Grid-connected inverter circulating current estimation and suppression method and device based on virtual synchronous machine

Abstract

The invention provides a grid-connected inverter circulating current estimation and suppression method and device based on a virtual synchronous machine, and the method comprises the steps: constructing an active power distribution network optical storage area cluster structure comprising a double-parallel grid-connected inverter interface converter and a hybrid energy storage device, and deducing a dynamic model of the double-parallel grid-connected inverter interface converter; extracting a closed-loop transfer function of the output power, the active power reference value, the actual output active power and the system frequency to form an extended state observer with a loop oscillation component estimation function; a total disturbance observed quantity is introduced into a control law based on an observation result of a model-free observer with oscillation estimation on a generalized disturbance reconstruction state, so that the system is converted into an active disturbance rejection state from a passive disturbance rejection state, and the model is simplified into a double-integrator series structure; pole assignment is also needed; according to the invention, the electric energy quality of the networking type optical storage cluster is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power distribution network, and in particular to a method and device for estimating and suppressing circulating current of a grid-connected inverter based on a virtual synchronous machine. Background Art

[0002] With the adjustment of energy structure, new energy forms represented by photovoltaics play a key role in the development of industrial modernization. With the help of the decline in system costs, as the proportion of photovoltaic power generation and power generation continues to increase in the future, the pace of my country's energy clean transformation continues to accelerate, and photovoltaics will gradually become the main power source.

[0003] At the same time, as an important means to enhance the flexibility and stability of the power system under the condition of high proportion of renewable energy access, the development of photovoltaic clusters will become an inevitable development trend. The large-scale development of photovoltaic storage clusters will also affect the power quality in the distribution network. Under normal circumstances, there are different load points in the power system, and the voltage will gradually decrease with the flow of power. When the distributed photovoltaic storage equipment cluster is running, the photovoltaic grid-connected inverter parallel operation technology can increase the rated power and accelerate the power conversion efficiency, but it will also derive the circulation phenomenon, which will lead to non-ideal energy consumption and affect the harmonic distortion rate (THD).

[0004] In the case of complex circulating currents, once reverse current and oscillation problems occur, the transmission power will be reduced, resulting in an increase in the voltage at the load point, and then voltage deviation. As a key part of the active distribution network, the distributed photovoltaic storage cluster will cause the current to change due to different output levels. At this time, it is difficult to control this with the system's own capabilities. When the distributed photovoltaic storage cluster is connected to or exits the system, the output power may change significantly, and the voltage may fluctuate and flicker. In addition, temperature changes, sunlight intensity, etc. will also cause current output power fluctuations and voltage fluctuations, seriously affecting the reliability and efficiency of the power system operation.

[0005] At present, the suppression of circulating current and oscillation is mainly achieved by optimizing and transforming the control link of the grid-connected inverter of the photovoltaic storage cluster in the active distribution network. According to the modeling analysis in the current research, the hybrid energy storage system composed of batteries and supercapacitors has significant nonlinear characteristics in the charging and discharging process. At the same time, the grid-connected inverter is also a nonlinear hybrid structure. Especially in the context of wide bandgap devices, the nonlinear characteristics of the active distribution network will be more obvious, which makes it difficult for the dynamic response capability of the conventional control scheme based on proportional-error feedback to meet the system's requirements for power quality. Based on this, it is necessary to invent a method to ensure that the circulating current oscillation disturbance of the grid-connected inverter of the distributed photovoltaic storage cluster in the active distribution network under uncertain loops is extracted and suppressed in time. Summary of the invention

[0006] This application proposes a method for estimating and suppressing the circulating current of a grid-connected inverter of a distributed photovoltaic storage cluster based on a virtual synchronous machine. This method can observe the oscillation disturbance of the circulating current of the grid-connected inverter and suppress generalized disturbances including oscillation disturbances, non-ideal modeling deviations and external disturbances. First, a model-free information ESO with oscillation estimation is designed for the generalized disturbances in the operation of the grid-connected inverter. Secondly, based on the observation results of the generalized disturbance reconstruction state by the model-free observer with oscillation estimation, the total disturbance observation is introduced into the control law to design a loop oscillation suppression strategy, so that the system is converted from a passive anti-disturbance state to an active anti-disturbance state, and the model is simplified to a dual-integrator series structure. On this basis, -k d z′2 replaces the The control law is thus reconstructed to eliminate the system oscillation phenomenon caused by the rapid change of the system given signal, and the feedback linear error type oscillation suppression rule is obtained; in addition, the parameter configuration method of the suppression rule is designed to enhance the adaptability of the circulating current estimation and suppression method to different working conditions.

[0007] A method for estimating and suppressing circulating current of a grid-connected inverter based on a virtual synchronous machine comprises the following steps:

[0008] Construct an active distribution network photovoltaic storage area cluster structure containing dual parallel grid-connected inverter interface converters and hybrid energy storage devices, and derive a dynamic model of dual parallel grid-connected inverter interface converters;

[0009] Extract the closed-loop transfer function of output power and active power reference value, actual output active power and system frequency, and perform corresponding self-disturbance rejection paradigm transformation of virtual synchronous mechanism network scheme, so as to form an extended state observer with loop oscillation component estimation function;

[0010] Construct a loop oscillation suppression strategy: Based on the observation results of the generalized disturbance reconstruction state by the model-free observer with oscillation estimation, the total disturbance observable is introduced into the control law, so that the system is converted from a passive anti-disturbance state to an active anti-disturbance state, and the model is simplified to a double integrator series structure; it is also necessary to determine the parameter setting rules of the loop oscillation suppression strategy through pole configuration, integrate the loop oscillation suppression strategy and the parameter setting rules, deal with the impact of loop oscillation, and improve the power quality of the grid-type photovoltaic storage cluster.

