Network-constructed converter impedance real-time reconstruction method, system, device and storage medium

Abstract

The application discloses a network-constructing type converter impedance real-time reconstruction method, system, device and storage medium, and comprises the following steps: establishing a network-constructing type VSG sequence impedance model; combining the change of grid strength and impedance analysis to obtain a network-constructing type VSG grid-connected open-loop transfer function containing grid strength, converting the network-constructing type VSG grid-connected open-loop transfer function from the S domain to the frequency domain, and judging the network-constructing type VSG grid-connected stability; analyzing the network-constructing type VSG grid-connected open-loop transfer function in the frequency domain by using an impedance sensitivity analysis method to obtain the sensitivity of each control parameter to impedance remodeling; optimizing the network-constructing type VSG grid-connected open-loop transfer function; and based on the sensitivity of each control parameter to impedance remodeling, controlling the pollen population generation direction of the improved pollen algorithm to real-time reconstruct the network-constructing type VSG impedance. The application effectively improves the stability of the network-constructing type VSG grid connection under the background of real-time change of grid strength, and has important significance for safe and stable operation of new energy grid connection.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of converter impedance modeling and resonance suppression, and particularly relates to a grid-connected converter impedance real-time reconstruction method, system, device and storage medium. BACKGROUND

[0002] With the development of new energy technologies such as wind power and photovoltaic power, large power electronic converters have become the core equipment for new energy grid connection. Grid-connected converters are widely used because of their voltage self-building capability. However, due to factors such as line switching, fault, load fluctuation and series compensation line, the equivalent input impedance of the power grid presents significant time-varying and uncertainty, which changes the impedance interaction characteristics between the VSG and the power system, leading to the occurrence of wide-frequency oscillation and affecting the stability of the power system.

[0003] For the grid-connected oscillation suppression problem of power electronic devices, existing researches mainly focus on optimizing control parameters under fixed working conditions and improving the internal control structure of the converter. Although these methods have significant effects under specific working conditions, they lack adaptability to real-time grid strength changes of the power system, i.e., short-circuit ratio changes. SUMMARY

[0004] To solve the above problems, the application provides a grid-connected converter impedance real-time reconstruction method, system, device and storage medium capable of enhancing the real-time stability of grid-connected converters.

[0005] To achieve the above purpose, the application is implemented by the following technical solutions:

[0006] The grid-connected converter impedance real-time reconstruction method provided by the application comprises the following steps:

[0007] establishing a grid-connected VSG serial impedance model;

[0008] based on the grid-connected VSG serial impedance model, combining grid strength changes and impedance analysis, obtaining a grid-connected VSG grid-connected open-loop transfer function containing grid strength, converting the grid-connected VSG grid-connected open-loop transfer function from the S domain to the frequency domain, and judging the grid-connected VSG grid-connected stability;

[0009] based on the unstable judgment result, using impedance sensitivity analysis method to analyze the grid-connected VSG grid-connected open-loop transfer function in the frequency domain, and obtaining the sensitivity of each control parameter to impedance remodeling;

[0010] using the improved pollen algorithm to optimize the grid-connected VSG grid-connected open-loop transfer function, controlling the pollen population generation direction of the improved pollen algorithm based on the sensitivity of each control parameter to impedance remodeling, and performing real-time reconstruction on the grid-connected VSG impedance.

[0011] The further improvement of the present application is to establish a network type VSG sequence impedance model, comprising:

[0012] Using the method of harmonic linearization, based on VSG power ring, voltage and current sampling delay, the following is constructed:

[0013] ;

[0014] For simplifying calculation, let:

[0015] ;

[0016] ;

[0017] ;

[0018] ;

[0019] In the formula: is the positive sequence impedance, is the negative sequence impedance, is the positive sequence disturbance current peak, is the negative sequence disturbance current peak, is the fundamental current peak, is the fundamental voltage peak, is the positive sequence disturbance voltage peak, is the negative sequence disturbance voltage peak, is the VSG internal potential amplitude, is the voltage ring PI controller transfer function, is the current sampling delay transfer function, , , are the fundamental, positive sequence and negative sequence current phase angles respectively, , are the positive sequence and negative sequence voltage phase angles respectively, , , are the LC filter filtering inductance, filtering capacitance and damping resistance respectively is the Laplace operator, is the power angle, represents the imaginary unit, is the counterclockwise rotation angle in complex space, is the active controller transfer function, is the rated angular frequency of the power grid, is the damping coefficient reference value, is the damping coefficient adjustable amount, is the virtual inertia reference value, is the virtual inertia adjustable amount.

[0020] Further improvement of the present application is that based on the network configuration type VSG sequence impedance model, the impedance analysis combined with the change of grid strength is used to obtain the grid strength containing network configuration type VSG grid-connected open-loop transfer function, the network configuration type VSG grid-connected open-loop transfer function is converted from S domain to frequency domain, and the network configuration type VSG grid-connected stability is judged, including:

[0021] The change of grid strength is characterized by short-circuit ratio, and the impedance analysis is combined to obtain the grid strength containing network configuration type VSG grid-connected open-loop transfer function, wherein the network configuration type VSG grid-connected open-loop transfer function includes positive sequence open-loop transfer function And negative sequence open-loop transfer function , is the grid strength, is the positive sequence impedance, is the negative sequence impedance, is the ratio of rated power to square of AC grid rated voltage;

[0022] The network configuration type VSG grid-connected open-loop transfer function is converted from S domain to frequency domain, and the generalized Nyquist criterion in frequency domain is used to judge the network configuration type VSG grid-connected stability.

