A method for estimating relative frequency of grid-connected point of grid-connected type converter under weak grid

Abstract

The application discloses a method for estimating the relative frequency of a grid-connected converter under a weak grid. Firstly, the active power and the reactive power of the grid-connected converter are measured, and the time derivative of the powers is calculated after the powers are filtered by a low-pass filter. Then, the operation state of the grid-connected converter is identified to determine whether the grid-connected converter is in a constant voltage control mode or a current limit control mode. Next, the power signals and their derivatives are input into a closed-loop observer based on the complex frequency theory. In the observer, the original relative frequency deviation is reconstructed by using an algebraic bias adjustment mechanism. Finally, the original relative frequency deviation is input into a signal conditioning unit composed of an integral unit, a low-pass filter and a differential unit for noise reduction and smoothing, and the final relative frequency deviation is output. The estimated relative frequency deviation can replace the traditional active power control loop with the absolute frequency deviation as the feedback quantity, and the active power reference value is corrected in real time to improve the synchronization stability of the grid-connected converter under the frequency fluctuation of the weak grid.

Description

Technical Field

[0001] This invention belongs to the field of equipment control technology for use with power sources or similar power supply systems, and specifically relates to a method for estimating the relative frequency of grid-connected converters in weak power grids. Background Technology

[0002] With the widespread application of grid-connected technologies, especially grid-connected converters (VSGs), in modern power systems, accurately estimating the frequency dynamics of their grid connection point under weak grid conditions is crucial for ensuring the synchronous stability of the power system. Currently used frequency acquisition methods (such as those based on phase-locked loops or virtual speeds within the VSG) typically only provide the "absolute frequency" relative to the rated synchronous reference coordinate system. In weak grids, power imbalances caused by severe faults can lead to fluctuations in grid frequency. In such cases, absolute frequency measurements cannot accurately reflect the "relative motion" between the VSG grid connection point and the weak grid reference system where frequency shifts occur. Although global information can be obtained through wide-area monitoring systems (WAMS), this method is costly and relies on system parameters such as impedance and inertia, which are often difficult to obtain accurately after a fault. Therefore, how to reconstruct the relative frequency of the VSG grid connection point using only local measurement information, without the aid of wide-area communication, has become a pressing technical challenge in current stability assessment and control. Summary of the Invention

[0003] To overcome the problems of existing technologies, such as the inability of absolute frequency measurement to reflect relative motion and the high cost and dependence on system parameters in wide-area communication, the present invention aims to provide a method for estimating the relative frequency at the grid connection point of a grid-connected converter in a weak power grid. The present invention utilizes only local power feedback to achieve real-time estimation of the relative frequency deviation between the grid-connected converter (VSG) and the weak power grid.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for estimating the relative frequency at the grid connection point of a grid-connected converter in a weak power grid includes the following steps: Step 1: Local signal acquisition Real-time measurement of active power output at grid connection point of grid-connected converter Reactive power output at the grid connection point of the grid-connected converter They are preprocessed using low-pass filters and then calculated. and ,in Active power The first derivative with respect to time, reactive power The first derivative with respect to time.

[0005] Step 2: Identification of Operating Modes of Grid-Type Converters The current operating mode of the grid converter is identified based on its operating status, and it is determined whether it is in constant voltage control (CVC) mode or current limiting control (CLC) mode.

[0006] Step 3: Construction of the equivalent reconstruction equation for relative frequency deviation Based on the mode determination result in step 2, determine the relative frequency deviation under the corresponding mode. The equivalent reconstruction equation.

[0007] a. In constant voltage control mode: 1) Relative frequency deviation The equivalent reconstruction equation is: , In the formula, Apparent power of grid-connected converters; For the imaginary part of the apparent power complex frequency of the grid-type converter; For the real part of the apparent power complex frequency of the grid-type converter, for The first derivative with respect to time; Let be the real part of the complex frequency of the internal potential vector of the grid-type converter. Let be the magnitude of the potential vector within the grid-type converter. for The first derivative with respect to time.

