Resonance suppression method for centralized new energy grid-connected system

Abstract

The resonant suppression method of centralized new energy grid-connected system solves the problem of resonant instability of the centralized new energy grid-connected system and belongs to the technical field of new energy power generation and power grid. The grid impedance and the total output impedance of the new energy grid-connected converter are identified online, the resonant frequency and the grid voltage resonant component are extracted through the impedance identification result, the required virtual impedance is dynamically adjusted based on the adaptive virtual impedance algorithm, the calculation result is sent into the control loop of the new energy grid-connected converter, the resonant suppression method based on the virtual impedance compensation is used, the output impedance of the grid-connected converter at the resonant frequency band is reshaped to suppress the resonance phenomenon, the wide frequency domain resonance suppression function is realized while ensuring the normal power output of the new energy grid-connected system, and an economic and feasible control scheme is provided for the safe and stable grid connection of the centralized new energy.

Description

TECHNICAL FIELD

[0001] The present application relates to a resonance suppression method of a centralized new energy grid-connected system, and belongs to the technical field of new energy power generation and power grid. BACKGROUND

[0002] With the increasing penetration of new energy power generation systems, the power grid is developing into a new type of power system characterized by a high proportion of new energy and high power electronics. However, most new energy power generation systems are located in remote end-of-grid areas, and the multi-stage transformer link and long-distance transmission line result in the weak grid characteristics of "low short-circuit ratio" and high impedance. As the main interface between new energy power generation devices and the power grid, grid-connected converters play an important role in new energy grid-connected applications. Photovoltaic and wind power, etc. are mainly connected to the grid in the form of current-controlled converters, which are different from traditional power frequency power generation equipment. The output impedance of grid-connected converters varies greatly at different frequencies. Since large-scale new energy power generation systems mainly use centralized grid connection, and the grid impedance under weak grid conditions cannot be ignored, the coupling interaction between the output impedance of multiple grid-connected converters and the grid impedance is prone to cause wide-band resonance problems, which seriously affects the safe and stable operation of new energy grid-connected systems.

[0003] To suppress the resonance phenomenon caused by the coupling of grid impedance and grid-connected converter output impedance and improve the stability of new power systems, existing researches mostly use virtual impedance compensation through grid-connected converters to achieve impedance reshaping to weaken the system impedance coupling effect. This method is an advanced control method that changes the current injected into the grid by the grid-connected converter without the need for additional actual impedance to change the frequency characteristics of the output impedance of the grid-connected converter. Existing virtual impedance compensation methods are mainly divided into fixed impedance compensation and adaptive impedance compensation. The stable operation capability of new power systems is often affected by system impedance coupling. Under the condition of centralized new energy grid connection, the real-time variation of system impedance caused by the fluctuation of new energy output and the existence of a large number of power electronic conversion devices. Therefore, how to obtain system impedance information online in real time, consider the influence of system impedance coupling on system stability, and realize adaptive impedance reshaping is one of the key problems in the field of new energy power generation grid connection. SUMMARY

[0004] In view of the resonance instability problem of the centralized new energy grid-connected system, the present application provides a resonance suppression method of a centralized new energy grid-connected system.

[0005] The resonance suppression method of the centralized new energy grid-connected system of the present application, the centralized new energy grid-connected system includes N grid-connected converters of new energy, N is a positive integer, the method comprises:

[0006] S1, online identify the total impedance Z of the output of the grid-connected converter of the new energy o (s) and the grid impedance Z g (s);

[0007] S2, calculate the required compensation virtual impedance Z v (s), including:

[0008] S21, extract the system resonance frequency ω o (s) and the grid impedance Z g (s) according to the total impedance Z of the output of the grid-connected converter of the new energy z ;

[0009] S22, obtain the reasonable value range [θ vmin ,θ vmax ] of the phase angle of the virtual impedance by using the virtual impedance phase angle planning model;

[0010] S23, on the basis of the reasonable value range [θ vmin ,θ vmax ], determine the optimal virtual impedance phase angle θ z according to the frequency band where the system resonance frequency ω v ;

[0011] S24, obtain the reasonable value range [Z vmin ,Z vmax ] of the amplitude of the virtual impedance by using the virtual impedance amplitude planning model;

[0012] S25, on the basis of the reasonable value range [Z vmin ,Z vmax ], determine the optimal virtual impedance amplitude Z v according to the demand of the resonance suppression speed of the current grid-connected point voltage resonance component;

[0013] S26, determine the required compensation virtual impedance Z v according to the optimal virtual impedance phase angle θ v and the optimal virtual impedance amplitude Z s is the Laplace operator;

[0014] S3, send the required compensation virtual impedance Z v (s) into the control loop of the grid-connected converter of the new energy, and control the grid-connected converter of the new energy.

[0015] As preferred, the virtual impedance phase angle planning model is:

[0016] θ vmin = min∠Z v (s), θ vmax = max∠Z v (s)

[0017]

[0018] Z v (s) is the virtual impedance to be compensated, ∠Z v (s) represents Z v (s) is the impedance phase angle of Z o (s c ) is the total output impedance of the new energy grid-connected converter at ω c after impedance reshaping, the intermediate variable s c = jω c , ω c is the cut-off frequency of Z g (s) / Z' o (s), and PM is the phase angle margin of Z g (s) / Z' o (s), R g is the virtual resistance, L g is the virtual inductance, θ lim is the minimum virtual impedance phase angle that N new energy grid-connected converters can compensate, Z lim is the minimum virtual impedance amplitude that N new energy grid-connected converters can compensate; ∠Z g (s c ) is the impedance phase angle of Z g (s c ), ∠Z' o (s c ) is the impedance phase angle of Z' o (s c ).