[0023] Further improvement of the present application is that the control parameters include active control loop virtual inertia J, damping coefficient , PI controller proportional coefficient , PI controller integral coefficient , and the expression of the sensitivity of each control parameter to impedance remodeling is:

[0024] ;

[0025] ;

[0026] In the formula, , respectively, the real part and the imaginary part of the positive sequence open-loop transfer function are the sensitivity set of each control parameter at different frequencies, , respectively, the real part and the imaginary part of the negative sequence open-loop transfer function are the sensitivity set of each control parameter at different frequencies, , respectively, the real part and the imaginary part of the positive sequence open-loop transfer function in frequency domain, , respectively, the real part and the imaginary part of the negative sequence open-loop transfer function in frequency domain.

[0027] Further improvement of the present application is that the improved pollen algorithm is used to optimize the network configuration type VSG grid-connected open-loop transfer function, including:

[0028] The objective function of the improved pollen algorithm is determined, including:

[0029] The first path objective function is:

[0030]

[0031] wherein: is the first path objective function, is the real part constraint of the grid-forming VSG open-loop transfer function, is the real part of the grid-forming VSG grid-connected open-loop transfer function, is the frequency, is the set of frequencies with the real part less than -1;

[0032] The second path objective function is:

[0033]

[0034] wherein: is the second path objective function, is the imaginary part constraint of the grid-forming VSG grid-connected open-loop transfer function, is the integral quantity under the unstable bandwidth, is the imaginary part of the grid-forming VSG grid-connected open-loop transfer function, is the penalty coefficient, wherein: , are the frequencies corresponding to the real part of the positive sequence and negative sequence open-loop transfer functions being equal to -1, respectively, is the resonant frequency;

[0035] The fitness function of the improved pollen algorithm is:

[0036]

[0037] wherein: is the fitness function, , are the weights of the first path objective function and the second path objective function, respectively;

[0038] The constraint function of the improved pollen algorithm is determined as:

[0039]

[0040] wherein: , are the lower limit and upper limit of the adjustable quantity of the active control loop virtual inertia J, , is the damping coefficient of the adjustable quantity, , is the PI controller proportion coefficient​​​​ The lower and upper limits of the adjustable amount, , Integral coefficients of the PI controller The lower and upper limits of the adjustable amount, This is an adjustable virtual inertia value. This is the adjustable amount of the damping coefficient. The adjustable value of the proportional coefficient of the PI controller. The integral coefficient of the PI controller is adjustable.

[0041] A further improvement of this invention lies in: based on the sensitivity of each control parameter to impedance reshaping, controlling the pollen population generation direction of the improved pollen algorithm, and reconstructing the network-type VSG impedance in real time, including:

[0042] Based on the sensitivity of each control parameter to impedance reshaping at the current resonant frequency, calculate the sensitivity weighting factor for each control parameter:

[0043] ;

[0044] In the formula: For control parameters, , These are the control parameters. Sensitivity weighting factors under positive-order and negative-order open-loop transfer functions , These represent the sensitivities of each control parameter under positive-sequence and negative-sequence open-loop transfer functions, respectively. For control parameters quantity;

[0045] Based on the sensitivity weighting factors of each control parameter, the direction of pollen population generation is controlled, and the initial position sequence of pollen individuals within the search area is generated:

[0046] ;

[0047] In the formula: Initial location sequence of pollen individuals , The first pollen individuals Minimum and maximum values ​​in It is a chaotic sequence. For control parameters Sensitivity weighting factors for positive-order and negative-order open-loop transfer functions .

[0048] The real-time impedance reconfiguration system for grid-type converters of the present invention includes:

[0049] A model establishing module is configured to establish a grid-type VSG sequence impedance model;

[0050] A stability judging module is configured to obtain a grid-type VSG grid-connected open-loop transfer function containing grid strength based on the grid-type VSG sequence impedance model, in combination with grid strength change and impedance analysis, convert the grid-type VSG grid-connected open-loop transfer function from an S domain to a frequency domain, and judge grid-type VSG grid-connected stability;

[0051] An analysis module is configured to analyze the grid-type VSG grid-connected open-loop transfer function in the frequency domain based on the judging result by using an impedance sensitivity analysis method, and obtain sensitivity of each control parameter to impedance remodeling.

[0052] An optimization module is configured to optimize the grid-type VSG grid-connected open-loop transfer function by using an improved pollen algorithm.

[0053] A reconstruction module is configured to control a pollen population generation direction of the improved pollen algorithm based on the sensitivity of each control parameter to impedance remodeling, and perform real-time reconstruction on the grid-type VSG impedance.

[0054] The electronic device of the present application comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the above-mentioned grid-type converter impedance real-time reconstruction method when executing the computer program.

[0055] The computer readable storage medium of the present application stores a computer program, and the computer program implements the steps of the above-mentioned grid-type converter impedance real-time reconstruction method when executed by the processor.