[0008] 2) As a simplified implementation, when the dynamic influence of voltage amplitude during the transient process is ignored, the relative frequency deviation under constant voltage control mode... The simplified form of the equivalent reconstruction equation is: .

[0009] b. In current-limited control mode: Relative frequency deviation The equivalent reconstruction equation is: .

[0010] Specifically, when the grid-connected system of the grid-connected converter includes a co-located grid-connected converter (GFL), in the constant voltage control mode of step 3, the active power and reactive power substituted into the equivalent reconfiguration equation are replaced by the total output active power and total output reactive power of the grid-connected converter and the grid-connected converter power supply cluster, respectively; in the current-limiting control mode of step 3, the active power and reactive power output at the grid connection point of the grid-connected converter are still used for reconfiguration.

[0011] Step 4: Relative Frequency Reconstruction Based on Closed-Loop Observer Preprocessed active power reactive power And the calculated active power First derivative with respect to time and reactive power First derivative with respect to time Input closed-loop observer: The closed-loop observer uses the numerator of the equivalent reconstruction equation as the target value and the estimated value of the deviation between the denominator of the equivalent reconstruction equation and the original relative frequency. The product of the target value and the feedback value is used as the error signal input to the proportional-integral controller. The output of the proportional-integral controller is the original relative frequency deviation estimate. and will The feedback value is updated by multiplying the denominator of the real-time feedback and the equivalent reconstruction equation to form a closed loop until the error converges to zero, thereby reconstructing the relative frequency deviation signal analytically to avoid the numerical singularity caused by explicit division.

[0012] Step 5: Signal Conditioning and Smoothing The original relative frequency deviation is smoothed by a signal conditioning process consisting of integration, low-pass filtering, and differentiation, and the final relative frequency deviation is output. .

[0013] Step 6: Relative Frequency Feedback and Reference Command Correction The final relative frequency deviation output from step 5 The feedback quantity is used by the proportional-integral controller to dynamically correct the active power reference command in real time, replacing the traditional feedback control logic based on the deviation between the absolute frequency and the rated frequency.

[0014] Compared with the prior art, the present invention has the following advantages: Current solutions for improving the synchronous stability of grid-connected converters in weak power grids primarily utilize the difference between the locally measured absolute frequency and the rated frequency as a feedback signal for power regulation. However, during transient processes with significant grid frequency fluctuations, accurately obtaining the slip frequency or relative motion information of the grid-connected converter's connection point relative to the dynamic grid reference frame typically requires a wide-area monitoring system (WAMS) to provide a global reference. This approach not only faces high communication and deployment costs but also heavily relies on parameters such as grid impedance and system inertia, which are difficult to accurately identify after a fault. It is challenging to implement in complex fault scenarios and often lacks robustness.

[0015] To address this, this invention proposes a method for estimating the relative frequency at the grid connection point of a grid-connected converter based on complex frequency theory. This method involves real-time acquisition of local active and reactive power and calculation of their time derivatives, using a closed-loop observer to reconstruct the relative frequency deviation of the grid-connected converter terminal relative to a weak grid reference system. Compared with existing schemes that rely on relative frequencies obtained from wide-area monitoring or directly use the difference between the internal absolute frequency and the rated frequency of the VSG for transient stability control, this invention offers the following advantages: 1) No communication required, relying only on local information: The relative frequency deviation can be resolved by local power signal feedback alone, without relying on wide-area monitoring systems and a posteriori system parameters, which has the characteristics of low cost and short latency; 2) High numerical stability: By using a closed-loop observer architecture, the complex division operation is transformed into a proportional-integral adjustment process for algebraic deviation, which avoids algorithm failure caused by computational overflow during power disturbances and significantly improves numerical robustness. 3) Balancing dynamic response and smoothness: The introduction of a signal conditioning stage consisting of integration, low-pass filtering, and differentiation effectively suppresses sampling ripple and numerical noise while minimizing the time delay caused by signal processing, ensuring real-time tracking capability of frequency fluctuations. 4) Adaptable to multiple operating conditions and cluster operation: Differentiated relative frequency reconfiguration logic is designed for constant voltage and current limiting control modes, and a processing scheme is provided for mixed grid-connected converters and grid-connected converters, which enhances the applicability of the method in complex heterogeneous grid-connected systems.