[0011] Furthermore, the method for constructing an active distribution network photovoltaic storage area cluster structure containing dual parallel grid-connected inverter interface converters and hybrid energy storage devices is as follows:

[0012]

[0013] In the formula, E k,i is the output fundamental voltage amplitude of the i-th grid-connected inverter under no-load condition; I out,1 and I out,2are the independent currents of two grid-connected inverter interface converters, is the total parallel current; T out,1 and T out,2 are the equivalent output impedances of the two grid-connected inverter interface converters; T line,1 and T line,1 is the line impedance of each; T total,1 and T total,2 are their respective total impedance values; Z kl is the load impedance, U kl is the load impedance voltage.

[0014] Furthermore, when deriving the dynamic model of the dual parallel grid-connected inverter interface converter, the independent currents of the dual parallel grid-connected inverter interface converter are decomposed as follows:

[0015]

[0016] Among them, I loop is the loop oscillation current caused by the relative voltage drop of the inverter; I load is the load current. When the system is running, I loop Oscillation interaction between dual parallel grid-connected inverter interface converters occurs when the distributed photovoltaic power generation carrying capacity at the dual parallel grid-connected inverter interface converters is unbalanced.

[0017] Furthermore, the closed-loop transfer function expressions of the output power, active power reference value, actual output active power and system frequency are as follows:

[0018] In order to suppress the oscillation phenomenon and avoid the introduction of harmonic components, it is necessary to effectively control the dual parallel grid-connected inverter interface converter with filtering link; the output power of the virtual synchronous mechanism network scheme and the active power reference value P ref , actual output active power P out The closed-loop transfer function expression of and system frequency ω is as follows:

[0019]

[0020] Where, J is the simulated moment of inertia; D is the damping coefficient; k f is the droop coefficient; τ is the filter absorption coefficient; P ref is the active power reference value; P out is the actual output active power; ω is the system frequency; ω n is the rated frequency; H1 and H2 are active power reference values ​​P ref And the actual output active power P out The pre-transfer function of .

[0021] Furthermore, the corresponding virtual synchronous mechanism network scheme self-disturbance rejection paradigm is as follows:

[0022] Based on the complex domain transfer function in (3), the time domain form of the active power-frequency control model can be obtained through inverse Laplace transform:

[0023]

[0024] To simplify the expression, the system state variable matrix is ​​defined as The state space expression paradigm of the active power-frequency control loop based on the virtual synchronous mechanism network scheme can be established by formula (4):

[0025]

[0026] In the formula, x1 and x2 are system state variables; u is the control state variable, satisfying u=ω n ; b0 is the control gain, satisfying b0=(D+k f ) -1 Jτ.

[0027] Furthermore, the extended state observer method with loop oscillation component estimation function is as follows:

[0028] Based on (5), all non-ideal components of the package are reconstructed into expanded state variables x3, and the expanded state space expression paradigm is:

[0029]

[0030] On the basis of (5), all non-ideal components are reconstructed into extended state variables x3, and the dynamic equation of the model-free information observer with oscillation estimation is designed as follows:

[0031]

[0032] Among them, b′=[β1′β2′β3′] T is the observation parameter vector; the vector [z1′z2′z3′] is the reconstructed observation state matrix, which is the observed values ​​of the first-order state variable x1, the second-order state variable x2 and the reconstructed extended state variable x3 of the system; A and B are the system matrix and the control matrix respectively; on this basis, the reconstructed variable x3 can be written as follows:

[0033]

[0034] In the formula, ψ m Modeling deviation for non-idealities; ψ l is the internal circulation oscillation disturbance; ψ d For external disturbance.

[0035] The parameter design principles of the model-free information observer are as follows:

[0036] In order to ensure the convergence of (6), the vector b′ must be properly configured so that the matrix A is a Hurwitz matrix. The pole placement method is used to adjust the observer bandwidth, and the characteristic roots of the extended state observer with the function of estimating the loop oscillation component are moved to the observation bandwidth. The parameter placement rule is written as:

[0037]

[0038] Calculated:

[0039]

[0040] Where I is the identity matrix; ω o is the observation bandwidth; s represents the complex domain independent variable in the Laplace transform.

[0041] Furthermore, the loop oscillation suppression strategy is designed as follows:

[0042] Based on the observation of the generalized disturbance reconstruction state by the model-free observer with oscillation estimation, a loop oscillation suppression strategy is designed to convert the total disturbance observation Introduced into the control law, it can be expressed as:

[0043]

[0044] Among them, k p and k d is the control gain; r and are the given reference input and its derivative function respectively.

[0045] When the system is converted from passive anti-disturbance to active anti-disturbance, the system can be simplified to a double integrator series structure. On this basis, the control law is reconstructed and -k d z′2 replaces the traditional PD control law In order to eliminate the system oscillation phenomenon caused by the rapid change of the system given signal, the feedback linear error oscillation suppression rule (FLE-OSR) can be obtained to meet the following requirements:

[0046]

[0047] Furthermore, the parameter design principles of the suppression rules are as follows:

[0048] Placing the second-order closed-loop pole of the controller at the controller bandwidth, the controller gain is solved as:

[0049]

[0050] Among them, ω cis the controller bandwidth; ξ is the damping ratio of the closed-loop system, which reflects the ability to suppress oscillation. Its value range is (0, 1]. The larger its value is, the stronger the system damping is. The smaller its value is, the stronger the system's rapid response capability is.