[0019] As a preferred, the optimal virtual impedance phase angle θ v is:

[0020]

[0021] wherein ω1 is the low-to-medium frequency resonance boundary angular frequency, and ω2 is the medium-to-high frequency resonance boundary angular frequency.

[0022] As a preferred, the virtual impedance amplitude planning model is:

[0023] Z vmin = min |Z v (s) |, Z vmax = max |Z v (s) |

[0024]

[0025] wherein Z v (s) is the virtual impedance to be compensated, ∠Z v(s) represents Z v (s) represents Z o (s c ) is the output total impedance of the new energy grid-connected converter at ω c (s) is the intermediate variable s c = jω c , ω c is Z g (s) / Z o (s) is the cut-off frequency of Z g (s) / Z o (s) is the phase angle margin of Z g R is the virtual resistance g L is the virtual inductance lim θ is the minimum virtual impedance phase angle that N new energy grid-connected converters can compensate lim ∠Z is the minimum virtual impedance amplitude that N new energy grid-connected converters can compensate; ∠Z g (s c ) is Z g (s c ) represents Z o (s c ) represents Z o (s c ) represents Z

[0026] As a preferred, the optimal virtual impedance amplitude Z v is:

[0027]

[0028] U pcc is the effective value of the grid-connected point voltage pcch U is the effective value of the grid-connected point voltage resonance component.

[0029] As a preferred, according to the new energy grid-connected converter output total impedance Z o (s) and the grid impedance Z g (s), the resonance frequency extraction model is used to extract the system resonance frequency ω z : the resonance frequency extraction model is:

[0030] ω z = minω

[0031]

[0032] Wherein, s = jω, ω represents the angular frequency, PM1 is the phase angle margin of Z g (s) / Z o (s);

[0033] Solving the resonance frequency extraction model obtains the system resonance frequency ω z .

[0034] As preferred, the method for online identifying the output total impedance Z o (s) and the grid impedance Z g (s) of the new energy grid-connected converter includes:

[0035] Let a single new energy grid-connected converter work in the grid-connected mode, and online identify the current grid positive sequence admittance G gp (s) and the negative sequence admittance G gn (s) through pulse response analysis;

[0036] When N grid-connected converters are simultaneously normally grid-connected, let the N grid-connected converters sequentially online identify the current grid point positive sequence admittance and negative sequence admittance through pulse response analysis:

[0037]

[0038] Wherein, G pi (s) and G ni (s) are the online identification results of the grid point positive sequence admittance and negative sequence admittance of the i-th grid-connected converter, G opk (s) and G onk (s) are the positive sequence output admittance and negative sequence output admittance of the k-th grid-connected converter;

[0039] Operating G pi (s) and G ni (s) of the N grid-connected converters obtains the new energy grid-connected converter positive sequence output total admittance G op (s) and the negative sequence output total admittance G on (s):

[0040]

[0041] Operating the grid positive sequence admittance G gp (s) and the new energy grid-connected converter positive sequence output total admittance G op (s) obtains the grid impedance Z g (s) and the new energy grid-connected converter output total impedance Z o (s):

[0042]

[0043] As preferred, the virtual impedance Z v (s) required for compensation is sent into the control loop of the grid-connected converter, and the method for controlling the grid-connected converter includes:

[0044] According to the system resonance frequency ωz Adjust the resonant frequency ω of the proportional-resonance controller n and the center frequency ω of the bandpass filter r ;

[0045] Using a bandpass filter to analyze the grid connection point voltage signal u pcc Filtering is performed to obtain the resonant component u of the grid connection point voltage. pcch ;

[0046] Based on the resonant component u of the grid connection point voltage pcch and the virtual impedance Z to be compensated v (s) Obtain the given value of the virtual impedance compensation current loop of the grid-connected converter.

[0047] Set the virtual impedance compensation current loop setpoint i of the grid-connected converter. href α and β axis components i hαref i hβref The difference between each value and its actual value is calculated, and the two differences are fed into a proportional-resonant controller to obtain the α and β axis modulation signals. The α and β axis modulation signals are then subjected to an inverse Clark transform to obtain the virtual impedance compensation modulation signal v. h The virtual impedance compensated modulated signal v h With power modulation signal v b The switching control signal of the grid-connected converter is obtained by superimposing PWM modulation, and the grid-connected converter is controlled according to the switching control signal.

[0048] Preferably, based on the given and actual values ​​of the grid-side filter inductor current and the converter-side filter inductor current of the grid-connected converter, the power modulation signal v is obtained through a dual-closed-loop fully decoupled control method in the dq coordinate system. b ,include:

[0049] The outer current control loop sets the current setpoint i of the grid-side filter inductor of the grid-connected converter. L2derf i L2qerf Each with its actual value i L2d i L2q The two differences are then fed into a proportional-integral controller for calculation, yielding the converter-side filter inductor current setpoint i. L1derf i L1qerf ;

[0050] The inner current control loop sets the current setpoint i of the filter inductor on the converter side. L1derf i L1qerf Each with its actual value i L1d i L1qThe two difference values are respectively sent into proportional-integral controllers for operation, and the operation results are respectively superimposed on the d-axis and q-axis decoupling feedforward components to obtain d-axis and q-axis modulation signals, and the d-axis and q-axis modulation signals are divided by the DC side voltage V dc The result after the normalization processing is obtained after Park inverse transformation to obtain the power modulation signal v b .

[0051] The application also provides a control system of the resonance suppression method based on the centralized new energy grid-connected system.

[0052] The system impedance online identification unit online identifies the total output impedance Z o (s) of the new energy grid-connected converter. g (s) of the power grid.

[0053] The adaptive virtual impedance calculation unit calculates the required compensation virtual impedance Z v (s).

[0054] The virtual impedance compensation unit sends the required compensation virtual impedance Z v (s) into the control loop of the grid-connected converter to control the grid-connected converter.

