Network construction type converter fault ride-through control method and system based on power angle sensitivity adaptive adjustment

Abstract

The invention discloses a power angle sensitivity adaptive adjustment-based fault ride-through control method and a power angle sensitivity adaptive adjustment-based fault ride-through control system for a network-constructed converter. The method comprises the steps of collecting a power grid phase angle and an inverter output phase angle, establishing a power angle linearization mathematical model, calculating a sensitivity coefficient of active power to the phase angle, and obtaining an equivalent synchronous power coefficient; identifying an equivalent synchronous power coefficient on line by using a recursive least square algorithm with a forgetting factor; constructing a self-adaptive regulator to calculate an expected value of a power angle amplification coefficient Ks based on a preset target synchronization stiffness; and introducing a projection operator to carry out amplitude limiting and filtering on the Ks so as to realize self-adaptive adjustment of the synchronous power coefficient. When the power grid breaks down or the phase angle fluctuates severely, the power angle amplification coefficient can be automatically adjusted according to real-time identification, excessive deviation of the power angle to the unstable interval can be restrained, the low-voltage ride-through capacity and transient stability of the grid-forming type converter are improved, and high reliability and robustness are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of power electronics, in particular to a grid-forming converter fault ride-through control method based on power angle sensitivity adaptive adjustment. BACKGROUND

[0002] With the increasing proportion of large-scale new energy grid-connected, the overall inertia and transient stability margin of the power system significantly decreases, and a large number of grid-connected converters need to have the stability support capability similar to synchronous generators under fault and disturbance conditions. The grid-forming converter has become a key interface device in the new power system because it can actively establish voltage amplitude and phase angle. However, during grid-connected operation, the active output of the grid-forming converter is highly dependent on the power angle adjustment. When the power grid fails, the voltage drops or the phase angle jumps, the power angle may quickly increase, and if it exceeds the stable operation interval, it is easy to cause step-out, exit grid-connected or produce oscillation, thereby seriously affecting the transient stability and fault ride-through performance of the system.

[0003] The existing grid-forming converter fault ride-through control method is mostly based on the fixed parameter virtual synchronous generator control, droop coefficient setting or simple amplitude limiting strategy, usually assuming that the synchronous power coefficient and grid parameters are basically constant, and it is difficult to reflect the changes of grid impedance, operating point and fault strength in time. With the change of working conditions, the fixed parameters may cause insufficient or excessive synchronous stiffness, resulting in power angle divergence, oscillation or even instability. On the other hand, the existing adaptive or parameter identification method mostly does not identify the power angle-active power sensitivity online, and lacks effective parameter constraint mechanism. The control parameters obtained by adaptive adjustment under extreme operating conditions may exceed the limit, further deteriorating the system stability. SUMMARY

[0004] The technical problem to be solved by the present application is to overcome the shortcomings of the prior art and provide a grid-forming converter fault ride-through control method based on power angle sensitivity adaptive adjustment, which identifies the power angle sensitivity (equivalent synchronous power coefficient) in real time during the grid-connected operation of the grid-forming converter, and combines adaptive adjustment and parameter limiting mechanism to perform bounded adaptive adjustment on the power angle amplification coefficient, thereby effectively suppressing power angle out-of-limit, improving fault ride-through capability and transient stability control under grid fault and phase angle mutation, to solve the problem that the grid-forming converter in the prior art lacks real-time identification and bounded adaptive adjustment of the equivalent synchronous power coefficient under grid fault and phase angle mutation, resulting in easy out-of-limit step-out of the power angle, insufficient fault ride-through capability and transient stability.