[0051] A grid-connected inverter circulating current estimation and suppression device based on a virtual synchronous machine includes:

[0052] The dynamic model acquisition module of the dual parallel grid-connected inverter interface converter is used to construct the active distribution network photovoltaic storage area cluster structure containing the dual parallel grid-connected inverter interface converter and the hybrid energy storage device, and derive the dynamic model of the dual parallel grid-connected inverter interface converter;

[0053] The extended state observer acquisition module is used to extract the closed-loop transfer function of the output power and the active power reference value, the actual output active power and the system frequency, and perform the corresponding virtual synchronous mechanism network scheme self-disturbance rejection paradigm transformation, thereby forming an extended state observer with a loop oscillation component estimation function;

[0054] The loop oscillation suppression strategy building module is used to introduce the total disturbance observation into the control law based on the observation results of the generalized disturbance reconstruction state by the model-free observer with oscillation estimation, so that the system can be converted from a passive anti-disturbance state to an active anti-disturbance state, and the model can be simplified to a double integrator series structure; it is also necessary to determine the parameter setting rules of the loop oscillation suppression strategy through pole configuration, integrate the loop oscillation suppression strategy and the parameter setting rules, deal with the influence of loop oscillation, and improve the power quality of the grid-type photovoltaic storage cluster.

[0055] A computing device comprising:

[0056] one or more processing units;

[0057] A storage unit for storing one or more programs,

[0058] Wherein, when the one or more programs are executed by the one or more processing units, the one or more processing units execute the above-mentioned method for estimating and suppressing circulating current of a grid-connected inverter based on a virtual synchronous machine.

[0059] A computer-readable storage medium having a non-volatile program code executable by a processor, wherein when the computer program is executed by the processor, the steps of the method for estimating and suppressing circulating current of a grid-connected inverter based on a virtual synchronous machine are implemented.

[0060] The advantages and positive effects of the present invention are:

[0061] The present invention constructs a dual parallel grid-connected inverter interface converter model and innovatively introduces the self-disturbance rejection paradigm transformation of the virtual synchronous mechanism network scheme, designs an extended state observer with a loop oscillation component estimation function, and realizes accurate observation of the circulating current; by introducing the total disturbance observation into the control law and simplifying it into a dual integrator series structure, the system realizes the conversion from passive interference rejection to active interference rejection, and achieves a -200dB suppression effect within the low-frequency disturbance range below 10Hz, and avoids the peak amplification phenomenon of the traditional linear self-disturbance rejection control scheme in the frequency range of 10-200Hz; at the same time, the parameter setting rules determined by the pole configuration further enhance the system adaptability. In a typical case, the circulating current oscillation of the parallel inverter can be reduced from 4.5A to below 0.4A within 1-2 cycles, and the output current harmonic distortion rate is only 0.66%, which effectively improves the power quality of the grid-type photovoltaic storage cluster. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] The technical solution of the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments, but it should be understood that these drawings are designed only for explanation purposes and are not intended to limit the scope of the present invention. In addition, unless otherwise specified, these drawings are intended only to conceptually illustrate the structural configurations described herein and are not necessarily drawn to scale.

[0063] Figure 1 An active power distribution network photovoltaic storage area cluster structure containing dual parallel grid-connected inverter interface converters provided by an embodiment of the present invention;

[0064] Figure 2 is a controlled object model provided by an embodiment of the present invention;

[0065] Figure 3 It is a comparison of frequency response curves in the complex plane between the model-free information extended state observer with oscillation estimation (improved solution) and the TLESO observer (conventional solution) provided in an embodiment of the present invention;

[0066] Figure 4 It is an equivalent model structure of a closed-loop control system of a circulating current self-disturbance rejection oscillation suppression strategy provided by an embodiment of the present invention;

[0067] Figure 5 Comparison of amplitude-frequency curves of the circulating current auto-disturbance rejection oscillation suppression solution (improved solution) and the linear auto-disturbance rejection solution (conventional solution) provided in an embodiment of the present invention;

[0068] Figure 6 is the output voltage waveform of the measurement phase provided by the embodiment of the present invention;

[0069] Figure 7 is the output current waveform of the measurement phase provided by the embodiment of the present invention;

[0070] Figure 8 It is an active distribution network experimental platform provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0071] In order to make the purpose, technical scheme and advantages of the implementation of the present invention clearer, the technical scheme in the embodiment of the present invention will be described in more detail below in conjunction with the drawings in the embodiment of the present invention. In the drawings, the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The described embodiments are part of the embodiments of the present invention, not all of the embodiments. The embodiments described below with reference to the drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limitations on the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0072] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0073] Example 1

[0074] This embodiment provides a method for estimating and suppressing circulating current of a grid-connected inverter based on a virtual synchronous machine, comprising the following steps:

[0075] Construct an active distribution network photovoltaic storage area cluster structure containing dual parallel grid-connected inverter interface converters and hybrid energy storage devices, and derive a dynamic model of dual parallel grid-connected inverter interface converters;

[0076] Extract the closed-loop transfer function of output power and active power reference value, actual output active power and system frequency, and perform corresponding self-disturbance rejection paradigm transformation of virtual synchronous mechanism network scheme, so as to form an extended state observer with loop oscillation component estimation function;

[0077] Construct a loop oscillation suppression strategy: Based on the observation results of the generalized disturbance reconstruction state by the model-free observer with oscillation estimation, the total disturbance observable is introduced into the control law, so that the system is converted from a passive anti-disturbance state to an active anti-disturbance state, and the model is simplified to a double integrator series structure; it is also necessary to determine the parameter setting rules of the loop oscillation suppression strategy through pole configuration, integrate the loop oscillation suppression strategy and the parameter setting rules, deal with the impact of loop oscillation, and improve the power quality of the grid-type photovoltaic storage cluster.

[0078] Specifically, an active distribution network photovoltaic storage area cluster structure containing dual parallel grid-connected inverter interface converters and hybrid energy storage devices is constructed, such as Figure 1 As shown, the voltage, current and impedance relationship of the dual parallel grid-connected inverter interface converter can be expressed as follows:

[0079]

[0080] In the formula, E k,i is the output fundamental voltage amplitude of the i-th grid-connected inverter under no-load condition; I out,i is the independent current of the grid-connected inverter interface converter, is the total parallel current; T out,i is the corresponding equivalent output impedance, T line,i is the line impedance, T total,i is the total impedance; Z kl is the load impedance, U kl is the load impedance voltage.