[0016] In summary, all the measurements required by this invention are derived from the local control system. The control logic is clear and easy to integrate into the existing grid-type converter controller. It has significant engineering application value for ensuring the safe and stable operation of new energy sources when they are connected to weak power grids on a large scale. Attached Figure Description

[0017] Figure 1 The flowchart of a method for estimating the relative frequency of a grid-connected converter under a weak power grid provided by the present invention is shown.

[0018] Figure 2 The relative frequency deviation acquisition structure block diagram provided by this invention illustrates the connection logic between the closed-loop observer and the signal conditioning stage.

[0019] Figure 3 This invention relates to a grid-connected system topology based on a hybrid connection of grid-connected and grid-structured converters.

[0020] Figure 4(a) shows the relative frequency deviation estimated by the present invention under CVC mode. Tracking effect on the true relative frequency; Figure 4(b) shows the relative frequency deviation estimated by the present invention in CLC mode. Tracking effect on true relative frequency; Figure 4(c) shows the tracking effect of traditional absolute frequency deviation in CVC mode. The relative frequency deviation estimated by this invention When used as a feedback quantity, the actual relative frequency deviation of the grid-connected system The dynamic response comparison; Figure 4(d) shows the results under CLC mode using traditional absolute frequency deviation. The relative frequency deviation estimated by this invention When used as a feedback quantity, the actual relative frequency deviation of the grid-connected system Dynamic response comparison. Detailed Implementation

[0021] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0022] This invention relates to a method for estimating the relative frequency of grid-connected converters in weak power grids, as shown in Figure 3. The system topology includes grid-connected converters and co-located grid-connected converters. Each converter is connected to the equivalent weak power grid point of common coupling (PCC) through its respective output impedance. The overall execution flow of the method described in this invention is as follows: Figure 1 As shown, the specific structural block diagram for obtaining the relative frequency deviation is as follows: Figure 2 As shown. The specific implementation steps of this method are as follows: Step 1: Local signal acquisition Real-time measurement of active power output at grid connection point of grid-connected converter Reactive power output at the grid connection point of the grid-connected converter They are preprocessed using low-pass filters and then calculated. and ,in Active power The first derivative with respect to time, reactive power The first derivative with respect to time. All of the above signals originate from the local control system of the grid-type converter, without relying on wide-area monitoring systems and a posteriori system parameters. They are characterized by low cost and short communication delay, overcoming the dependence of traditional relative frequency acquisition schemes on wide-area communication infrastructure.

[0023] Step 2: Identification of VSG operating mode for grid-type converters Reference Figure 1 The process involves identifying the current operating mode of the grid-connected converter (VSG) based on its operating status, and determining whether it is in constant voltage control (CVC) mode or current limiting control (CLC) mode.

[0024] Step 3: Construction of the equivalent reconstruction equation for relative frequency deviation Based on the mode determination result in step 2, determine the relative frequency deviation under the corresponding mode. The equivalent reconstruction equation is derived. This step, based on complex frequency theory, starts from the locally measurable power signal of the grid-type converter and derives an analytical relationship that expresses the relative frequency deviation as the ratio of the power signal and its time derivative. This transforms the relative frequency, which originally required wide-area reference information, into an algebraic expression that can be reconstructed solely by local measurements. This overcomes the dependence of traditional methods on parameters that are difficult to accurately identify after faults, such as grid impedance and system inertia.