[0056] The present application has the following advantages: the present application establishes a grid-type VSG sequence impedance model, and performs sensitivity analysis on the grid-type VSG sequence impedance model, and uses sensitivity to guide the population generation direction of the improved pollen algorithm, ensuring the accuracy of the improved pollen algorithm in reconstructing the converter impedance while taking into account the efficiency. The present application effectively improves the stability of the grid-type VSG grid connection under the background of real-time change of grid strength, and has important significance for safe and stable operation of new energy grid connection. BRIEF DESCRIPTION OF DRAWINGS

[0057] Figure 1 is the method flowchart in the embodiment of the present application;

[0058] Figure 2 is the grid-type VSG main circuit topology diagram in the embodiment of the present application;

[0059] Figure 3 is the grid-type VSG control block diagram in the embodiment of the present application;

[0060] Figure 4is a small-signal equivalent circuit diagram of a positive sequence of a network-form VSG in the embodiment of the application;

[0061] Figure 5 is a small-signal equivalent circuit diagram of a negative sequence of a network-form VSG in the embodiment of the application. DETAILED DESCRIPTION

[0062] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and should not be used to limit the present application. In addition, the technical features involved in each embodiment of the present application described below can be combined with each other as long as they do not conflict with each other.

[0063] As shown in Figure 1 , the network-form converter impedance real-time reconstruction method of the embodiment includes:

[0064] Step 1, a network-form VSG sequence impedance model based on VSG power ring, voltage and current sampling delay factors is established by a harmonic linearization method, wherein the VSG power ring includes an active control loop and a reactive control loop;

[0065] Step 2, based on the network-form VSG sequence impedance model, the change of short-circuit ratio is used to represent the change of grid strength, and the impedance analysis method is combined to obtain a network-form VSG grid-connected open-loop transfer function containing grid strength, the network-form VSG grid-connected open-loop transfer function is converted from S domain to frequency domain, and the stability of the network-form VSG grid connection is judged by using the generalized Nyquist criterion in the frequency domain;

[0066] Step 3, in the case that the network-form VSG grid connection has instability risk, the impedance sensitivity analysis method is used to analyze the network-form VSG grid-connected open-loop transfer function in the frequency domain, and the sensitivity of impedance remodeling of each control parameter at the current resonant frequency is obtained;

[0067] Step 4, the improved pollen algorithm is used to optimize the network-form VSG grid-connected open-loop transfer function;

[0068] Step 5, based on the sensitivity of each control parameter at the current resonant frequency obtained in step 3, the pollen population generation direction of the improved pollen algorithm is controlled, the real-time optimal adjustment amount of each control parameter is obtained, and the impedance of the network-form VSG is reconstructed in real time by adjusting each control parameter in real time.

[0069] The network-form VSG main circuit topology used in the embodiment is shown in Figure 2 , Figure 2 , is the DC side voltage of the network-form VSG, and is a stable value. 、 and VSG internal voltage. , VSG output current, i.e. inductor current. , , , Voltage at PCC. , , Grid voltage. , , LC filter inductor, capacitor and damping resistor, respectively. , Grid equivalent circuit inductor and resistor, respectively. , Positive / negative sequence small signal perturbation device.

[0070] In this embodiment, the grid-connected converter adopts VSG control. Considering the influence of voltage (voltage inner loop), active control loop, reactive control loop and current sampling delay, a grid-connected VSG sequence impedance model is established, and the grid-connected VSG control block diagram is shown in Figure 3 .

[0071] The active control loop simulates the inertia and primary frequency modulation characteristics of the synchronous virtual generator, and the mathematical model is as follows:

[0072] ;

[0073] In the formula: is the rated angular frequency of the grid, is the VSG output angular frequency, is the damping coefficient, wherein, is the damping coefficient reference value, is the adjustable amount of the damping coefficient, is the virtual inertia of the active control loop, wherein, is the virtual inertia reference value, is the adjustable amount of the virtual inertia, is the Laplace operator, is the VSG output instantaneous active power, is the converter output active power given value, is the phase angle of the VSG three-phase modulation wave.

[0074] In view of the fact that impedance remodelling in the embodiment has an impact on active control loop parameters, i.e. changes the active control loop parameters, and further causes a change in the VSG output angular frequency. The embodiment introduces a speed change rate feedback on the basis of expression (1), and further compensates for the change in the VSG output angular frequency caused by impedance remodelling. The expression of the speed change rate feedback is as follows:

[0075] ;

[0076] In the formula: is the active power reference value of the active loop, is the power feedback introduced when the VSG output angular frequency changes, is the power feedback coefficient, is the VSG output angular frequency change rate.

[0077] The rotor motion equation considering impedance real-time reconstruction is obtained by combining expression (1) and expression (2), and is as follows:

[0078] ;

[0079] The reactive control loop simulates the primary voltage regulation characteristic of a synchronous generator, and the mathematical model is as follows:

[0080] ;

[0081] In the formula: is the converter output reactive power given value, is the VSG output instantaneous reactive power, is the reactive damping coefficient, is the rated voltage amplitude, is the output voltage amplitude, is the reactive inertia coefficient, is the VSG internal potential amplitude;

[0082] As can be seen from Figure 2 , the d-axis and q-axis modulation voltages are:

[0083] ;

[0084] In the formula: , is the d-axis and q-axis modulation voltage, , is the VSG output voltage d-axis and q-axis component, is the voltage loop PI controller transfer function, is the PI controller proportional coefficient, wherein, is a reference value, is an adjustable amount, is a PI controller integral coefficient, wherein is a reference value, is an adjustable amount.