[0081] When deriving the dynamic model of the dual parallel grid-connected inverter interface converter, the independent currents of the dual parallel grid-connected inverter interface converter are decomposed as follows:

[0082]

[0083] Among them, I loop is the loop oscillation current caused by the relative voltage drop of the inverter; I load is the load current. I loop Oscillation interaction between dual parallel grid-connected inverter interface converters occurs when the distributed photovoltaic power generation carrying capacity at the dual parallel grid-connected inverter interface converters is unbalanced.

[0084] The closed-loop transfer function expressions of output power, active power reference value, actual output active power and system frequency are as follows:

[0085] In order to suppress the oscillation phenomenon and avoid the introduction of harmonic components, it is necessary to effectively control the dual parallel grid-connected inverter interface converter with filtering link. Output power and active power reference value P of the virtual synchronous mechanism network scheme ref , actual output active power P out and system frequency ω n The closed-loop transfer function expression is as follows:

[0086]

[0087] Where, J is the simulated moment of inertia; D is the damping coefficient; k f is the droop coefficient; τ is the filter absorption coefficient; P ref is the active power reference value; P out is the actual output active power; n is the system frequency; H1 and H2 are active power reference values ​​P ref And the actual output active power P out The pre-transfer function of .

[0088] The corresponding virtual synchronous mechanism network scheme self-disturbance rejection paradigm is as follows:

[0089] Based on the complex domain transfer function in (3), the time domain form of the active power-frequency control model can be obtained through inverse Laplace transform:

[0090]

[0091] To simplify the expression, the system state variable matrix is ​​defined as The state space expression paradigm of the active power-frequency control loop based on the virtual synchronous mechanism network scheme can be established by formula (4):

[0092]

[0093] In the formula, x1 and x2 are system state variables; u is the control state variable, satisfying u=ω n ; b0 is the control gain, satisfying b0=(D+k f ) -1 Jτ.

[0094] The extended state observer method with loop oscillation component estimation function is as follows:

[0095] Based on (5), all non-ideal components of the package are reconstructed into expanded state variables x3, and the expanded state space expression paradigm is:

[0096]

[0097] Accordingly, the dynamic equation of the model-free information observer with oscillation estimation is designed as:

[0098]

[0099] Among them, b′=[β1′β2′β3′] T is the observation parameter vector; the vector [z1′z2′z3′] is the reconstructed observation state matrix, which represents the observed values ​​of the system's first-order state variable x1, second-order state variable x2, and reconstructed extended state variable x3; A and B are the system matrix and control matrix, respectively. On this basis, the reconstructed variable x3 can be written as follows:

[0100]

[0101] In the formula, ψ m Modeling deviation for non-idealities; ψ l is the internal circulation oscillation disturbance; ψ d For external disturbance.

[0102] The parameter design principles of the model-free information observer are as follows:

[0103] In order to ensure the convergence of (6), the vector b′ must be properly configured so that the matrix A is a Hurwitz matrix. The pole placement method is used to adjust the observer bandwidth, and the characteristic roots of the extended state observer with the loop oscillation component estimation function are moved to the observation bandwidth. The parameter placement rule is written as:

[0104] sI-A m |=(s+ω o ) 3 , (twenty two)

[0105] Calculated

[0106]

[0107] Where I is the identity matrix; ω o is the observation bandwidth; s represents the complex domain independent variable in the Laplace transform.

[0108] The loop oscillation suppression strategy is designed as follows:

[0109] Based on the observation of the generalized disturbance reconstruction state by the model-free observer with oscillation estimation, a loop oscillation suppression strategy is designed to convert the total disturbance observation Introduced into the control law, it can be expressed as:

[0110]

[0111] Among them, k p and k d is the control gain; r and are the given reference input and its derivative function respectively.

[0112] When the system is converted from passive anti-disturbance to active anti-disturbance, the system can be simplified to a double integrator series structure. On this basis, the control law is reconstructed and -k d z′2 replaces the traditional PD control law In order to eliminate the system oscillation phenomenon caused by the rapid change of the system given signal, the feedback linear error oscillation suppression rule (FLE-OSR) can be obtained to meet the following requirements:

[0113]

[0114] The parameter design principles of suppression rules are as follows:

[0115] Placing the second-order closed-loop pole of the controller at the controller bandwidth, the controller gain is solved as:

[0116] k p =ω c 2 ,k d =2ξω c (26)

[0117] Among them, ω c is the controller bandwidth; ξ is the damping ratio of the closed-loop system, which reflects the ability to suppress oscillation. Its value range is (0, 1]. The larger its value is, the stronger the system damping is. The smaller its value is, the stronger the system's rapid response capability is.

[0118] Frequency domain response capability analysis from the perspective of disturbance observation

[0119] From (7), we can get the reconstructed variable x3 including all non-ideal components of the system as:

[0120]

[0121] Through Laplace transform, the time domain form of (13) can be transformed into the following complex frequency domain form:

[0122] X3(s)+b0U(s)=s 2 X1(s). (28)

[0123] Solving the differential equations in (6), we can obtain the equivalent representation of the total disturbance estimate Z3(s) in the conventional linear extended state observer and the total disturbance estimate Z3′ in the model-free information extended state observer with oscillation estimation as follows:

[0124]

[0125] Furthermore, the equivalent representation of the estimated values ​​of each order state variables in the model-free information extended state observer with oscillation estimation in the frequency domain can be obtained:

[0126]

[0127] Combining the above formulas, when using the linear extended state observer and the model-free information extended state observer with oscillation estimation for disturbance observation, the transfer models G1(s) and G2(s) between the generalized external disturbance and the extracted disturbance value are respectively:

[0128]

[0129] By comparing the total disturbance response functions of the two observers, it can be seen that by correcting the highest-order error feedback part in the linear extended state observer, a pair of poles and zeros are added to the total disturbance response function, and their positions are directly related to the observer bandwidth ω0.