[0055] The application provides a system impedance online identification method, which is suitable for a centralized new energy grid-connected system containing multiple grid-connected devices, can online obtain real-time power grid impedance and total output impedance of the new energy grid-connected device, provides accurate impedance information model for system stability analysis and improvement, and realizes accurate monitoring and effective management of system stable operation capability.

[0056] The application realizes direct control and elimination of system resonance components through the resonance suppression method based on virtual impedance compensation, fully utilizes device spare capacity and control resources while ensuring normal power output of the new energy grid-connected device, reshapes the output impedance of the new energy grid-connected device at the resonance frequency band, solves the resonance problem of the new energy grid-connected device, and improves the stable operation capability of the power system.

[0057] The application fully considers system impedance, resonance frequency and resonance components through the adaptive virtual impedance algorithm, dynamically adjusts the required compensation virtual impedance, minimizes the influence of virtual impedance compensation on power output control of the new energy grid-connected device under the premise of ensuring system stability, improves system resonance suppression speed, reduces power loss, and improves the efficiency, safety and economic benefit of system operation. BRIEF DESCRIPTION OF DRAWINGS

[0058] Figure 1 It is a structural schematic diagram of the centralized new energy grid-connected system.

[0059] Figure 2 A schematic diagram of the circuit structure of a grid-connected converter for new energy sources;

[0060] Figure 3 Flowchart of a method for online impedance identification and adaptive resonance suppression in centralized new energy grid-connected systems;

[0061] Figure 4 Flowchart of the online impedance identification method for centralized new energy grid-connected systems;

[0062] Figure 5 The equivalent circuit models of the system before and after virtual impedance compensation are shown, where (a) is before virtual impedance compensation and (b) is after virtual impedance compensation.

[0063] Figure 6 Here is a flowchart of the adaptive virtual impedance algorithm;

[0064] Figure 7 This is a control block diagram for the online impedance identification and adaptive resonance suppression method. Detailed Implementation

[0065] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0066] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0067] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but this is not intended to limit the scope of the invention.

[0068] A schematic diagram of a centralized new energy grid-connected system is shown below. Figure 1 As shown, N new energy power generation devices are connected in parallel to the new energy grid connection point through N new energy grid-connected converters, forming a centralized new energy power generation system, which is then connected to the power grid and transmission system via transformers. pcc For the grid connection point voltage, i pcc R is the grid connection point current, i.e., the output current of the centralized new energy power generation system. g and L g u is the equivalent impedance of the transmission line. g with i g These represent the grid-side voltage and current, respectively.

[0069] The N new energy grid-connected converters in the centralized new energy generation system adopt LCL grid-connected converter topology, and the circuit structure diagram is shown in Figure 2 The main structure of the grid-connected converter is a three-phase bridge converter, and the DC side is composed of a capacitor C dc and a new energy generation device in parallel, V dc and I dc are the DC side voltage and DC side output current respectively. Each grid-connected converter is connected to the new energy grid connection point in parallel through an LCL filter, L1 is the filter inductance on the converter side, L2 is the filter inductance on the grid side, C d is the filter capacitor, i L1 is the current flowing through the inductor L1, that is, the AC side output current of the grid-connected converter, i L2 is the current flowing through the inductor L2, that is, the grid-connected output current of the grid-connected converter, and u C is the filter capacitor voltage.

[0070] The resonance suppression method of the centralized new energy grid-connected system of the embodiment online identifies the grid impedance and the total output impedance of the new energy grid-connected converter, extracts the system resonance frequency and the grid voltage resonance component through the impedance identification result, dynamically adjusts the virtual impedance required for compensation based on the adaptive virtual impedance algorithm, sends the calculation result into the new energy grid-connected converter control loop, uses the resonance suppression method based on virtual impedance compensation, reshapes the grid-connected converter output impedance at the resonance frequency band to suppress the resonance phenomenon, ensures the normal power output of the new energy grid-connected system, and realizes the wide frequency domain resonance suppression function, providing an economical and feasible control scheme for the safe and stable grid connection of centralized new energy. Specifically includes:

[0071] Step one, in the centralized new energy grid-connected system containing N grid-connected converters, online identify the total output impedance Z o (s) of the new energy grid-connected converter and the grid impedance Z g (s):

[0072] First, let a single new energy grid-connected converter work in grid-connected mode, and online identify the current grid positive sequence admittance G gp (s) and negative sequence admittance G gn (s) through impulse response analysis.

[0073] In the implementation case, the impedance online identification method based on impulse response analysis is used for each grid-connected converter in the centralized new energy grid-connected system. This method injects a disturbance pulse into the grid-connected point through a certain new energy grid-connected converter to obtain the system impedance or admittance information excluding the grid-connected converter. Since the grid-connected converter generally adopts current-mode control, a current pulse is used as the disturbance pulse in this case. The specific process includes:

[0074] Step 1: detecting the current output current i of the current new energy power generation system pcc whether the peak is reached, if the peak is reached, step 2 is performed, otherwise, step 1 is continuously executed;

[0075] Step 2: applying a disturbance signal to the current injected into the power grid by the new energy grid-connected converter;

[0076] Step 3: sampling the grid-connected point voltage u pcc and the new energy power generation system output current i pcc , obtaining the current disturbance and voltage response signal, and subtracting the current and voltage signals without disturbance from them to remove the influence of the inherent harmonics of the power grid;

[0077] Step 4: performing discrete Fourier transform (DFT) analysis on the sampled voltage response and current disturbance signals to obtain their respective frequency domain components;

[0078] Step 5: separating the positive and negative sequence frequency domain components of the voltage response and current disturbance, and the ratio of the positive sequence frequency domain component in the current disturbance to the positive sequence frequency domain component in the voltage response is the positive sequence frequency domain admittance at the grid-connected point, and similarly the negative sequence frequency domain admittance at the grid-connected point can be obtained, and the positive and negative sequence frequency domain admittances are inverted to obtain the positive and negative sequence frequency domain impedances.