[0085] Based on the above principle, the network type VSG sequence impedance model is established, the positive and negative sequence disturbance voltage is injected at the PCC point, taking phase A as an example, assuming that the A-phase output voltage and current at this time are:

[0086] ;

[0087] In the formula: is the A-phase output voltage at the moment, is the A-phase output current at the moment, the subscript 1 represents the fundamental component, the subscript p represents the positive sequence component, and the subscript n represents the negative sequence component, , , are the peak value of the fundamental voltage, the peak value of the positive sequence disturbance voltage, and the peak value of the negative sequence disturbance voltage respectively, , are the positive sequence voltage phase angle and the negative sequence voltage phase angle respectively, , , are the peak value of the fundamental current, the peak value of the positive sequence disturbance current, and the peak value of the negative sequence disturbance current respectively, , , are the fundamental current phase angle, the positive sequence current phase angle, and the negative sequence current phase angle respectively, , , are the fundamental frequency, the positive sequence frequency, and the negative sequence frequency respectively.

[0088] The above A-phase output voltage is converted to the frequency domain and is represented as:

[0089] ;

[0090] In the formula: represents the A-phase output voltage in the frequency domain.

[0091] The above A-phase output current is converted to the frequency domain. Since the disturbance voltage is given by the device, the disturbance current is obtained by sampling, in order to ensure the accuracy of the impedance model, the delay of current sampling is considered, and the current frequency domain is:

[0092] ;

[0093] In the formula: is the A-phase output current in the frequency domain, is the current sampling delay transfer function, wherein is the analog sampling delay, is the analog pulse width modulation (PWM) delay, is the analog sampling low-pass filter, is the sampling period, is the cut-off angular frequency of the low-pass filter, is the counterclockwise rotation angle in the complex space, is the imaginary unit,

[0094] In order to facilitate calculation, in the embodiment, the following is used The instantaneous power theory is used to calculate the output instantaneous active power and reactive power of the grid-forming VSG in the coordinate system:

[0095] ;

[0096] In the formula: and are the output currents of the VSG in the coordinate system, and are the output voltages of the VSG in the coordinate system.

[0097] The expressions (8) and (9) are combined, and the VSG output instantaneous active power is obtained by the frequency domain convolution theorem:

[0098] ;

[0099] In the formula: , is the current sampling delay transfer function after frequency offset, , , are the fundamental, positive sequence and negative sequence current phase angles, respectively, , are the positive sequence and negative sequence voltage phase angles, respectively.

[0100] The expression (10) is substituted into the expression (2) to obtain the phase angle of the VSG three-phase modulation wave in the frequency domain , that is, Since the phase angle of the VSG three-phase modulation wave exists phase angle disturbance, that is, wherein, is the phase angle disturbance caused by small signal, is the phase angle of the fundamental component in the grid-forming VSG three-phase modulation wave.

[0101] The small signal induced phase angle disturbance in frequency domain The expression is:

[0102] ;

[0103] In the formula: is the active controller transfer function, .

[0104] From expression (11), the influence of positive and negative sequence disturbance voltage and current on is decoupled, so the relationship between positive sequence disturbance voltage and current, negative sequence disturbance voltage and current and is analyzed independently, and the following is obtained:

[0105] ;

[0106] Without small signal interference, the three-phase frequency domain voltage is converted to the dq coordinate system by Park transformation:

[0107] ;

[0108] ;

[0109] In the formula: , are the d-axis and q-axis components of three-phase frequency domain voltage in frequency domain without considering the small signal induced phase angle disturbance , and dc is the direct current.

[0110] Based on the influence of small signal induced phase angle disturbance on Park transformation of phase angle, combined with the d-axis and q-axis voltage values of three-phase frequency domain voltage without considering small signal disturbance, the following is derived:

[0111] ;

[0112] In the formula: is the Park transformation formula, represents the zero sequence voltage component.

[0113] Expressions (13), (14), and (15) are obtained by combining:

[0114] ;

[0115] ;

[0116] For simplifying the calculation expression, where:

[0117] ;

[0118] ;

[0119] In the formula: is the d-axis component of VSG output voltage in frequency domain, is the q-axis component of VSG output voltage in frequency domain.

[0120] The expression (5), (16), (17) is obtained by simultaneous equations in frequency domain d-axis, q-axis modulation signal , :

[0121] ;

[0122] ;

[0123] From the Park transformation, , combined with , transformed into:

[0124] ;

[0125] In the formula: , is the d-axis, q-axis component of equivalent modulation voltage considering small signal phase angle disturbance, is the A-phase modulation voltage.

[0126] The expression (11), (18), (19) is obtained by simultaneous equations in frequency domain d-axis, q-axis component of equivalent modulation voltage considering small signal phase angle disturbance , :

[0127] ;

[0128] ;

[0129] The expression (20), (21), (22) is obtained by simultaneous equations in frequency domain A-phase modulation voltage :

[0130] ;

[0131] wherein: is the power angle, .

[0132] wherein: Figure 2 The master circuit equation is:

[0133] ;

[0134] wherein: is the A-phase output voltage, is the A-phase output current.

[0135] For the purpose of simplifying the expression form, let , and the expression (23) and (24) are combined to obtain the network-constructed VSG serial impedance model:

[0136] ;

[0137] wherein: is the PI controller transfer function, , is the current sampling delay transfer function, , is the positive sequence impedance, is the negative sequence impedance.

