Self-adaptive weighted feed-forward control system of LCL type grid-connected inverter under weak power grid

Abstract

According to the adaptive weighted feedforward control system of the LCL type grid-connected inverter under the weak power grid, the problems of phase lag and instability are avoided, and the efficiency and stability of the grid-connected inverter under the weak power grid are effectively improved; the system comprises a detection control module which collects and preprocesses voltage signals at two ends of a filter capacitor; constructing a feed-forward function, introducing a weighting coefficient to perform weighting processing on the feed-forward function, further adjusting the amplitude of a rotating vector in the feed-forward function through the weighting coefficient, inputting the preprocessed voltage signal into the optimized feed-forward function, and converting the preprocessed voltage signal into a weighted feed-forward signal; and the impedance remodeling and adjusting module reconstructs the output impedance of the inverter according to the weighted feed-forward signal, further judges whether the inverter is stable or not, and adaptively adjusts the weighting coefficient in real time based on the phase margin when the inverter is judged to be unstable.

Description

Technical Field

[0001] This invention relates to the fields of power electronics and new energy power generation technology, specifically to an adaptive weighted feedforward control system for LCL-type grid-connected inverters under weak power grid conditions. Background Technology

[0002] With the widespread application of renewable energy and the integration of distributed energy sources such as photovoltaic power generation and wind power generation, more and more grid-connected inverters are being used in power systems. These grid-connected inverters usually need to operate in weak grid environments, which refer to power systems with large voltage fluctuations, high impedance, and insufficient short-circuit capacity. The stability and performance of traditional grid-connected inverters are often affected under such grid conditions, especially when there are large voltage fluctuations, frequency disturbances, and impedance changes in the grid, which can easily lead to system instability or reduced efficiency.

[0003] LCL-type grid-connected inverters are commonly used in power systems due to their excellent filtering performance. However, in weak grid environments, the resonant characteristics of the LCL filter may adversely affect the inverter's stability due to the high grid impedance. Capacitor voltage full feedforward control is a commonly used harmonic suppression method, but under digital control, the introduction of the feedforward loop will cause phase lag in the control system, significantly reducing the inverter's adaptability in weak grid environments. Impedance-based stability criteria indicate that in weak grid scenarios, correcting the inverter's output impedance can improve system stability. Existing research has proposed a weighted proportional-differential grid voltage feedforward scheme to extend the passive frequency band of the inverter's output impedance, but this scheme suffers from reduced system harmonic suppression capability and complex control parameter design. Therefore, it is urgent to design an adaptive control strategy to cope with changes under weak grid conditions, which has become a pressing problem. This invention addresses the above problems by proposing an adaptive weighted feedforward control system for LCL-type grid-connected inverters in weak grid environments. Summary of the Invention

[0004] In order to solve the technical problems mentioned in the background art, the purpose of this invention is to provide an adaptive weighted feedforward control system for LCL-type grid-connected inverters under weak power grid conditions.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: An adaptive weighted feedforward control system for an LCL-type grid-connected inverter operating under weak grid conditions. The LCL-type grid-connected inverter consists of a voltage source inverter, an LCL filter, a point of common coupling, and the grid-side equivalent impedance. The LCL filter consists of an inverter-side filter inductor L1, a filter capacitor C, and a grid-side filter inductor L2. The system includes: The detection and control module acquires and preprocesses the voltage signal across the filter capacitor; constructs a feedforward function based on the parameters of the LCL filter, and introduces weighting coefficients to weight the feedforward function. The magnitude of the rotation vector in the feedforward function is further adjusted by the weighting coefficients to obtain an optimized feedforward function. The preprocessed voltage signal is input into the optimized feedforward function to convert it into a weighted feedforward signal. The impedance reshaping adjustment module reconstructs the output impedance of the inverter based on the weighted feedforward signal, and further determines whether the inverter is stable. When it is determined to be unstable, it adaptively adjusts the weighting coefficient in real time based on the phase margin.

[0006] Furthermore, the feedforward function includes a second-order differential term and a proportional term; the weighting coefficients include a first weighting coefficient and a second weighting coefficient, which respectively weight the second-order differential term and the proportional term. The magnitude of the rotation vector in the optimized feedforward function The expression is: in, Angular frequency represents the frequency characteristics of a signal; and Indicates the first weighting coefficient and the second weighting coefficient; The optimized feedforward function expression is as follows: in, These are the PWM modulation coefficients; and Representing the proportional term and the second-order differential term, respectively corresponding to DC component and dynamic component; This represents the optimized feedforward function; s This represents the complex frequency variable of the Laplace transform.