[0130] As an example, in this embodiment, in order to intuitively compare the disturbance observation capabilities of the model-free extended state observer with oscillation estimation (MLESO) and the linear extended state observer (TLESO), the attached Figure 3 Discussed ino Frequency response curves of the two observers in the complex plane when =250.

[0131] It can be seen that both TLESO and the model-free information extended state observer with oscillation estimation in this embodiment have no encircling point (-1, j0), which means that the system remains stable. Among them, the Nyquist curve of TLESO presents a smaller closed trajectory, and the curve is relatively flat and compact. The observation gain fluctuates in a lower range, indicating that TLESO has a large signal attenuation in the observation process of the total disturbance.

[0132] In addition, the model-free information extended state observer with oscillation estimation is more extended than TLESO, especially showing higher gain in the intermediate frequency part where the oscillation disturbance appears. The peak response index is improved by about 1.35dB from the original -6.1dB, and the curve has a larger offset near the real axis, showing a more sensitive dynamic response to frequency changes.

[0133] The improved model-free extended state observer with oscillation estimation is superior to the TLESO observer in terms of dynamic response speed and peak response capability in the mid-frequency band. In addition, this characteristic can be reproduced under any parameters within the stable range and is not limited to a specific parameter condition.

[0134] From the transfer function between the actual disturbance value and the disturbance estimation value, it can be seen that the transfer functions used to describe the disturbance estimation deviation in the TLESO scheme and the grid-connected inverter circulating current estimation and suppression scheme based on virtual synchronous machine self-disturbance rejection paradigm transformation are:

[0135]

[0136] Different oscillation disturbance functions x3 are selected to compare the TLESO and the grid-connected inverter circulating current estimation scheme of the distributed photovoltaic storage cluster based on the virtual synchronous machine self-disturbance rejection paradigm transformation. The response results are shown in the two right columns of Table 1.

[0137] Table 1 Comparison of inverter circulating current response results caused by oscillation disturbance function x3 excitation of different powers

[0138]

[0139] In the table, the polynomial satisfies f en (t) = k e0 t 0 +k e1 t 1 +…+k en t n (n≤5); n is the power of the polynomial; k e0 ~k en are the coefficients of each power polynomial; fen (t) is a polynomial expression; k is the polynomial proportional coefficient.

[0140] As shown in Table 1, when the generalized disturbance x3 is an oscillation type of power or less, BCR-ESO can effectively observe the disturbance. However, when TLESO faces a power oscillation disturbance, a continuous static error kω will occur. o -1 , which manifests as non-ideal harmonic components.

[0141] When the generalized disturbance x3 is an oscillation type of quadratic power or above, both schemes will have continuous deviations during the disturbance observation process. However, the grid-connected inverter circulating current estimation scheme for clusters of distributed photovoltaic storage stations based on the virtual synchronous machine self-disturbance rejection paradigm transformation has a higher accuracy in extracting oscillations. TLESO only has conventional disturbance description states and cannot effectively compensate for high-order power oscillation disturbances. Therefore, the observation error of the circulating current oscillation component continues to increase during the operation of the grid-connected inverter, which will lead to poor output power quality. Therefore, during the cluster operation process, the grid-connected inverter circulating current estimation scheme for clusters of distributed photovoltaic storage stations based on the virtual synchronous machine self-disturbance rejection paradigm transformation has more accurate observation results for disturbances than TLESO.

[0142] According to the state space expression of the model-free information extended state observer with oscillation estimation and the output of the active disturbance rejection control law, it can be derived that:

[0143]

[0144] Among them, R(s) is the complex domain representation of the reference signal; Y(s) is the complex domain representation of the output; U(s) is the complex domain representation of the control quantity; G r (s) is the equivalent transfer function of the reference signal transmission channel; G u (s) is the equivalent transfer function of the control signal transmission channel; H(s) is the equivalent transfer function of the feedback signal transmission channel, which specifically satisfies the following:

[0145]

[0146] Among them, G p (s) is a first-order integrator model. It can be seen that the complex domain expression from the reference signal to the control quantity is:

[0147]

[0148] The corresponding poles and zeros are:

[0149]

[0150] Among them, G ff(s) is the complex domain expression transfer function from the reference signal to the control quantity; is a pair of multiple root poles; For the pole of independence; is the zero point for three replicates.

[0151] It can be seen that for the proposed grid-connected inverter circulating current estimation and suppression scheme based on virtual synchronous machine self-disturbance rejection paradigm transformation, the real parts of the poles and zeros of the open-loop transfer function from the reference signal to the control quantity are less than or equal to zero, so the minimum phase system G ff (s) is reversible. The equivalent simplified form of the obtained closed-loop control system is shown in the attached figure. Figure 4 shown.

[0152] Based on this, the transfer function of the closed-loop system using the circulating current estimation and suppression scheme can be derived as:

[0153]

[0154] Among them, R eq,Ⅰ (s) is the complex domain form of the reference signal equivalent interference; F eq,Ⅱ (s) is the complex domain form of the equivalent interference of the control signal; Y F (s) is the overall transfer function of the closed-loop system; Y F,Ⅰ (s) and Y F,Ⅱ (s) are the decomposition functions of the reference signal equivalent interference and the control signal equivalent interference, which can be written as follows:

[0155]

[0156]

[0157] It can be seen that the output of the closed-loop control system consists of a disturbance term and a tracking term. According to the superposition theorem and the final value theorem, the proposed grid-connected inverter circulating current estimation and suppression scheme based on the virtual synchronous machine self-disturbance rejection paradigm transformation satisfies the dynamic final value of the total disturbance response:

[0158]

[0159] Under the premise of interference term convergence, the proposed scheme for estimating and suppressing circulating current of grid-connected inverters based on the paradigm shift of virtual synchronous machine anti-disturbance can realize the tracking of reference signal without overshoot. At this time, the ability of the system to suppress interference term is the main factor affecting the performance of circulating current estimation and suppression.