[0138] In this embodiment, the network-constructed VSG is divided into two parts, i.e. the network-constructed VSG and the power grid. Based on the impedance model of the network-constructed VSG and the equivalent impedance of the power grid, the impedance ratio of the target network-constructed VSG is analyzed according to the Nyquist criterion, so as to analyze the interactive stability of the converter connected to the power grid.

[0139] In this embodiment, the external characteristic of the network-constructed VSG converter is taken as a controlled voltage source, and combined with the expression (25), the small signal description of the converter connected to the power grid is as shown in Figure 4 and Figure 5 . Wherein, is the ideal voltage source of the equivalent Thevenin circuit; is the ideal voltage source of the power grid; , wherein, is the equivalent input impedance of the power grid, is the positive sequence equivalent impedance of the power grid, is the negative sequence equivalent impedance of the power grid. Due to the influence of factors such as line switching, fault, load fluctuation and series compensation line, the equivalent input impedance of the power grid presents significant time-varying and uncertainty, which causes the real-time change of the strength of the power grid. In this embodiment, the short-circuit ratio is used to represent the change of the strength of the power grid:

[0140] ;

[0141] wherein: is the AC grid rated voltage, is the rated power, is the grid impedance unit value.

[0142] The grid strength characterized by short circuit ratio (SCR) is in linear relationship with the grid equivalent input impedance, and the ratio of rated power to the square of AC grid rated voltage is , and the grid strength is . .

[0143] According to Figure 4 , Figure 5 , the grid-connected current expression of grid-forming VSG is:

[0144] ;

[0145] wherein: is the equivalent positive sequence current, is the equivalent negative sequence current.

[0146] From the above, the small-signal stability of the converter connected to the grid depends on and , combined with the relationship between the grid strength and the grid equivalent input impedance, the stability depends on whether the Nyquist curve of the grid-forming VSG grid-connected open-loop transfer function meets the generalized Nyquist criterion, wherein the grid-forming VSG grid-connected open-loop transfer function includes the positive sequence open-loop transfer function and the negative sequence open-loop transfer function , and , , that is , are the positive sequence open-loop transfer function and the negative sequence open-loop transfer function, respectively. , are converted from the S domain to the frequency domain, and , , wherein , are the real part and the imaginary part of the positive sequence open-loop transfer function in the frequency domain, respectively, , are the real part and the imaginary part of the negative sequence open-loop transfer function in the frequency domain, respectively.

[0147] According to the generalized Nyquist criterion in the frequency domain, the grid-connected grid-forming VSG meets the following formula:

[0148] ;

[0149] The current control parameters of the network-forming VSG and the current grid strength SCR are obtained, and the expressions (25) and (28) are used to calculate the resonant frequencies of the positive sequence and the negative sequence open-loop transfer functions , the calculation of , whether the network-forming VSG is in a risk of instability.

[0150] The impedance remodeling should track the change of the grid impedance in real time. In order to ensure the efficiency of the impedance remodeling, the embodiment performs sensitivity analysis on the control parameters affecting the output impedance of the impedance model of the network-forming VSG, and the sensitivity analysis specifically includes:

[0151] The sensitivity analysis is performed on the open-loop transfer function of the network-forming VSG, that is, the sensitivity of the open-loop transfer function of the network-forming VSG with respect to the control parameters is derived, and the expression is:

[0152] ;

[0153] The control parameters in the embodiment include the virtual inertia of the active control loop, the damping coefficient , the proportional coefficient of the PI controller , the integral coefficient of the PI controller , that is, , , , , , wherein is a matrix composed of the positive and negative sequence open-loop transfer functions, is a matrix composed of the real part and the imaginary part of the positive sequence open-loop transfer function in the frequency domain, is a matrix composed of the real part and the imaginary part of the negative sequence open-loop transfer function in the frequency domain.

[0154] The absolute value of the sensitivity of the impedance to the control parameter indicates the influence of the change of the control parameter on the impedance, and the positive sign of the sensitivity indicates that the increase of the parameter will cause the increase of the impedance ratio, and vice versa.

[0155] After the above impedance sensitivity analysis method, the sensitivity of the positive and negative sequence impedance ratio to each control parameter is obtained, and the Laplace operator is substituted to obtain the sensitivity of the control parameter at different frequencies, as follows:

[0156] ;

[0157] ​ ;

[0158] In the formula: 、 respectively represent the positive sequence open-loop transfer function real part and imaginary part at different frequencies. The control parameter sensitivity set, 、 respectively represent the negative sequence open-loop transfer function real part and imaginary part at different frequencies. The control parameter sensitivity set.

[0159] Let , , the resonant frequency of the positive sequence and negative sequence open-loop transfer function 、 , 、 Substitute formula (30) (31) to get the sensitivity of each control parameter to impedance remodeling at the current resonant frequency .

[0160] In the background of real-time changes in grid strength, the improved pollen algorithm is used to control the control parameters of the grid-forming VSG in real time, and the real-time matching of the converter output impedance and the grid impedance is realized, including:

[0161] The objective function of the improved pollen algorithm has two paths, one path is to improve the impedance ratio real part in the frequency domain, and the second path is to reduce the impedance ratio imaginary part, that is, to reduce the frequency range of and The Nyquist curve in the frequency domain moves to the right.