[0025] a. In constant voltage control mode: 1) Relative frequency deviation The equivalent reconstruction equation is: , In the formula, Apparent power of grid-connected converters; For the imaginary part of the apparent power complex frequency of the grid-type converter; For the real part of the apparent power complex frequency of the grid-type converter, for The first derivative with respect to time; Let be the real part of the complex frequency of the internal potential vector of the grid-type converter. Let be the magnitude of the potential vector within the grid-type converter. for The first derivative with respect to time.

[0026] 2) As a simplified implementation, when the dynamic influence of voltage amplitude during the transient process is ignored, the relative frequency deviation under constant voltage control mode... The simplified form of the equivalent reconstruction equation is: .

[0027] This simplified form only includes locally measurable data. , Its time derivative completely eliminates the influence on the internal potential amplitude. E The dependence on its derivatives further reduces the difficulty of engineering implementation.

[0028] b. In current-limited control mode: Relative frequency deviation The equivalent reconstruction equation is: .

[0029] Specifically, when the VSG grid-connected system includes Figure 3When the grid-connected converter (GFL) is co-located as shown, in the constant voltage control mode of step 3, the active power and reactive power substituted into the equivalent reconfiguration equation are replaced with the total output active power and total output reactive power of the VSG and GFL power cluster, respectively; in the current limiting control mode of step 3, the active power and reactive power output from the VSG grid connection point are still used for reconfiguration.

[0030] Step 4: Relative Frequency Reconstruction Based on Closed-Loop Observer like Figure 2 As shown, based on complex frequency theory, the preprocessed active power... reactive power And the calculated active power First derivative with respect to time and reactive power First derivative with respect to time Input closed-loop observer: The observer uses the numerator of the equivalent reconstruction equation as the target value and the estimated value of the deviation between the denominator of the equivalent reconstruction equation and the original relative frequency. The product of the target value and the feedback value is used as the error signal input to the proportional-integral controller (PIC). The output of the PIC is the original relative frequency deviation estimate. and will The feedback value is updated by multiplying the denominator of the real-time feedback and the equivalent reconstruction equation to form a closed loop until the error converges to zero, thereby reconstructing the relative frequency deviation signal analytically to avoid the numerical singularity caused by explicit division.

[0031] Step 5: Signal Conditioning and Smoothing The original relative frequency deviation estimate Input smoothing processing device, through such Figure 2 The signal conditioning stage, consisting of the integral, low-pass filter, and differential stages shown, performs smoothing processing to output the final relative frequency deviation. The signal conditioning stage utilizes the inverse properties of integral and derivative functions combined with a low-pass filter to suppress numerical noise while achieving noise reduction and smoothing of the relative frequency estimate.

[0032] Step 6: Relative Frequency Feedback and Reference Command Correction The final relative frequency deviation Δ output from step 5 ω As a feedback quantity, the active power reference command is dynamically corrected in real time by the proportional-integral controller, replacing the traditional feedback control logic based on the deviation between the absolute frequency and the rated frequency. Thus, the slip frequency of the grid-connected point of the grid-type converter relative to the dynamic grid reference system can be obtained without relying on a wide-area measurement system.

[0033] Example: To verify the effectiveness of the method of this invention, a grid-connected system model including a grid-connected converter (VSG) and a grid-connected converter (GFL) was built on the PSCAD / EMTDC simulation platform. The rated capacity of the system was set as follows: 150 MW for the grid-connected converter VSG and 50 MW for the grid-connected converter GFL. The severe disturbance conditions were set as follows: the voltage of the weak grid bus drops to 0.1 per unit for 200 milliseconds; at the same time, the frequency response of the weak grid simulates the response characteristics of a synchronous generator under the frequency drop condition, and its lowest frequency point is set to -1.7 Hz.