[0160] Attached Figure 5 The range of the amplitude-frequency curve of the transfer function between the disturbance and the output is shown. The curve intuitively shows that when the observer bandwidth ω oWhen the disturbance suppression characteristics of RELADRC and TLADRC control systems vary widely within the range of 300 to 600, they are poor.

[0161] When o =300, when the grid-connected inverter introduces low-frequency disturbances (below 10Hz) caused by the slow-changing behavior of the load, it can be seen from the frequency sweep results that the negative gain range of the traditional TLADRC controller to the disturbance is -80 to -70dB, while the negative gain of the grid-connected inverter circulating current estimation and suppression scheme based on the virtual synchronous machine self-disturbance rejection paradigm transformation is -200dB, showing a low-frequency disturbance suppression capability far higher than that of TLADRC.

[0162] At frequencies around 10-200Hz, TLADRC shows a spike amplification phenomenon, which means that the traditional disturbance suppression scheme cannot effectively suppress the mid-frequency disturbance caused by control dynamic response, system load mutation, bus voltage fluctuation, etc. o The peak amplification phenomenon of TLADRC is alleviated when the power factor is increased to 600. In comparison, the equivalent output impedance of the grid inverter circulating current estimation and suppression scheme based on the virtual synchronous machine self-disturbance rejection paradigm transformation under the same bandwidth is lower. In the high-frequency noise disturbance range greater than 1kHz, the disturbance suppression capabilities of the two control systems are similar. At this time, the equivalent output impedance is mainly affected by ω o Size restrictions.

[0163] Compared with the TLADRC scheme, the newly added poles of the grid-connected inverter circulating current estimation and suppression scheme based on the virtual synchronous machine self-disturbance rejection paradigm transformation occupy a better longitudinal center position, providing good damping support for the tracking system. o Under the amplification condition, the equivalent output impedance of the grid-connected inverter circulating current estimation and suppression scheme based on the virtual synchronous machine self-disturbance rejection paradigm transformation is lower, which means that the external disturbance is greatly weakened when it is transmitted to the output in the system, and the output of the grid-connected inverter circulating current estimation and suppression scheme based on the virtual synchronous machine self-disturbance rejection paradigm transformation will not be significantly affected by the disturbance.

[0164] In order to verify the feasibility of the proposed scheme for estimating and suppressing circulating current of grid-connected inverters based on the virtual synchronous machine self-disturbance rejection paradigm transformation, an active distribution network experimental platform was built based on the RT-LAB simulation platform (see Appendix). Figure 8 ), in the experiment, the DC side voltage was set to Udc=700V, the grid side bus voltage was 220V, and the fundamental frequency was 50Hz. The control parameters were set by trial and error based on the loop method. The virtual synchronous machine's filter inductance was L=4mH, parasitic resistance r=0.015Ω, filter capacitor C=50μF, and switching frequency was 10kHz. The line impedance was Z=1+j0.1. In all subsequent experiments, the switching frequency was controlled to 50kHz.

[0165] Taking the control process of a dual-parallel photovoltaic grid-connected inverter as an example, the reference voltage parameters of the parallel inverter interface and the line impedance parameters are kept consistent, the maximum power point voltage of the photovoltaic array is set to 250V, the ambient temperature is 25 degrees Celsius, the virtual moment of inertia is 0.2kg·m2, the virtual damping coefficient is 10W / (rad / s), and the output voltage and current waveforms of the phases are measured (see Appendix Figure 5 , Attachment Figure 6 ).

[0166] It can be seen that within 1 to 2 cycles, the closed-loop control system of the grid-connected inverter circulating current estimation and suppression scheme based on the virtual synchronous machine self-disturbance rejection paradigm transformation achieves the synchronization of voltage and current (about 0.03s). After the transition process, the maximum value of the output voltage remains near 311V, which meets the expected control target of the steady-state voltage. At this time, the harmonic distortion rate of the load-side output current is 0.66%, which is far lower than the standard value requirement of 0.05 and meets the power quality requirements.

[0167] As the harmonic order increases, the content of high-order harmonics decreases. In the initial synchronization stage, the oscillation current between parallel inverters surges, and the peak value of the circulating current is 4.5A, which is relatively large. After introducing the grid-connected inverter circulating current estimation and suppression scheme based on virtual synchronous machine self-disturbance rejection paradigm transformation, after a transition process of 1 to 2 cycles, the circulating current oscillation gradually stabilizes, and the oscillation peak is controlled below 0.4A.

[0168] In summary, this embodiment proposes a scheme for estimating and suppressing the circulating current of a grid-connected inverter based on the self-disturbance rejection paradigm transformation of a virtual synchronous machine. This method can observe the oscillation disturbance of the circulating current of the grid-connected inverter, and suppress generalized disturbances including oscillation disturbances, non-ideal modeling deviations and external disturbances. First, a model-free information ESO with oscillation estimation is designed for generalized disturbances during the operation of the grid-connected inverter. Secondly, based on the observation results of the generalized disturbance reconstruction state by the model-free observer with oscillation estimation, the total disturbance observation is introduced into the control law to design a loop oscillation suppression strategy, so that the system is converted from a passive anti-disturbance state to an active anti-disturbance state, and the model is simplified to a dual-integrator series structure. On this basis, -k d z′2 replaces the Thus, the control law is reconstructed to eliminate the system oscillation phenomenon caused by the rapid change of the system given signal, and the feedback linear error type oscillation suppression rule is obtained. In addition, the parameter configuration method of the suppression rule is designed to enhance the adaptability of the circulating current estimation and suppression method to different working conditions.