[0162] The embodiment optimizes the grid-forming VSG grid-connected open-loop transfer function through two paths at the same time, avoids the case that the grid-forming VSG grid-connected open-loop transfer function exists and , and improves the stability margin of the grid-forming VSG grid-connected while ensuring the stability of the grid-forming VSG grid-connected.

[0163] The objective function of the improved pollen algorithm in the embodiment is:

[0164] The first path objective function is:

[0165] At the resonant frequency , ensure , and reduce the unstable bandwidth:

[0166] ;

[0167] In the formula: is the first path objective function, is the real part constraint of the grid-forming VSG grid-connected open-loop transfer function, ,in, The penalty coefficient is... The value should satisfy , For the gain margin parameter, , In frequency , The real part of the next-order open-loop transfer function. The set of frequencies whose real part is less than -1. Let be the real part of the open-loop transfer function of the VSG in a grid-connected configuration. For frequency.

[0168] Second path objective function:

[0169] Reduce the imaginary part of the grid-connected open-loop transfer function of the VSG. The amplitude, and the resonant frequency Move right:

[0170] ;

[0171] In the formula: The objective function for the second path is... For the integral value under unstable bandwidth, the prerequisite for stable network connection of the VSG is to ensure that this integral value is negative. The penalty coefficient is... ,in, , Let be the frequencies corresponding to the real part of the positive-sequence and negative-sequence open-loop transfer functions being equal to -1, respectively. , , For the imaginary part constraint of the open-loop transfer function of the VSG grid-connected system. ,in, , In frequency , The imaginary part of the open-loop transfer function in positive and negative order. As a penalty coefficient, ensure that in hour To ensure the stability of the converter while maintaining a certain phase margin, among which... This is the phase margin parameter.

[0172] The improved fitness function of the pollen algorithm is:

[0173] ;

[0174] In the formula: For the fitness function, , The first path objective function and the second path objective function are respectively weighted.

[0175] Considering the adjustable range of each parameter of the network type VSG, the improved pollen algorithm constraint function is designed as follows:

[0176]

[0177] In the formula: Jmin and Jmax are the lower limit and the upper limit of the adjustable range of the virtual inertia J of the active control loop, Kmin and Kmax are the lower limit and the upper limit of the adjustable range of the damping coefficient, Kpmin and Kpmax are the lower limit and the upper limit of the adjustable range of the PI controller proportion coefficient, Ki min and Ki max are the lower limit and the upper limit of the adjustable range of the PI controller integral coefficient, Jmin, Jmax, Kmin, Kmax, Kpmin, Kpmax, Ki min and Ki max are the lower limit and the upper limit of the adjustable range of the virtual inertia, Kmin and Kmax are the lower limit and the upper limit of the adjustable range of the damping coefficient, Kpmin and Kpmax are the lower limit and the upper limit of the adjustable range of the PI controller proportion coefficient,

[0178] Step 5 specifically includes:

[0179] Step 5.1, a chaotic (Tent) mapping generates an initial population;

[0180] In this embodiment, the Tent mapping is used to generate a chaotic sequence. Given the upper limit and the lower limit of the pollen individual (i.e., the upper limit and the lower limit of the adjustable range of each control parameter), the Tent is used to take values within the upper limit and the lower limit of the adjustable range of each control parameter during initialization:

[0181]

[0182] In the formula: represents a randomly generated random sequence in the interval [0, 1], is a chaotic sequence, and the constant .

[0183] Step 5.2, combined with sensitivity for weighted expansion;

[0184] According to the sensitivity of each control parameter to impedance remodeling at the current resonant frequency ​​​​​​​​​​​​​, the sensitivity weight factor of each control parameter is calculated:

[0185] ;

[0186] wherein: , is the sensitivity weight factor of control parameter under the positive sequence and negative sequence open-loop transfer function, , is the sensitivity of each control parameter under the positive sequence and negative sequence open-loop transfer function, wherein, , is calculated by expression (30) and expression (31), is the control parameter number.

[0187] Based on the sensitivity weight factor of each control parameter under the positive sequence and negative sequence open-loop transfer function, the pollen population generation direction is controlled, and the pollen individual initial position sequence in the search area is generated:

[0188] (38);

[0189] wherein: is the pollen individual initial position sequence, , is the minimum value and maximum value in the generation pollen individual , is a chaotic sequence, is the sensitivity weight factor of control parameter under the positive sequence and negative sequence open-loop transfer function, .

[0190] Step 5.3, calculate the dynamic switching probability;

[0191] In this embodiment, the dynamic switching probability P is set as an exponential function related to the iteration number, so that the pollen algorithm focuses on cross-pollination in the early stage and focuses on self-pollination in the later stage, and improves the convergence speed in the later stage.

[0192] (39);

[0193] wherein: and T represent the current iteration number and the maximum iteration number.

[0194] Step 5.4, iteration optimization:

[0195] If , it is cross-pollination, otherwise it is self-pollination. The expression of cross-pollination is:

[0196] (40);

[0197] wherein: , denote the i-th pollen individual in the j-th generation, +1, the (i+1)-th pollen individual in the j-th generation, is a random individual in the population different from the pollen, is the best pollen in the population, and the control parameter is a D-dimensional pollination intensity vector, D is the dimension of the problem to be solved, and each dimension is a random number obeying a levy distribution;

[0198] (41);

[0199] wherein: is a standard gamma function; the constant , is a random step size.