[0034] Figures 4(a) and 4(b) show the tracking effect of the estimated value of the present invention on the true relative frequency deviation in CVC mode and CLC mode, respectively; Figures 4(c) and 4(d) show the results of using the traditional absolute frequency deviation in CVC mode and CLC mode, respectively. The relative frequency deviation estimated by this invention A comparison of the system frequency response when active power regulation is performed as a feedback quantity. Figures 4(a) and 4(b) show that, in both CVC and CLC modes, the relative frequency deviation estimated by this invention closely tracks the actual relative frequency. In contrast, traditional internal variables This fails to reflect the true frequency dynamics. Figures 4(c) and 4(d) show that when using traditional internal variables... When used as a feedback quantity, VSG exhibits continuous oscillations or even instability after a disturbance; while the quantity estimated by this invention... After being fed back to the active power loop, the VSG can maintain synchronization and stability throughout the entire frequency recovery process, and the frequency deviation quickly converges to zero.

[0035] In summary, this invention achieves dynamic compensation for frequency offset in weak grids with a fully localized and parameter-free control architecture, providing a low-cost and highly reliable transient stability improvement solution for grid-connected converters to weak grids.

Claims

1. A method for estimating the relative frequency of a grid-connected converter in a weak power grid, characterized in that: Includes the following steps: Step 1: Measure the active power output at the grid connection point of the grid-connected converter in real time. Reactive power output at the grid connection point of the grid-connected converter They are preprocessed using low-pass filters and then calculated. and ,in Active power The first derivative with respect to time, reactive power The first derivative with respect to time; Step 2: Identify the current operating mode of the grid converter based on its operating status, and determine whether the grid converter is in constant voltage control mode or current limiting control mode. Step 3: Based on the mode determination result in Step 2, determine the relative frequency deviation under the corresponding mode. The equivalent reconstruction equation; Step 4: Convert the pre-processed active power reactive power And the calculated active power First derivative with respect to time and reactive power First derivative with respect to time Input closed-loop observer: The closed-loop observer uses the numerator of the equivalent reconstruction equation as the target value and the estimated value of the deviation between the denominator of the equivalent reconstruction equation and the original relative frequency. The product of the target value and the feedback value is used as the error signal input to the proportional-integral controller (PIC). The output of the PIC is the original relative frequency deviation estimate. and will The feedback value is updated by multiplying the denominator of the real-time feedback and the equivalent reconstruction equation to form a closed loop until the error converges to zero. Step 5: Convert the original relative frequency deviation estimate into... The signal is smoothed by a signal conditioning process consisting of integration, low-pass filtering, and differentiation, and the final relative frequency deviation is output. ; Step 6: Calculate the final relative frequency deviation output from Step 5. As feedback, the active power reference command is dynamically corrected in real time by the proportional-integral controller.

2. The method for estimating the relative frequency of a grid-connected converter under a weak power grid according to claim 1, characterized in that: In step 3, the relative frequency deviation under constant voltage control mode The equivalent reconstruction equation is: ， In the formula, Apparent power of grid-connected converters; For the imaginary part of the apparent power complex frequency of the grid-type converter; For the real part of the apparent power complex frequency of the grid-type converter, for The first derivative with respect to time; Let be the real part of the complex frequency of the internal potential vector of the grid-type converter. Let be the magnitude of the potential vector within the grid-type converter. for The first derivative with respect to time.

3. The method for estimating the relative frequency of a grid-connected converter under a weak power grid according to claim 2, characterized in that: When the dynamic influence of voltage amplitude during the transient process is ignored, the relative frequency deviation under constant voltage control mode The simplified form of the equivalent reconstruction equation is: 。 4. The method for estimating the relative frequency of a grid-connected converter under a weak power grid according to claim 1, characterized in that: In step 3, the equivalent reconstruction equation for the relative frequency deviation Δω under the current-limited control mode is: 。 5. The method for estimating the relative frequency of a grid-connected converter under a weak power grid according to claim 1, characterized in that: When a grid-connected converter system includes a co-located grid-connected converter, in the constant voltage control mode of step 3, the active power and reactive power substituted into the equivalent reconfiguration equation are replaced by the total output active power and total output reactive power of the grid-connected converter and the grid-connected converter cluster, respectively; in the current-limiting control mode of step 3, the active power and reactive power output at the grid connection point of the grid-connected converter are still used for reconfiguration.

Images