[0169] The theoretical analysis process, such as the frequency domain response capability analysis from the perspective of disturbance observation and the generalized disturbance suppression capability analysis of the equivalent input model, proves the effectiveness of the circulating current estimation and suppression scheme proposed in this embodiment. In addition, a test platform for the circulating current estimation and suppression scheme of the grid-connected inverter based on the self-disturbance rejection paradigm transformation of the virtual synchronous machine was built, and multi-condition simulation experiments with parameter perturbation / non-perturbation were carried out to verify the anti-disturbance capability of the circulating current estimation and suppression scheme. Theoretical analysis and experimental results show that the circulating current estimation and suppression scheme of the grid-connected inverter based on the self-disturbance rejection paradigm transformation of the virtual synchronous machine proposed in this embodiment is more suitable for the operation environment of the active distribution network photovoltaic storage cluster containing complex unknown disturbances.

[0170] Example 2

[0171] The embodiment of the invention is a device for estimating and suppressing circulating current of a grid-connected inverter based on a virtual synchronous machine, comprising:

[0172] The dynamic model acquisition module of the dual parallel grid-connected inverter interface converter is used to construct the active distribution network photovoltaic storage area cluster structure containing the dual parallel grid-connected inverter interface converter and the hybrid energy storage device, and derive the dynamic model of the dual parallel grid-connected inverter interface converter;

[0173] The extended state observer acquisition module is used to extract the closed-loop transfer function of the output power and the active power reference value, the actual output active power and the system frequency, and perform the corresponding virtual synchronous mechanism network scheme self-disturbance rejection paradigm transformation, thereby forming an extended state observer with a loop oscillation component estimation function;

[0174] The loop oscillation suppression strategy building module is used to introduce the total disturbance observation into the control law based on the observation results of the generalized disturbance reconstruction state by the model-free observer with oscillation estimation, so that the system can be converted from a passive anti-disturbance state to an active anti-disturbance state, and the model can be simplified to a double integrator series structure; it is also necessary to determine the parameter setting rules of the loop oscillation suppression strategy through pole configuration, integrate the loop oscillation suppression strategy and the parameter setting rules, deal with the influence of loop oscillation, and improve the power quality of the grid-type photovoltaic storage cluster.

[0175] A computing device is also provided, comprising:

[0176] one or more processing units;

[0177] A storage unit for storing one or more programs,

[0178] Among them, when the one or more programs are executed by the one or more processing units, the one or more processing units execute the grid-connected inverter circulating current estimation and suppression method based on the virtual synchronous machine in this embodiment; it should be noted that the computing device may include but is not limited to a processing unit and a storage unit; those skilled in the art can understand that the computing device including a processing unit and a storage unit does not constitute a limitation on the computing device, and may include more components, or a combination of certain components, or different components. For example, the computing device may also include input and output devices, network access devices, buses, etc.

[0179] A computer-readable storage medium having a non-volatile program code executable by a processor, wherein when the computer program is executed by the processor, the steps of the method for estimating and suppressing the circulating current of a grid-connected inverter based on a virtual synchronous machine in the present embodiment are implemented; it should be noted that the readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof; the program contained on the readable medium may be transmitted using any appropriate medium, including, but not limited to, wireless, wired, optical cable, RF, etc., or any suitable combination thereof. For example, the program code for performing the operation of the present invention may be written in any combination of one or more programming languages, wherein the programming language includes an object-oriented programming language, such as Java, C++, etc., and also includes a conventional procedural programming language, such as C language or a similar programming language. The program code may be executed entirely on a user computing device, partially on a user device, as an independent software package, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0180] Finally, it should be pointed out that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Although the present invention has been described in detail with reference to the above embodiments, a person skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for estimating and suppressing circulating current of a grid-connected inverter based on a virtual synchronous machine, characterized in that: The steps include: Construct an active distribution network photovoltaic storage area cluster structure containing dual parallel grid-connected inverter interface converters and hybrid energy storage devices, and derive a dynamic model of dual parallel grid-connected inverter interface converters; Extract the closed-loop transfer function of output power and active power reference value, actual output active power and system frequency, and perform corresponding self-disturbance rejection paradigm transformation of virtual synchronous mechanism network scheme, so as to form an extended state observer with loop oscillation component estimation function; Construct a loop oscillation suppression strategy: Based on the observation results of the generalized disturbance reconstruction state by the model-free observer with oscillation estimation, the total disturbance observable is introduced into the control law, so that the system is converted from a passive anti-disturbance state to an active anti-disturbance state, and the model is simplified to a double integrator series structure; it is also necessary to determine the parameter setting rules of the loop oscillation suppression strategy through pole configuration, integrate the loop oscillation suppression strategy and the parameter setting rules, deal with the impact of loop oscillation, and improve the power quality of the grid-type photovoltaic storage cluster.

2. The method for estimating and suppressing circulating current of a grid-connected inverter based on a virtual synchronous machine according to claim 1, characterized in that: The method for constructing a cluster structure of an active distribution network photovoltaic storage area containing dual parallel grid-connected inverter interface converters and a hybrid energy storage device is as follows: In the formula, E k,i is the output fundamental voltage amplitude of the i-th grid-connected inverter under no-load condition; I out,1 and I out,2 are the independent currents of two grid-connected inverter interface converters, is the total parallel current; T out,1 and T out,2 are the equivalent output impedances of the two grid-connected inverter interface converters; T line,1 and T line,1 are the respective line impedances; T total,1 and T total,2 are their respective total impedance values; Z kl is the load impedance, U kl is the load impedance voltage.

3. The method for estimating and suppressing circulating current of a grid-connected inverter based on a virtual synchronous machine according to claim 2, characterized in that: When deriving the dynamic model of the dual parallel grid-connected inverter interface converter, the independent currents of the dual parallel grid-connected inverter interface converter are decomposed as follows: Among them, I loop is the loop oscillation current caused by the relative voltage drop of the inverter; I load is the load current. When the system is running, I loop Oscillation interaction between dual parallel grid-connected inverter interface converters occurs when the distributed photovoltaic power generation carrying capacity at the dual parallel grid-connected inverter interface converters is unbalanced.