[0079] Further, when N new energy grid-connected converters are normally grid-connected, the N grid-connected converters are sequentially analyzed online to identify the positive and negative sequence admittances at the grid-connected point by pulse response analysis, as shown in the following formula:

[0080]

[0081] wherein, G pi (s) and G ni (s) are the online identification results of the positive and negative sequence admittances at the grid-connected point of the i-th grid-connected converter, G opk (s) and G onk (s) are the positive and negative sequence output admittances of the k-th grid-connected converter. It can be seen that the online identification result of the grid-connected point admittance of the i-th new energy grid-connected converter is the sum of the admittances of all components of the system except the grid-connected converter itself.

[0082] Further, the online identification results of the N new energy grid-connected converters are operated to obtain the following formula:

[0083]

[0084] wherein, G op (s) and G on (s) are the total positive and negative sequence output admittances of the new energy grid-connected converter, i.e. the sum of the positive and negative sequence output admittances of the N new energy grid-connected converters.

[0085] According to the above formula, the N new energy grid-connected converter admittance online identification results are operated to obtain the positive sequence output total admittance G op (s) and the negative sequence output total admittance G on (s) as shown in the following formula:

[0086]

[0087] Further, the grid positive and negative sequence admittance and the new energy grid-connected converter positive and negative sequence output total admittance are calculated to obtain the grid impedance Z g (s) and the output total impedance Z o (s) of the centralized new energy grid-connected system as shown in the following formula:

[0088]

[0089] Under the condition of three-phase balance, the new energy generation grid-connected system often works in the positive sequence environment, and the main factor affecting the system stability is the positive sequence impedance characteristic, so in this embodiment, only the influence of the system positive sequence admittance or impedance on the system is analyzed.

[0090] The centralized new energy grid-connected system impedance online identification method flow chart is shown in Figure 4 , and the specific steps are as follows:

[0091] Step one: make a single new energy grid-connected converter work in the grid-connected mode, and analyze the current grid positive sequence admittance G gp (s) and the negative sequence admittance G gn (s) through pulse response online identification;

[0092] Step two: when N new energy grid-connected converters are normally grid-connected, it is judged whether i≤N (the initial value of i is 1) is established, if yes, go to step three, if not, execute step four;

[0093] Step three: make the i-th new energy grid-connected converter online identify the positive sequence admittance G pi (s) and the negative sequence admittance G ni (s) at the current grid-connected point, update i=i+1 and go to step two;

[0094] Step four: operate the N new energy grid-connected converter admittance online identification results to obtain the new energy grid-connected converter positive sequence output total admittance G op (s) and the negative sequence output total admittance G on (s);

[0095] Step two, calculate the required virtual impedance Z v (s) by using the adaptive virtual impedance algorithm, including:

[0096] The adaptive virtual impedance algorithm combines the grid impedance and the online identification result of the output impedance of the new energy grid-connected converter to obtain the system resonance frequency and the resonance component of the grid-connected point voltage, and dynamically adjusts the required virtual impedance compensation to achieve resonance suppression in a wide frequency range and improve the resonance suppression speed of the system. The specific idea of the algorithm is as follows: the system resonance frequency is obtained through the online identification result of the impedance, the virtual impedance phase angle θ v is adjusted according to the frequency band where the current system resonance frequency is located, the system resonance frequency is sent into a band-pass filter to obtain the resonance component of the grid-connected point voltage, and the virtual impedance amplitude Z v is adjusted according to the demand of the current resonance component for the resonance suppression speed; finally, the required virtual impedance Z v (s) is calculated according to the adjusted virtual impedance phase angle θ v and amplitude Z v .

[0097] Step two: according to the total output impedance Z o (s) of the centralized new energy grid-connected converter and the grid impedance Z g (s), the resonance frequency ω z is extracted by using a resonance frequency extraction model, and the resonance frequency extraction model is as follows:

[0098] ω z =minω

[0099]

[0100] wherein, s=jω, ω represents the angular frequency, and PM1 is the phase angle margin of Z g (s) / Z o (s).

[0101] The system resonance frequency ω z is obtained by solving the resonance frequency extraction model.

[0102] The system resonance frequency ω z of this step has two functions, one is in step 23, and the other is to obtain the resonance component of the grid-connected point voltage in step 25.

[0103] Step two: the reasonable value range [θ vmin , θ vmax ] of the virtual impedance phase angle is obtained by using a virtual impedance phase angle planning model.

[0104] Further, according to the influence law of the impedance coupling relationship on the system stability, the reasonable value range of the virtual impedance is obtained, Figure 5 and the equivalent circuit model of the system before and after the virtual impedance compensation is given, wherein I s(s) represents the output current of the centralized new energy power generation system, I g (s) represents the grid current, U g (s) is the grid voltage, Z o (s) represents the total output impedance of the new energy grid-connected converter, Z g (s) is the grid impedance, Z v (s) is the compensated virtual impedance, and satisfies Z v (s)=R v +sL v .according to Figure 5 The equivalent model of (a) yields the following expression for the grid current before virtual impedance compensation:

[0105]

[0106] Since the coupling effect between the grid impedance and the total output impedance of the grid-connected converter for new energy sources is the main reason for system resonance, and under the conditions that the grid voltage is stable and the output power, circuit parameters, and control loop of the centralized new energy power generation system remain unchanged, I in the above equation can be considered as... s (s), U g (s), Z o If H(s) is a stable term, then the stability of the grid current depends on the latter part of the above equation, H(s), as shown below:

[0107]