[0007] Furthermore, the process of reconstructing the inverter's output impedance is as follows: 1) Establish a small-signal model of the inverter. Perform linear small-signal modeling of the main circuit and control loop in the complex frequency domain. The current-voltage transfer relationship can be derived using Kirchhoff's laws as follows: in, This indicates the current in the inverter-side filter inductor. This represents the PWM voltage output by the inverter; This represents the voltage across the filter capacitor. Indicates the grid-connected current; Indicates the voltage at the point of common coupling; , and Represents the parasitic resistance of inductors and capacitors; 2) Integrate the optimized feedforward function into the inverter's control loop and derive the output impedance after incorporating the optimized feedforward function; 3) Measure the output impedance of the actual inverter by frequency sweep and compare it with the output impedance of the inverter derived from the small-signal model of the inverter. If the amplitude error of the output impedance of the inverter derived from the small-signal model of the inverter is ≤5% and the phase error is ≤3°, it proves that the output impedance of the inverter derived from the small-signal model of the inverter is accurate.

[0008] Furthermore, the process for determining whether the inverter is stable is as follows: 1) Define the ratio of the grid-side equivalent impedance to the inverter output impedance as an open-loop transfer function, expressed as: in, Represent the open-loop transfer function; Indicates the equivalent impedance on the grid side; Indicates the inverter output impedance; 2) Count the number of poles in the right half-plane of the open-loop transfer function. P The open-loop transfer function has no poles in the right half-plane, therefore... ; 3) Plot the Bode plot based on the output impedance expression. The Bode plot includes the amplitude-frequency response (open-loop gain) and the phase-frequency response (open-loop phase angle). Finally, determine the crossover frequency, and using the phase angle at the crossover frequency as the calculation reference, define the phase margin expression as follows: in, Indicates phase margin; This represents the phase angle of the inverter output impedance at the crossover frequency; This indicates the angular frequency corresponding to the crossover frequency; 4) Determine the trajectory of the open-loop transfer function. Net number of points N ; When phase margin The open-loop transfer function trajectory does not enclose points, net turns ; When phase margin Time-open-loop transfer function trajectory encircling points, net turns ; Introduction and Afterwards, PM increases, at which point the open-loop transfer function trajectory no longer encloses the loop. point, Then the inverter is open-loop stable; 5) Through Verify closed-loop stability; already determined. , therefore It satisfies the closed-loop stability condition.

[0009] Furthermore, the adaptive adjustment process of the weighting coefficients is as follows: A preset phase margin threshold is defined. When the detected phase margin is lower than the preset phase margin threshold, the following action is taken: Decrease by 0.05 Increase the step size by 0.05 until the phase margin is higher than the phase margin threshold.

[0010] Compared with the prior art, the advantages of the present invention are as follows: 1. This invention optimizes the feedforward function by weighting it, and can dynamically adjust the control strategy according to the actual situation of the power grid, effectively improving the adaptability of the inverter in weak power grids. 2. This invention optimizes the output impedance of the inverter by using a weighted feedforward signal, thus avoiding the phase lag and instability problems caused by impedance mismatch in traditional methods. 3. By combining current dual closed-loop control and weighted feedforward control, this invention can not only effectively suppress grid harmonics, but also achieve phase compensation, thus avoiding instability caused by grid voltage fluctuations. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a system workflow diagram of the present invention; Figure 2 This is a flowchart of the detection and control module of the present invention; Figure 3 This is a flowchart of the impedance reshaping adjustment module of the present invention. Detailed Implementation