[0200] In the self-pollination stage, a plurality of mutation strategies are added, mutation is performed synchronously, the mutation range is increased, and comparison is performed through calculation of fitness, and the optimal one is selected, so as to avoid falling into a local optimum. The specific strategies are as follows:

[0201] Basic algorithm mutation strategy:

[0202] (42);

[0203] wherein: , are two random individuals in the population different from the pollen, is a random number obeying a uniform distribution on [0, 1].

[0204] Directional mutation strategy:

[0205] (43);

[0206] wherein: , denote the i-th pollen individual in the j-th generation, +1, the (i+1)-th pollen individual in the j-th generation; is the best pollen in the population; is a random number obeying a uniform distribution on [0, 1]; is a random individual in the population different from the pollen. Difference evolution mutation strategy:

[0207]

[0208] ​​ (44);

[0209] wherein: , denote the first +1, generation of pollen individuals; is the optimal pollen in the population; , are two random individuals different from the pollen in the population; is a random number obeying uniform distribution on [0,1]. Step 5.4, evaluating the new generation of pollen, updating the optimal pollen, and obtaining the optimal solution

[0210] .

[0211] By adjusting the control parameters in real time , the impedance of the grid-forming VSG is reshaped in real time, ensuring that the control parameters are matched with the grid impedance in real time under the change of the grid strength, and improving the grid-connection stability of the grid-forming VSG.

[0212] The above method is realized based on the grid-forming converter impedance real-time reconstruction system in the embodiment, and includes:

[0213] A model establishing module is configured to establish a grid-forming VSG sequence impedance model;

[0214] A stability judging module is configured to obtain a grid-forming VSG grid-connection open-loop transfer function containing the grid strength based on the grid-forming VSG sequence impedance model, combining the change of the grid strength and impedance analysis, convert the grid-forming VSG grid-connection open-loop transfer function from the S domain to the frequency domain, and judge the grid-connection stability of the grid-forming VSG;

[0215] An analysis module is configured to analyze the grid-forming VSG grid-connection open-loop transfer function in the frequency domain based on the judging result by using the impedance sensitivity analysis method, and obtain the sensitivity of each control parameter to impedance reshaping;

[0216] An optimization module is configured to optimize the grid-forming VSG grid-connection open-loop transfer function by using the improved pollen algorithm;

[0217] A reconstruction module is configured to control the pollen population generation direction of the improved pollen algorithm based on the sensitivity of each control parameter to impedance reshaping, and reshape the impedance of the grid-forming VSG in real time.

[0218] ​Those skilled in the art should understand that the modules or steps of the present application described above can be realized by a general computer device, or alternatively, they can be realized by program codes executable by a computing device, so that they can be stored in a storage device and executed by a computing device, or they can be respectively manufactured as individual integrated circuit modules, or a plurality of modules or steps among them can be manufactured as a single integrated circuit module. The present application is not limited to any specific combination of hardware and software.

[0219] Those skilled in the art can understand that unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with those in the context of the prior art, unless otherwise defined herein, and should not be interpreted in an idealized or overly formal sense.

[0220] The above specific embodiments further illustrate the purposes, technical solutions and beneficial effects of the present application. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for real-time impedance reconstruction of a grid-type converter, characterized in that: The method comprises the following steps: establishing a sequence impedance model of a grid-forming VSG; based on the sequence impedance model of the grid-forming VSG, combining grid strength variation and impedance analysis, obtaining a grid strength containing grid-forming VSG grid-connected open-loop transfer function, converting the grid-forming VSG grid-connected open-loop transfer function from S domain to frequency domain, and judging the grid-forming VSG grid-connected stability; based on the unstable judgment result, using impedance sensitivity analysis method to analyze the grid-forming VSG grid-connected open-loop transfer function in frequency domain, and obtaining the sensitivity of each control parameter to impedance remodeling; using the improved pollen algorithm to optimize the grid-forming VSG grid-connected open-loop transfer function, controlling the pollen population generation direction of the improved pollen algorithm based on the sensitivity of each control parameter to impedance remodeling, and remodeling the impedance of the grid-forming VSG in real time, including: determining the objective function of the improved pollen algorithm, including: the first path objective function: ； In the formula: is a first path target function, is a real part constraint of a grid-connected open-loop transfer function of a grid-forming VSG, is a real part of a grid-connected open-loop transfer function of a grid-forming VSG, is a frequency, is a set of frequencies with a real part less than -1. the second path objective function: ； In the formula: is a second path target function, is a virtual part constraint of a grid-connected open-loop transfer function of a grid-forming VSG, is an integral quantity under unstable bandwidth, is a virtual part of a grid-connected open-loop transfer function of a grid-forming VSG, is a penalty coefficient, wherein, , are respectively a positive sequence and a negative sequence open-loop transfer function corresponding to a frequency at which a real part is equal to -1, is a resonance frequency; the fitness function of the improved pollen algorithm is: ； In the formula: is the fitness function, , are the weights of the first path objective function and the second path objective function, respectively. determining the constraint function of the improved pollen algorithm: ； In the formula: , are the lower limit and the upper limit of the adjustable quantity of the virtual inertia J of the active control loop, , is a damping coefficient are the lower limit and the upper limit of the adjustable quantity, , is a PI controller proportional coefficient are the lower limit and the upper limit of the adjustable quantity, , is a PI controller integral coefficient are the lower limit and the upper limit of the adjustable quantity, is the adjustable quantity of the virtual inertia, is the adjustable quantity of the damping coefficient, is the adjustable quantity of the PI controller proportional coefficient, is the adjustable quantity of the PI controller integral coefficient. based on the sensitivity of each control parameter to impedance remodeling at the current resonant frequency, calculating the sensitivity weight factor of each control parameter: ； In the formula: is a control parameter, , is a control parameter is a sensitivity weight factor under the positive sequence open-loop transfer function, , is a sensitivity of each control parameter under the positive sequence open-loop transfer function, is a control parameter is the number of control parameters based on the sensitivity weight factor of each control parameter, controlling the pollen population generation direction, and generating the initial position sequence of pollen individuals in the search area: ； In the formula: a sequence of initial positions of pollen individuals, , are the minimum value, maximum value, respectively, of the pollen individuals of the nth generation are the minimum value, maximum value, respectively, is a chaotic sequence, is a control parameter sensitivity weight factors of positive sequence, negative sequence open-loop transfer function, .