[0108] H(s) can be viewed as a vector with a forward gain of 1 and a feedback gain of Z. g (s) / Z o The closed-loop transfer function of the negative feedback control system (s) is, according to control theory, only when Z... g (s) / Z o (s) When the improved Nyquist stability criterion is satisfied: at the cutoff frequency ω c Z g (s) and Z o (s) If the phase difference is less than 180°, i.e., the phase margin PM > 0, then the system is stable. The expression for the phase margin PM is:

[0109] PM = 180° - (∠Z) g (jω c )-∠Z o (jω c ))

[0110] As can be seen from the above equation, changes in impedance coupling can affect system stability, thus triggering resonance. Impedance reshaping can be achieved through virtual impedance compensation using a new energy grid-connected converter, suppressing system resonance. Figure 5The equivalent model of (b) shows that the virtual impedance compensation is essentially adjusting the total output impedance of the new energy grid-connected converter to ensure that the system has sufficient phase margin. In the present application, each new energy grid-connected converter needs to simultaneously perform power output control and virtual impedance compensation. In order to minimize the impact of virtual impedance compensation on the power output control of the new energy grid-connected converter, a resistive-inductive virtual impedance compensation is adopted to achieve the safe and stable operation of the new energy grid-connected converter. When selecting the parameters of the resistive-inductive virtual impedance, both the phase angle and the amplitude can be considered. The phase angle of the virtual impedance can reflect the proportion of the resistive component and the inductive component in the virtual impedance. The smaller the phase angle, the greater the proportion of the resistive component, and the smaller the proportion of the inductive component. Meanwhile, the resistive and inductive virtual impedance corresponds to the consumption of active and reactive power, respectively. When the system resonance frequency is low, the resistive-inductive virtual impedance has good resonance suppression characteristics, but the proportion of the resistive component should be reduced as much as possible to reduce its consumption of active power and avoid affecting the power output control. As the system resonance frequency gradually increases, the impedance value corresponding to the inductive virtual impedance increases, and the proportion of the inductive component should be reduced to ensure good resonance suppression characteristics. In addition, the smaller the amplitude of the virtual impedance, the faster the suppression speed of the grid resonance, and the more obvious the resonance suppression effect. However, too small amplitude of the virtual impedance will cause more current to be absorbed, resulting in a decrease in the overall efficiency of the system. Therefore, a suitable size of the amplitude of the virtual impedance needs to be selected to suppress the distribution network resonance while minimizing power loss. In summary, the phase angle and amplitude of the virtual impedance need to be reasonably valued to ensure the normal power output of the new energy grid-connected converter while minimizing the power consumption caused by the virtual impedance compensation and maximizing the system resonance suppression performance. First, the reasonable value range of the phase angle of the virtual impedance can be calculated by using the virtual impedance phase angle planning model as shown below;

[0111] The virtual impedance phase angle planning model is:

[0112] θ vmin =min∠Z v (s),θ vmax =max∠Z v (s)

[0113]

[0114] Z v (s) is the virtual impedance to be compensated, ∠Z v (s) represents the impedance phase angle of Z v (s), Z' o (s c ) represents the total output impedance of the new energy grid-connected converter at ω c after impedance remodeling, and the intermediate variable s c =jω c, ω c is Z g (s) / Z' o (s) of the virtual impedance, PM is Z g (s) / Z' o (s) of the virtual impedance, R g is a virtual resistance, L g is a virtual inductance, ∠Z g (s c ) is Z g (s c ) of the impedance phase angle, ∠Z' o (s c ) is Z' o (s c ) of the impedance phase angle, θ lim is the minimum virtual impedance phase angle that N new energy grid-connected converters can compensate, Z lim is the minimum virtual impedance amplitude that N new energy grid-connected converters can compensate, which can comprehensively consider the remaining capacity resources, efficiency indicators and loss requirements of the new energy grid-connected converters in addition to power output. From the above formula, the value range [θ vmin , θ vmax ] of the virtual impedance phase angle that can ensure the system to be in a stable state under the current impedance condition can be obtained.

[0115] Step two three, on the basis of the reasonable value range [θ vmin , θ vmax ], the optimal virtual impedance phase angle θ v is determined according to the frequency band in which the system resonance frequency ω z is located:

[0116]

[0117] Wherein, ω1 is the low frequency resonance boundary angular frequency, ω2 is the high frequency resonance boundary angular frequency. In the embodiment, ω1 takes the angular frequency corresponding to 100Hz, and ω2 takes the angular frequency corresponding to 1kHz. Step two four, the reasonable value range [Z vmin , Z vmax ] of the virtual impedance amplitude is obtained by using the virtual impedance amplitude planning model:

[0118] Z vmin = min |Z v (s) |, Z vmax = max |Z v (s) |

[0119]

[0120] Step two five, in the reasonable value range [Z vmin , Zvmax the grid-connected point voltage effective value U pcc and the grid-connected point voltage resonance component effective value U pcch , according to the demand of the current grid-connected point voltage resonance component on the resonance suppression speed, determine the optimal virtual impedance amplitude Z v :

[0121]

[0122] Step two, according to the optimal virtual impedance phase angle θ v and the optimal virtual impedance amplitude Z v , determine the required compensation virtual impedance Z v (s):

[0123]

[0124] s is the Laplace operator;

[0125] Step three, send the required compensation virtual impedance Z v (s) into the control loop of the grid-connected converter, and control the grid-connected converter based on the resonance suppression method of virtual impedance compensation;

[0126] The resonance suppression method based on virtual impedance compensation absorbs the resonance current in the system by controlling N grid-connected new energy converters, so as to equivalent the N grid-connected new energy converters to additional resistive-inductive virtual impedance in the resonance frequency band, so as to reshape the output impedance of the grid-connected converter and suppress the system resonance, which specifically includes:

[0127] Firstly, the collected grid-connected point voltage signal u pcc is filtered by a band-pass filter to obtain the resonance component u pcch of the grid-connected point voltage, and the grid-connected point voltage effective value U pcc and the grid-connected point voltage resonance component effective value U pcch are calculated, and the transfer function G u (s) of the band-pass filter is as follows:

[0128]

[0129] Wherein, ω r is the center frequency, and ω u is the band-pass width. The function of the band-pass filter is to extract the signal component with the frequency of ω r , so the value of ω r should be consistent with the system resonance frequency, and the band-pass width ω u should be consistent with the actual impedance identification frequency step accuracy.