[0013] To achieve the above objectives, this invention provides an adaptive weighted feedforward control system for an LCL-type grid-connected inverter operating under weak grid conditions. Please refer to [link to relevant documentation]. Figures 1 to 3 The LCL-type grid-connected inverter consists of a voltage source inverter, an LCL filter, a point of common coupling, and an equivalent impedance on the grid side. The LCL filter is composed of an inverter-side filter inductor. L1 Filter capacitor C and grid-side filter inductor L2 In this embodiment, the composition is as follows: , and Connect the input terminal of the LCL filter to the output terminal of the voltage source inverter, and connect the output terminal of the LCL filter to the grid-side equivalent impedance via the point of common coupling. In a weak power grid, inductors are used. Typical values; The detection and control module acquires and preprocesses the voltage signal across the filter capacitor; constructs a feedforward function based on the parameters of the LCL filter, and introduces weighting coefficients to weight the feedforward function. The magnitude of the rotation vector in the feedforward function is further adjusted by the weighting coefficients to obtain an optimized feedforward function. The preprocessed voltage signal is input into the optimized feedforward function to convert it into a weighted feedforward signal. Real-time detection of filter capacitor using a voltage sensor The voltage signal at both ends, the accuracy of the voltage sensor must meet the requirements. sampling frequency This voltage contains information about the resonant characteristics of the LCL filter and grid disturbances, as well as the equivalent impedance on the grid side under weak grid conditions. When the voltage increases, the phase shift of the voltage signal directly reflects the stability of the inverter. The acquired voltage signal is low-pass filtered using a second-order RC low-pass filter circuit with a cutoff frequency of [missing information]. resistance ,capacitance Filtering out switching ripple and sensor noise, the sensor output is amplified by an operational amplifier. Signal scaling to the digital controller ADC interface Input range, avoid signal overflow, and obtain preprocessed voltage signal. ; The feedforward function is constructed based on the parameters of the LCL filter. The feedforward function includes a second-order differential term and a proportional term. The weighting coefficients include a first weighting coefficient and a second weighting coefficient, which are used to weight the second-order differential term and the proportional term, respectively. In this embodiment, the first weighting coefficient Second weighting coefficient In traditional feedforward functions, both the proportional and differential terms have a weight of 1, leading to accumulated phase lag under digital delay. In this embodiment, the magnitude balance is broken by splitting the weights. First weighting coefficient Used to adjust the weights of the second-order derivative. This allows the high-frequency output impedance phase to change from... Degree increased to Above a certain degree, avoid high-frequency instability; the high-frequency band is greater than [a certain value]. ; Second weighting coefficient Used to adjust the weight of the proportional term. A value close to 1 preserves low-frequency gain, ensuring harmonic suppression and preventing an increase in low-frequency total harmonic distortion due to excessive attenuation of the feedforward function's proportional term; the low-frequency range is less than... ; Using the first weighting coefficient Second weighting coefficient Adjusting the magnitude of the rotation vector in the feedforward function Represented as: in, Angular frequency represents the frequency characteristics of the signal and is used to describe the amplitude variation of the optimized feedforward function at different frequencies. By adjusting the amplitude of the rotating vector in the optimized feedforward function, the two vectors with equal amplitudes and opposite directions in the traditional scheme are transformed into superimposed vectors with unequal amplitudes, thus offsetting the phase lag introduced by digital delay; in this embodiment, under weak power grid conditions... This increases the traditional solution's degree from 3.4 to 47.2 degrees, far exceeding the stability threshold; The optimized feedforward function expression is as follows: in, This is the PWM modulation coefficient, which is 1 in this embodiment; and Representing the proportional term and the second-order differential term, respectively corresponding to DC component and dynamic component; and This represents the first and second weighting coefficients, with a range of values. , ; This represents the optimized feedforward function; s The complex frequency variable represents the Laplace transform; The preprocessed voltage signal is input into the optimized feedforward function and converted into a weighted feedforward signal; the core function of the weighted feedforward signal is to compensate for phase lag and reshape the output impedance of the inverter.

