, calculate the impedance weight value of a single power generation unit; The impedance weight difference is obtained by calculating the difference between the impedance weight value and the benchmark weight value corresponding to the system impedance network matrix under normal steady-state conditions. The power generation unit with the largest impedance weight difference is identified as a candidate resonant source, and the candidate resonant source is used as the target power generation unit to output the initial screening result.
3. The method according to claim 2, characterized in that, Determining the resonant source type of the converter cluster system based on the transfer function includes: Based on the difference between the transfer function and its own converter transfer function, a transfer coefficient matrix is constructed, and the maximum value of the diagonal elements in the transfer coefficient matrix is extracted. If the maximum value of the diagonal element is greater than the first preset threshold and the maximum value of the off-diagonal element is less than the second preset threshold, then the type of the resonant source is determined to be a single power generation unit resonance. If the maximum value of the diagonal element is less than or equal to the first preset threshold, and the maximum value of the off-diagonal element is greater than or equal to the second preset threshold, then the type of the resonance source is determined to be internal station interactive resonance. If the maximum value of the diagonal element is less than or equal to the first preset threshold, and the maximum value of the off-diagonal element is less than the second preset threshold, then the resonant source type is determined to be external power grid resonance.
4. The method according to claim 1, characterized in that, The step of driving the local controller and cluster coordination controller in the converter cluster system to perform differentiated harmonic collaborative compensation control based on the resonant source type includes: If the resonant source type is a single power generation unit resonance, the local controller of the target power generation unit is driven to switch to transient virtual impedance control mode or adjust the phase-locked loop bandwidth to provide active damping suppression. If the type of resonance source is internal interactive resonance within the power station, the cluster coordination controller is driven to dynamically allocate harmonic compensation weights and reactive power compensation weights according to the impedance characteristics of each power generation unit, so as to collaboratively suppress interactive resonance.
5. The method according to claim 4, characterized in that, The local controller driving the target power generation unit switches to transient virtual impedance control mode or adjusts the phase-locked loop bandwidth, including: A virtual impedance component that is positively correlated with the target frequency corresponding to the type of resonant source is superimposed in the outer voltage loop or inner current loop of the target power generation unit; Alternatively, the bandwidth of the phase-locked loop can be adaptively reduced based on the grid voltage drop depth to weaken the voltage outer loop tracking characteristics and improve the current inner loop accuracy. The process of driving the cluster coordination controller to dynamically allocate harmonic compensation weights and reactive power compensation weights based on the impedance characteristics of each power generation unit includes: Calculate the negative impedance amplitude of each power generation unit at the target frequency, and use the reciprocal of the negative impedance amplitude as the compensation weight allocation factor for the corresponding power generation unit. Based on the compensation weight allocation factor, the harmonic compensation current reference value and reactive power compensation current reference value of each of the power generation units are generated.
6. The method according to claim 1, characterized in that, Before acquiring the port impedance data of each generator unit in the converter cluster system, the method further includes: The power resonance signal in the region where forced resonance occurs is obtained, and the forced resonance signal and random noise signal in the power resonance signal are separated based on higher-order statistical analysis to obtain the forced resonance center frequency; The phase information corresponding to the forced resonance center frequency is extracted based on time-frequency analysis, and the transient energy flow direction of the system is calculated based on the phase information to determine whether the forced resonance disturbance source is located in the converter cluster system.
7. The method according to claim 6, characterized in that, The step of separating the forced resonance signal and random noise signal from the power resonance signal based on higher-order statistical analysis to obtain the forced resonance center frequency includes: Calculate the third-order cumulant of the power resonant signal, and construct a bispectral diagonal slice based on the third-order cumulant. The frequency point with the largest amplitude in the bispectral diagonal slice is determined as the forced resonant center frequency. The extraction of phase information corresponding to the forced resonance center frequency based on time-frequency analysis includes: The Morlet wavelet is selected as the basis wavelet, and the wavelet parameters are set so that the wavelet center frequency is equal to the forced resonance center frequency. The power resonant signal is subjected to continuous wavelet transform using the Morlet wavelet to extract the wavelet coefficients near the forced resonant center frequency. The phase information is obtained based on the wavelet coefficients.
8. A harmonic collaborative compensation device suitable for hybrid converter clusters, characterized in that, For performing the method as described in any one of claims 1-7, comprising: The acquisition module is used to acquire the port impedance data of each power generation unit in the converter cluster system; The initial screening module is used to determine the impedance weight difference of each of the power generation units based on the port impedance data, and to perform initial screening of the resonant source according to the impedance weight difference to obtain the initial screening result. The response module is used to inject a small disturbance signal into the control loop of the target power generation unit in response to the initial screening result meeting the preset triggering condition, extract the transfer function based on the output current response after the small disturbance signal is injected, and determine the resonant source type of the converter cluster system according to the transfer function. The processing module is used to drive the local controller and the cluster coordination controller in the converter cluster system to perform differentiated harmonic collaborative compensation control based on the type of the resonant source.
9. An electronic device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method as described in any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed in a computer, causes the computer to perform the method described in any one of claims 1-7.