[0130] Further, the required compensation virtual impedance Z v (s) is obtained by an adaptive virtual impedance algorithm, and the current given value is obtained by using the following relationship:

[0131]

[0132] where i href is the given value of the virtual impedance compensation current loop of each new energy grid-connected converter, that is, the resonance current required to be absorbed by each new energy grid-connected converter; Z v (s) is the current required compensation virtual impedance, at this time, the N grid-connected converters are equivalent to additional resistive and inductive impedance in the resonance frequency band.

[0133] Further, to accurately track the alternating current resonance current at a specific frequency, the current control is realized in the αβ coordinate system, and the current controller uses a proportional-resonant (PR) controller, and the transfer function G i (s) is as follows:

[0134]

[0135] where K pa and K ra are the proportional coefficient and the resonant coefficient of the PR controller, ω n is the resonance frequency of the PR controller, and ω d is the resonance bandwidth. The PR controller is used to track and control the signal component with the frequency of ω n , so the value of ω n should be consistent with the system resonance frequency, and the resonance bandwidth ω d should be consistent with the actual impedance identification frequency step accuracy.

[0136] To improve the extraction and tracking accuracy of the resonance component, the obtained system resonance frequency ω z is sent to the band-pass filter and the proportional-resonant controller, and the frequency control parameters are adjusted as follows:

[0137] ω r = ω n = ω z

[0138] Based on the above, the required compensation virtual impedance Z v (s) is sent to the control loop of the new energy grid-connected converter, and the process of controlling the new energy grid-connected converter includes:

[0139] Step three, according to the system resonance frequency ω z , the resonance frequency ω n of the proportional-resonant controller and the center frequency ω r of the band-pass filter are adjusted.;

[0140] Step three two, using a band-pass filter to filter the grid point voltage signal u pcc , to obtain the resonant component of the grid point voltage u pcch ;

[0141] Step three three, according to the resonant component of the grid point voltage u pcch And the required compensation virtual impedance Z v (s) to get the new energy grid-connected converter virtual impedance compensation current loop given value

[0142] Step three four, the α, β axis components i href Of the new energy grid-connected converter virtual impedance compensation current loop given value i hαref , i hβref Respectively, the difference between the actual value is obtained Two difference values are respectively sent into the proportional-resonant controller to obtain α, β axis modulation signal, and the Clark inverse transformation is performed on the α, β axis modulation signal to obtain the virtual impedance compensation modulation signal v h , the virtual impedance compensation modulation signal v h And the power modulation signal v b Superimposed PWM modulation to obtain the switching control signal of the new energy grid-connected converter, and the new energy grid-connected converter is controlled according to the switching control signal.

[0143] The embodiment based on the above method provides a control system, and the overall control block diagram is as shown in Figure 7 , including a signal acquisition and processing unit, a coordinate transformation unit, a system impedance online identification unit, an adaptive virtual impedance calculation unit, a power current given calculation unit, a power output control unit, and a virtual impedance compensation unit.

[0144] The centralized new energy power generation system contains N LCL type new energy grid-connected converters, and each grid-connected converter is independently controlled to realize system impedance online identification, virtual impedance compensation and power output functions, Figure 7 The control flow block diagram of each grid-connected converter is given.

[0145] The signal acquisition and processing unit is used to acquire the current centralized new energy power generation system output current i pcc And the grid point voltage u pcc At the same time, the system impedance online identification unit and the virtual impedance compensation unit are sent in, the current new energy grid-connected converter grid-side filter inductor current i L2 , converter-side filter inductor current i L1 And filter capacitor voltage u c The coordinate transformation unit is used to acquire the current new energy grid-connected converter DC side voltage Vdc The voltage u at the current grid connection point is sent to the power output control unit and obtained through a phase-locked loop (PLL). pcc The phase angle θ is fed into the coordinate transformation unit and the power output control unit to calculate the effective value U of the grid-connected point voltage. pcc RMS value of the resonant component of the grid connection point voltage U pcch The data is fed into the adaptive virtual impedance calculation unit.

[0146] The coordinate transformation unit is used to adjust the setpoint i of the virtual impedance compensation current loop of the new energy grid-connected converter. href And the grid-side filter inductor current i of the new energy grid-connected converter L2 Perform a Clark transform to obtain its α and β axis components i. hαref i hβref and i L2α i L2β The signal is fed into the virtual impedance compensation unit, utilizing the grid connection point voltage u. pcc The phase angle θ affects the grid-side filter inductor current i of the LCL-type new energy grid-connected converter. L2 The current i of the filter inductor on the converter side L1 Filter capacitor voltage u c and grid connection point voltage u pcc Performing the Park transform yields its d-axis current components and q-axis voltage components: i L2d i L2q i L1d i L1q u cd u cq u pccd u pccq The current is fed into the power current setting calculation unit and the power output control unit.

[0147] The system impedance online identification unit is used to utilize Figure 4 The process shown obtains the grid impedance Z. g (s) and the total output impedance Z of the new energy grid-connected converter o (s) is fed into the adaptive virtual impedance calculation unit.

[0148] The power current setpoint calculation unit is used to calculate the power current setpoint value using instantaneous power theory and the following formula, and then send it to the power output control unit.