[0014] The impedance reshaping adjustment module reconstructs the output impedance of the inverter based on the weighted feedforward signal, and further determines whether the inverter is stable. When it is determined to be unstable, it adaptively adjusts the weighting coefficient in real time based on the phase margin. Under the traditional feedforward strategy, the equivalent impedance on the grid side Phase drop at high frequencies Degree, easily related to output impedance Resonance is formed; The weighted feedforward signal reshapes the inverter output impedance into a low-frequency band by adjusting its phase characteristics. High frequency band The smooth characteristics of the temperature gradient avoid the equivalent impedance of the grid side. Phase cancellation occurs; The process of reconstructing the inverter's output impedance is as follows: 1) Establish a small-signal model of the inverter. Perform linear small-signal modeling of the main circuit and control loop in the complex frequency domain. Based on Kirchhoff's laws, ignore the on-state voltage drop of power devices, and incorporate the parasitic resistance of inductors and capacitors into the model to ensure that the dynamic characteristics closely match reality. The current-voltage transfer relationship can be derived: in, This indicates the current in the inverter-side filter inductor. This represents the PWM voltage output by the inverter; This represents the voltage across the filter capacitor. Indicates the grid-connected current; Indicates the voltage at the point of common coupling; , and Represents the parasitic resistance of inductors and capacitors; 2) Integrate the optimized feedforward function into the inverter's control loop, and derive the output impedance expression after incorporating the optimized feedforward function; the output impedance is defined as the ratio of the point of common coupling voltage to the grid-connected current, i.e. Integrating feedforward By incorporating the small-signal model, simplifying it through a control block diagram, and re-deriving the model... and Based on the relationship, a new expression for the output impedance is obtained; 3) The accuracy of the small-signal model is verified by measuring the output impedance of the actual inverter through frequency domain sweep and comparing it with the output impedance of the inverter derived by the model. If the amplitude error of the output impedance of the inverter derived by the model is ≤5% and the phase error is ≤3°, it proves that the output impedance of the inverter derived by the small-signal model is accurate and verifies the accuracy of the small-signal model. By detecting the frequency response of the inverter output impedance and the grid-side equivalent impedance, it is determined whether the inverter is stable. If the inverter is determined to be stable, a trigger signal is provided for adaptive adjustment. The process for determining whether an inverter is stable is as follows: 1) Define the open-loop transfer function as the ratio of the grid-side equivalent impedance to the inverter output impedance, expressed as: in, Represent the open-loop transfer function; Indicates the equivalent impedance on the grid side; Indicates the inverter output impedance; 2) Count the number of poles in the right half-plane of the open-loop transfer function. P In this embodiment, the open-loop transfer function has no poles in the right half-plane, therefore... ; 3) Analyze the encirclement characteristics by examining the frequency response trajectory of the open-loop transfer function. In this embodiment, the output impedance expressions under different control strategies are first obtained through control block diagram simplification and Laplace transform. A Bode plot is then drawn based on the output impedance expressions. The Bode plot includes the amplitude-frequency characteristic (open-loop gain) and the phase-frequency characteristic (open-loop phase angle). Finally, the crossover frequency is determined. The crossover frequency is the frequency corresponding to an open-loop gain of 1. The phase angle at the crossover frequency is used as the calculation benchmark, and the phase margin expression is defined as follows: in, Indicates phase margin; This represents the phase angle of the inverter output impedance at the crossover frequency; The angular frequency corresponding to the crossover frequency is the frequency at which the equivalent impedance on the grid side is equal to the amplitude of the inverter output impedance. 4) Determine the trajectory of the open-loop transfer function. Net number of points N ; When phase margin The open-loop transfer function trajectory does not enclose points, net turns ; When phase margin Time-open-loop transfer function trajectory encircling points, net turns ; In this embodiment, the traditional capacitor voltage full feedforward strategy, when When, PM is calculated to be... At this point, the open-loop transfer function trajectory is close to being surrounded. The point is that it greatly weakens the adaptability to weak power grids and is on the verge of instability; while the weighted full feedforward strategy of this invention introduces and Afterwards, PM was upgraded to At this point, the open-loop transfer function trajectory does not enclose the loop at all. point, Then the inverter is open-loop stable; 5) Through Verify closed-loop stability; this has been determined in this embodiment. , therefore This satisfies the closed-loop stability condition; When the inverter is determined to be unstable, adaptive adjustment is achieved based on the phase margin closed loop. This adjustment is made when the grid-side equivalent impedance... When increasing the phase margin leads to a decrease, automatic adjustment is performed. and ; A preset phase margin threshold is defined. When the detected phase margin is below the threshold, [the system] will [follow the] preset phase margin threshold. Decrease by 0.05 Increase the adjustment step by 0.05 until the phase margin rises back above the threshold; in this embodiment, when the grid-side equivalent impedance When varying within the range of 0.5mH to 2.0mH, the phase margin is kept stable between 40 and 50 degrees without any risk of instability; Reducing this will make the amplitude of the rotating vector of the feedforward function in the high-frequency band closer to the hysteresis vector of the digital delay, thus offsetting part of the phase hysteresis and shifting the high-frequency phase of the inverter's output impedance from... Degree increased to Above a certain degree, avoid creating resonance conditions with the equivalent impedance on the power grid side; Increase the low-frequency gain of the retained proportional term. This enhances the feedforward signal's ability to compensate for grid fundamental disturbances, suppresses the interactive coupling between the inverter's output impedance and the grid-side equivalent impedance in the low-frequency band, and improves the phase margin near the fundamental frequency.