4. The method for estimating and suppressing circulating current of a grid-connected inverter based on a virtual synchronous machine according to claim 1, characterized in that: The closed-loop transfer function expressions of output power, active power reference value, actual output active power and system frequency are as follows: Output power and active power reference value P of virtual synchronous mechanism network solution ref , actual output active power P out The closed-loop transfer function expression of and system frequency ω is as follows: Where, J is the simulated moment of inertia; D is the damping coefficient; k f is the droop coefficient; τ is the filter absorption coefficient; P ref is the active power reference value; P out is the actual output active power; ω is the system frequency; ω n is the rated frequency; H1 and H2 are active power reference values ​​P ref And the actual output active power P out The pre-transfer function of .

5. The method for estimating and suppressing circulating current of a grid-connected inverter based on a virtual synchronous machine according to claim 4, characterized in that: The corresponding virtual synchronous mechanism network scheme self-disturbance rejection paradigm is as follows: Based on the complex domain transfer function in (3), the time domain form of the active power-frequency control model can be obtained through inverse Laplace transform: To simplify the expression, the system state variable matrix is ​​defined as The state space expression paradigm of the active power-frequency control loop based on the virtual synchronous mechanism network scheme can be established by formula (4): In the formula, x1 and x2 are system state variables; u is the control state variable, satisfying u=ω n ; b0 is the control gain, satisfying b0=(D+k f ) -1 Jτ.

6. The method for estimating and suppressing circulating current of a grid-connected inverter based on a virtual synchronous machine according to claim 5, characterized in that: The extended state observer method with loop oscillation component estimation function is as follows: Based on (5), all non-ideal components are reconstructed into expanded state variables x3, and the expanded state space expression paradigm is: On the basis of (5), all non-ideal components of the package are reconstructed into extended state variables x3, and the dynamic equation of the model-free information observer with oscillation estimation is designed as follows: Among them, b′=[β1′β2′β3′] T is the observation parameter vector; the vector [z1′z2′z3′] is the reconstructed observation state matrix, which is the observed values ​​of the first-order state variable x1, the second-order state variable x2 and the reconstructed extended state variable x3 of the system; A and B are the system matrix and the control matrix respectively; on this basis, the reconstructed variable x3 can be written as follows: In the formula, ψ m Modeling deviation for non-idealities; ψ l is the internal circulation oscillation disturbance; ψ d is an external disturbance; The parameter design principles of the model-free information observer are as follows: In order to ensure the convergence of (6), the vector b′ must be properly configured so that the matrix A is a Hurwitz matrix. The pole placement method is used to adjust the observer bandwidth, and the characteristic roots of the extended state observer with the function of estimating the loop oscillation component are moved to the observation bandwidth. The parameter placement rule is written as: |sI-A|=(s+ω o ) 3 , (9) Calculated: Where I is the identity matrix; ω o is the observation bandwidth; s represents the complex domain independent variable in the Laplace transform.

7. The method for estimating and suppressing circulating current of a grid-connected inverter based on a virtual synchronous machine according to claim 6, characterized in that: The loop oscillation suppression strategy is designed as follows: Based on the observation of the generalized disturbance reconstruction state by the model-free observer with oscillation estimation, a loop oscillation suppression strategy is designed to convert the total disturbance observation Introduced into the control law, it can be expressed as: Among them, k p and k d is the control gain; r and are the given reference input and its derivative function respectively; When the system is converted from passive anti-disturbance to active anti-disturbance, the system can be simplified to a double integrator series structure. On this basis, the control law is reconstructed and -k d z′2 replaces the traditional PD control law In order to eliminate the system oscillation phenomenon caused by the rapid change of the system given signal, the feedback linear error oscillation suppression rule (FLE-OSR) can be obtained to meet the following requirements:

8. The method for estimating and suppressing circulating current of a grid-connected inverter based on a virtual synchronous machine according to claim 7, characterized in that: The parameter design principles of suppression rules are as follows: Placing the second-order closed-loop pole of the controller at the controller bandwidth, the controller gain is solved as: k p =ω c 2 ,k d =2 out c (13) Among them, ω c is the controller bandwidth; ξ is the damping ratio of the closed-loop system, which reflects the ability to suppress oscillation. Its value range is (0, 1]. The larger its value is, the stronger the system damping is. The smaller its value is, the stronger the system's rapid response capability is.

9. A device for estimating and suppressing circulating current of a grid-connected inverter based on a virtual synchronous machine, characterized in that: include: The dynamic model acquisition module of the dual parallel grid-connected inverter interface converter is used to construct the active distribution network photovoltaic storage area cluster structure containing the dual parallel grid-connected inverter interface converter and the hybrid energy storage device, and derive the dynamic model of the dual parallel grid-connected inverter interface converter; The extended state observer acquisition module is used to extract the closed-loop transfer function of the output power and the active power reference value, the actual output active power and the system frequency, and perform the corresponding virtual synchronous mechanism network scheme self-disturbance rejection paradigm transformation, thereby forming an extended state observer with a loop oscillation component estimation function; The loop oscillation suppression strategy building module is used to introduce the total disturbance observation into the control law based on the observation results of the generalized disturbance reconstruction state by the model-free observer with oscillation estimation, so that the system can be converted from a passive anti-disturbance state to an active anti-disturbance state, and the model can be simplified to a double integrator series structure; it is also necessary to determine the parameter setting rules of the loop oscillation suppression strategy through pole configuration, integrate the loop oscillation suppression strategy and the parameter setting rules, deal with the influence of loop oscillation, and improve the power quality of the grid-type photovoltaic storage cluster.

10. A computing device, characterized in that: include: one or more processing units; A storage unit for storing one or more programs, When the one or more programs are executed by the one or more processing units, the one or more processing units execute the method as claimed in any one of claims 1 to 8.

Images