[0149]

[0150] Among them, P ref and Q ref These are the active power setpoint and reactive power setpoint of the grid-connected converter for new energy sources, i L2drefThe given value for the d-axis current of the grid-side filter inductor in the new energy grid-connected converter is i, which is the given value for the active power current. L2qref This is the given value for the q-axis current of the grid-side filter inductor in the new energy grid-connected converter, which is also the given value for the reactive power current.

[0151] The power output control unit employs a dual-loop, fully decoupled current control method in the dq coordinate system to achieve independent control and output of active and reactive power in the renewable energy grid-connected converter. The outer current control loop sets the grid-side filter inductor current setpoint i of the renewable energy grid-connected converter. L2derf i L2qerf Each with its actual value i L2d i L2q The difference is calculated and fed into a proportional-integral (PI) controller to obtain the setpoint i for the filter inductor current on the converter side. L1derf i L1qerf The inner current control loop sets the current setpoint i of the filter inductor on the converter side. L1derf i L1qerf Each with its actual value i L1d i L1q The difference is calculated and fed into a proportional-integral (PI) controller for calculation. The calculation results are then superimposed on the d-axis and q-axis decoupling feedforward components G. d (s), G q (s) Obtain the d-axis and q-axis modulation signals, and divide the modulation signals by the DC-side voltage V of the new energy grid-connected converter. dc After standardization and inverse Park transform, the power modulation signal v is obtained. b And fed into the virtual impedance compensation unit. d, q-axis decoupling feedforward components G d (s), G q (s) is shown in the following formula:

[0152]

[0153] Where L2 is the grid-side filter inductor of the LCL-type new energy converter, L1 is the converter-side filter inductor, and C d ω0 is the filter capacitor and ω0 is the fundamental angular frequency.

[0154] Adaptive virtual impedance calculation unit, used to utilize Figure 6 The system resonant frequency ω is calculated using the process shown. z and the virtual impedance Z to be compensated v (s) is fed into the virtual impedance compensation unit.

[0155] The virtual impedance compensation unit first uses a bandpass filter to process the acquired grid connection point voltage signal u. pcc Filtering is performed to obtain the resonant component u of the grid connection point voltage. pcchThe resonance component u of the grid-connected point voltage is sent to the signal collection and processing unit pcch Divided by the required compensation virtual impedance Z v (s) Get the new energy grid-connected converter virtual impedance compensation current loop given value i href The α, β axis current loop given value i is sent to the coordinate transformation unit hαref , i hβref Subtract the actual value i L2α , i L2β respectively, and the difference is sent to the proportional-resonant controller for operation to obtain the α, β axis modulation signal. The modulation signal is subjected to Clark inverse transformation to obtain the virtual impedance compensation modulation signal v h The power modulation signal v b is superimposed with the virtual impedance compensation modulation signal v h , and the superimposed signal is sent to the PWM modulation module for modulation to obtain the switching control signal, which controls the new energy grid-connected converter to perform virtual impedance compensation and power output. In this way, the grid is supported in power, the resonance current is absorbed, the impedance is reshaped, and the system stability is improved.

[0156] Although the present application is described herein with reference to particular embodiments, it is to be understood that these embodiments are merely exemplary of the principles and applications of the present application. It is therefore to be understood that numerous modifications can be made to the illustrative embodiments and that other arrangements can be devised without departing from the spirit and scope of the present application as defined by the appended claims. It is to be understood that different combinations of the features described herein can be made with different dependent claims and features described herein. It is also to be understood that features described in connection with a single embodiment can be used in other described embodiments.

Claims

1. A method for resonance suppression of a centralized new energy grid-connected system, the centralized new energy grid-connected system comprising N new energy grid-connected converters, N being a positive integer, characterized in that, The method comprises: S1, online identify new energy grid-connected converter output total impedance Z o (s) and grid impedance Z g (s); S2, calculating the required compensation virtual impedance Z v (s), comprising: S21, total impedance Z of the new energy grid-connected converter output o (s) and grid impedance Z g (s), extract the system resonance frequency ω z ; S22, acquire a reasonable value range [θ vmin ,θ vmax ] of the virtual impedance phase angle by using the virtual impedance phase angle planning model. vmin vmax S23, on the basis of reasonable value range [θ vmin , θ vmax ], according to the frequency band where the system resonance frequency ω z is located, determine the best virtual impedance phase angle θ v ; S24, acquiring a reasonable value range [Z of the virtual impedance amplitude by using a virtual impedance amplitude planning model vmin ,Z vmax ]; S25, on the basis of reasonable value range [Z vmin ,Z vmax ] of the virtual impedance, according to the demand of the current grid-connected point voltage resonance component on the resonance suppression speed, determine the optimal virtual impedance amplitude Z v ; S26, determining the optimal virtual impedance phase angle θ v with the optimal virtual impedance magnitude Z v , determining the required compensation virtual impedance s is the Laplacian operator; S3, the virtual impedance Z v (s) sending into a control loop of the new energy grid-connected converter to control the new energy grid-connected converter. 2.The resonance suppression method of the centralized new energy grid-connected system according to claim 1, characterized in that, The virtual impedance phase angle planning model is: θ vmin = min ∠Z v (s), θ vmax = max ∠Z v (s) Z v (s) is the virtual impedance needed to be compensated, ∠Z v (s) represents the impedance angle of Z v (s), Z' o (s c ) is the total output impedance of the new energy grid-connected converter at ω c after the impedance reshaping, the intermediate variable s c = jω c , ω c is the cut-off frequency of Z g (s) / Z' o (s), PM is the phase angle margin of Z g (s) / Z' o (s), R g is the virtual resistance, L g is the virtual inductance, θ lim is the minimum virtual impedance angle that N new energy grid-connected converters can compensate, Z lim is the minimum virtual impedance amplitude that N new energy grid-connected converters can compensate; ∠Z g (s c ) is the impedance angle of Z g (s c ), ∠Z' o (s c ) is the impedance angle of Z' o (s c ). 3.The resonance suppression method of the centralized new energy grid-connected system according to claim 2, characterized in that, Optimum virtual impedance phase angle θ v is: Wherein, ω1 is the low frequency resonance boundary angular frequency, ω2 is the high frequency resonance boundary angular frequency.