[0015] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An adaptive weighted feed-forward control system for LCL-type grid-connected inverter under weak grid, the LCL-type grid-connected inverter is composed of voltage source inverter, LCL filter, point of common coupling and grid-side equivalent impedance , the LCL filter is composed of inverter-side filter inductance L1 , filter capacitance C and grid-side filter inductance L2 , characterized in that, The system comprises: The detection control module collects and pre-processes the voltage signal across the filter capacitor; constructs a feedforward function based on the parameters of the LCL filter, introduces a weighting coefficient to weight the feedforward function, further adjusts the amplitude of the rotating vector in the feedforward function through the weighting coefficient, obtains an optimized feedforward function, and inputs the pre-processed voltage signal into the optimized feedforward function to convert it into a weighted feedforward signal; The impedance remodeling adjustment module reconstructs the output impedance of the inverter according to the weighted feedforward signal, further judges whether the inverter is stable, and when it is judged to be unstable, adjusts the weighting coefficient in real time based on the phase margin.

2. The system of claim 1, wherein, The feedforward function comprises a second-order differential term and a proportional term; the weighting coefficient comprises a first weighting coefficient and a second weighting coefficient, which are used to weight the second-order differential term and the proportional term, respectively; the magnitude of the rotation vector in the optimized feedforward function is expressed as wherein is the angular frequency, representing the frequency characteristic of the signal; and denotes a first weighting factor and a second weighting factor; The expression of the optimized feedforward function is: wherein, is a PWM modulation factor; and denotes a proportional term and a second order derivative term, respectively, corresponding to a direct current component and a dynamic variation component; denotes an optimized feedforward function; s denotes a complex frequency variable of the Laplace transform.

3. The system of claim 2, wherein, The process of reconstructing the output impedance of the inverter is as follows: 1) Establish a small-signal model of the inverter, linearly model the main circuit and control loop in the complex frequency domain, and derive the transfer relationship between current and voltage through Kirchhoff's law, which is expressed as: wherein, represents the inverter-side filter inductance current; represents the PWM voltage of the inverter output; represents the filter capacitance voltage; represents the grid-connected current; represents the point of common coupling voltage; , and represents the parasitic resistance of the inductance, capacitance; 2) Integrate the optimized feedforward function into the control loop of the inverter, and derive the output impedance after integrating the optimized feedforward function; 3) Measure the output impedance of the actual inverter by frequency domain sweep, compare it with the output impedance of the inverter derived from the model, and if the amplitude error of the output impedance of the inverter derived from the small-signal model of the inverter is ≤5% and the phase error is ≤3°, it is proved that the output impedance of the inverter derived from the small-signal model of the inverter is accurate.

4. The system of claim 3, wherein, The process of judging whether the inverter is stable is as follows: 1) The ratio of the equivalent impedance of the grid side to the output impedance of the inverter is the open-loop transfer function, which is expressed as: wherein, represents an open-loop transfer function; represents an equivalent grid-side impedance; represents an inverter output impedance; 2) Count the number of right half plane poles of the open loop transfer function P ; the open loop transfer function has no right half plane poles, hence ; 3) Draw a Bode plot according to the output impedance expression, which includes the amplitude-frequency characteristic, i.e., the open-loop gain, and the phase-frequency characteristic, i.e., the open-loop phase angle; finally determine the cross-over frequency, take the phase angle at the cross-over frequency as the calculation reference, and define the phase margin expression as: wherein, denotes the phase margin; denotes the phase angle of the inverter output impedance at the cross-over frequency; denotes the angular frequency corresponding to the cross-over frequency; 4) determining the net number of loops of the open-loop transfer function trajectory around the point N ; When the phase margin The open loop transfer function trajectory does not encircle The net number of turns ; When the phase margin The open loop transfer function trajectory encircles The net number of turns ; Introduction and After that, PM increases, at which point the open-loop transfer function trajectory does not enclose the point, then the inverter is open-loop stable; 5) by verifying closed loop stability; it has been determined , therefore , the closed loop stability condition is met.

5. The system of claim 4, wherein, The process of adjusting the weighting coefficient in real time is as follows: a pre-set phase margin threshold, when detecting that the phase margin is lower than the phase margin threshold, adjusting the phase margin by decreasing 0.05, increasing 0.05, until the phase margin is higher than the phase margin threshold.

Images