2. The network configuration type converter impedance real-time reconstruction method according to claim 1, characterized in that: The method for establishing a sequence impedance model of a grid-forming VSG comprises the following steps: using the method of harmonic linearization, based on VSG power ring, voltage and current sampling delay, the following is obtained: ； for simplifying calculation, let: ； ； ； ； wherein: is the positive sequence impedance, is the negative sequence impedance, is the positive sequence disturbance current peak, is the negative sequence disturbance current peak, is the fundamental current peak, is the fundamental voltage peak, is the positive sequence disturbance voltage peak, is the negative sequence disturbance voltage peak, is the internal voltage amplitude of the VSG, is the voltage loop PI controller transfer function, is the current sampling delay transfer function, , , are the fundamental, positive sequence, and negative sequence current phase angles, respectively, , are the positive sequence and negative sequence voltage phase angles, respectively, , , are the LC filter filter inductance, filter capacitance, and damping resistance, respectively, is the Laplace operator, is the power angle, denotes the imaginary unit, is the angle of rotation in the complex space counterclockwise, is the active controller transfer function, is the grid rated angular frequency, is the damping coefficient reference value, is the damping coefficient adjustable quantity, is the virtual inertia reference value, is the virtual inertia adjustable quantity.

3. The network configuration type converter impedance real-time reconstruction method according to claim 1, characterized in that: based on the sequence impedance model of the grid-forming VSG, combining grid strength variation and impedance analysis, obtaining a grid strength containing grid-forming VSG grid-connected open-loop transfer function, converting the grid-forming VSG grid-connected open-loop transfer function from S domain to frequency domain, and judging the grid-forming VSG grid-connected stability, including: The grid strength change is represented by a short-circuit ratio, and an impedance analysis is combined to obtain a grid-forming VSG grid-connected open-loop transfer function including positive sequence open-loop transfer function and negative sequence open-loop transfer function , is the grid strength, is the positive sequence impedance, is the negative sequence impedance, is the ratio of rated power to square of rated voltage of alternating current grid.​ converting the grid-forming VSG grid-connected open-loop transfer function from S domain to frequency domain, using the generalized Nyquist criterion in frequency domain to judge the grid-forming VSG grid-connected stability.

4. The network configuration type converter impedance real-time reconstruction method according to claim 1, characterized in that: The control parameters include active control loop virtual inertia J, damping coefficient and PI controller proportional coefficient , PI controller integral coefficient The expression of the sensitivity of each control parameter to impedance remodeling is: ； ； In the formula: , are respectively the real part and the imaginary part of the positive sequence open-loop transfer function, and are respectively the sensitivity set of each control parameter under different frequencies, , are respectively the real part and the imaginary part of the negative sequence open-loop transfer function, and are respectively the sensitivity set of each control parameter under different frequencies, , are respectively the real part and the imaginary part of the positive sequence open-loop transfer function in the frequency domain, , are respectively the real part and the imaginary part of the negative sequence open-loop transfer function in the frequency domain.

5. A system for real-time impedance reconstruction of a network-forming converter based on the method of any of claims 1 to 4. The method comprises the following steps: a model establishing module for establishing a sequence impedance model of a grid-forming VSG; a stability judging module for, based on the sequence impedance model of the grid-forming VSG, combining grid strength variation and impedance analysis, obtaining a grid strength containing grid-forming VSG grid-connected open-loop transfer function, converting the grid-forming VSG grid-connected open-loop transfer function from S domain to frequency domain, and judging the grid-forming VSG grid-connected stability; an analysis module for, based on the judgment result, using impedance sensitivity analysis method to analyze the grid-forming VSG grid-connected open-loop transfer function in frequency domain, and obtaining the sensitivity of each control parameter to impedance remodeling; an optimization module for using the improved pollen algorithm to optimize the grid-forming VSG grid-connected open-loop transfer function; a remodeling module for, based on the sensitivity of each control parameter to impedance remodeling, controlling the pollen population generation direction of the improved pollen algorithm, and remodeling the impedance of the grid-forming VSG in real time.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: The processor executes the computer program to realize the steps of the grid-forming converter impedance real-time remodeling method in any one of claims 1 to 4.

7. A computer readable storage medium storing a computer program, characterized in that: The computer program is executed by the processor to realize the steps of the grid-forming converter impedance real-time remodeling method in any one of claims 1 to 4.

Images