4. The resonance suppression method for centralized new energy grid-connected system according to claim 1, characterized in that, The virtual impedance amplitude planning model is: Z vmin = min |Z v (s) |, Z vmax = max |Z v (s) | wherein Z v (s) is the virtual impedance to be compensated, ∠Z v (s) represents the impedance phase angle of Z v (s), Z' o (s c ) is the total output impedance of the new energy grid-connected converter at ω c after the impedance reshaping, the intermediate variable s c = jω c , ω c is the cut-off frequency of Z g (s) / Z' o (s), PM is the phase angle margin of Z g (s) / Z' o (s), R g is the virtual resistance, L g is the virtual inductance, θ lim is the minimum virtual impedance phase angle that can be compensated by N new energy grid-connected converters, Z lim is the minimum virtual impedance amplitude that can be compensated by N new energy grid-connected converters; ∠Z g (s c ) is the impedance phase angle of Z g (s c ), ∠Z' o (s c ) is the impedance phase angle of Z' o (s c ).

5. The resonance suppression method for centralized new energy grid-connected system according to claim 4, characterized in that, Optimal virtual impedance magnitude Z v is: U pcc is the grid point voltage effective value, U pcch is the grid point voltage resonance component effective value. 6.The method of claim 1, wherein, According to the total impedance Z of the new energy grid-connected converter output o (s) and the grid impedance Z g (s), the system resonance frequency ω z : The resonance frequency extraction model is: ω z = min ω where s = jco, co represents the angular frequency, PM1is the Z g (s) / Z o phase margin of G(s). Solving the model for the resonant frequency gives the system resonant frequency ω z .

7. The resonance suppression method of the centralized new energy grid-connected system according to claim 1, characterized in that, Online identification of the output total impedance Z of a new energy grid-connected converter o (s) and grid impedance Z g (s), comprising: Let a single new energy grid-connected converter work in grid-connected mode, identify the current grid positive sequence admittance G gp (s) and negative sequence admittance G gn (s) online through impulse response analysis; When N grid-connected converters are simultaneously normally grid-connected, the N grid-connected converters are sequentially passed through pulse response analysis to identify the positive sequence admittance and the negative sequence admittance at the current grid-connected point: wherein G pi (s) and G ni (s) are the online identification results of positive sequence and negative sequence admittance at the point of common coupling of the i-th grid-connected converter, respectively, and G opk (s) and G onk (s) are the positive sequence output admittance and negative sequence output admittance of the k-th grid-connected converter, respectively. G pi (s) of N grid-connected converters ni (s) are calculated, and the total positive sequence output admittance G op (s) and the total negative sequence output admittance G on (s) of the new energy grid-connected converter are obtained: The positive sequence admittance G gp of the power grid The positive sequence output total admittance G op of the new energy grid-connected converter g The positive sequence output total impedance Z o of the new energy grid-connected converter 8.The method of claim 1, wherein, a virtual impedance Z v (s) A method of controlling a grid-connected inverter by feeding into a control loop of the grid-connected inverter. According to the system resonance frequency ω z Adjusting the resonance frequency ω of the proportional-resonant controller n And the center frequency ω of the band-pass filter r ; The grid-connected point voltage signal u is filtered by a band-pass filter pcc The resonant component u of the grid-connected point voltage is obtained by filtering pcch ; According to the resonant component u of the grid-connected point voltage pcch and the virtual impedance Z required for compensation v (s) obtaining a given value of the grid-connected converter virtual impedance compensation current loop The given value i of the grid-connected converter virtual impedance compensation current loop href The α, β axis components i of the given value hαref , i hβref The two difference values are respectively sent into the proportional-resonance controller to obtain the α, β axis modulation signals, and the Clark inverse transformation is performed on the α, β axis modulation signals to obtain the virtual impedance compensation modulation signal v h The virtual impedance compensation modulation signal v h is superimposed with the power modulation signal v b After the PWM modulation, the switching control signal of the grid-connected converter is obtained, and the grid-connected converter is controlled according to the switching control signal. 9.The resonance suppression method of the centralized new energy grid-connected system according to claim 8, characterized in that, According to the given value and the actual value of the grid-side filter inductor current and the converter-side filter inductor current of the grid-connected converter, a power modulation signal v b , comprising: The current control outer loop gives a given value i L2derf of the grid-side filter inductance current of the grid-connected converter L2qerf The actual value i L2d of the grid-side filter inductance current is subtracted respectively, and the two difference values are sent into proportional-integral controllers respectively to obtain the given value i L2q of the converter-side filter inductance current i L1derf , i L1qerf ; The current control inner loop gives a given value i L1derf , L1qerf The actual value i L1d , L1q is subtracted respectively, and the two difference values are sent into proportional-integral controllers respectively for operation, and the operation results are superimposed with d, q axis decoupling feedforward components respectively to obtain d, q axis modulation signals, and the d, q axis modulation signals are divided by the grid-connected converter DC side voltage V dc for normalization processing, and the normalized processing results are obtained after Park inverse transformation to obtain power modulation signals v b .

10. The control system of the resonance suppression method of the centralized new energy grid-connected system according to any one of claims 1 to 9, characterized in that, Comprise: A system impedance online identification unit online identifies the total impedance Z of the new energy grid-connected converter output o (s) and the grid impedance Z g (s); an adaptive virtual impedance calculation unit calculating a virtual impedance Z required for compensation v (s); a virtual impedance compensation unit for providing a virtual impedance Z v (s) sending into a control loop of the grid-connected converter to control the grid-connected converter.

Images