[0005] To solve the above technical problems, the present application adopts the following technical solutions:

[0006] Firstly, the application proposes a network configuration type converter fault ride-through control method based on power angle sensitivity adaptive adjustment, which is applied to the grid-connected operation process of the network configuration type converter. When the power grid fails or the phase angle of the power grid suddenly changes, the power angle amplification coefficient Ks is automatically adjusted according to the real-time identified power angle sensitivity, so that the equivalent synchronous power coefficient is kept within the preset safe range, and the loss of step instability caused by the power angle exceeding 90° is avoided; including:

[0007] The power grid phase angle and the inverter output phase angle are collected, a power angle linearization mathematical model is established, the sensitivity coefficient ΔP / Δθ of the active power to the phase angle is calculated, and the equivalent synchronous power coefficient is obtained;

[0008] The recursive least square algorithm with a forgetting factor is used to identify the equivalent synchronous power coefficient online;

[0009] Based on the pre-set target synchronous stiffness, an adaptive regulator is constructed to calculate the expected value of the power angle amplification coefficient Ks;

[0010] The projection operator is introduced to limit and filter Ks, and the adaptive adjustment of the synchronous power coefficient is realized.

[0011] Further, the power grid phase angle and the inverter output phase angle are collected, a power angle linearization mathematical model is established, the sensitivity coefficient ΔP / Δθ of the active power to the phase angle is calculated, and the equivalent synchronous power coefficient is obtained, including:

[0012] Under the condition of ignoring the line resistance, the equivalent circuit between the network configuration type inverter and the power grid can be represented as a voltage source E v ∠θ v Through the reactance X g The grid voltage U g ∠θ g The active power of the network configuration type converter is calculated by the following method,

[0013] ,

[0014] Where, θ v is the output voltage phase angle of the converter, θ g is the grid voltage phase angle, and δ is the power angle. The small signal linearization of the active power near the steady-state power angle δ0 is:

[0015] ,

[0016] Where, is the synchronous power coefficient, then:

[0017] ,

[0018] When |δ0|<90°, K Pδ> 0, the system has synchronous stiffness; when | δ0 | > 90°, K Pδ < 0, the system enters the unsynchronized region.

[0019] Considering the power-frequency-phase angle control structure of the virtual synchronous generator (VSG), the small signal model thereof can be simplified as:

[0020] ,

[0021] Wherein, J is the virtual rotational inertia, D is the virtual damping coefficient, ΔP c is the active power feedback, K s is the power angle amplification coefficient, and Δθ is the inverter port phase angle disturbance.

[0022] In the small signal range, Δδ can be approximately considered as K s · Δθ, and ΔP ≈ K Pδ · K s · Δθ is obtained, and the equivalent synchronous power coefficient is calculated by the following method,

[0023] ,

[0024] Further, the online identification of the equivalent synchronous power coefficient by using the recursive least squares algorithm with a forgetting factor comprises the following steps:

[0025] Step 1, constructing a linear identification model. Discretize ΔP ≈ K Pδ · K s · Δθ, and select the phase angle disturbance as the input and the grid-connected active power disturbance as the output according to the relationship between the power angle and the active power in the method, and construct a first-order linear identification model at the sampling time k:

[0026] ,

[0027] ,

[0028] Wherein, y(k) is the active power disturbance at the kth sampling time, K Pθ (k) is the parameter vector to be identified, and e(k) is the modeling error.

[0029] Step 2, establishing a recursive least squares update formula with a forgetting factor. On the basis of the given initial parameter estimate (0) and the covariance matrix P(0), introduce a forgetting factor λ ∈ (0, 1) for each sampling time k, k = 1, 2, … K, and update according to the following recursive formula:

[0030] ,

[0031] Wherein, K(k) is the gain vector, KPθ (k) is the parameter estimation result after the kth update, P(k) is the covariance matrix, and the forgetting factor λ is used to apply an exponential decay weight to historical data to improve the tracking ability of the algorithm to changes such as faults.

[0032] Step 3, identify the effect and parameter convergence test. First, calculate the identification residual at each sampling time:

[0033] ,

[0034] Construct a weighted mean square performance index based on the residual:

[0035] ,

[0036] When J(k) is less than a pre-set threshold, and the parameter change satisfies | K Pθ (k) - K Pθ (k-1) | < ε, it is considered that the current forgetting factor setting is reasonable, and the online identification result of the power angle sensitivity ΔP / Δθ is reliable; if J(k) is large for a long time or the parameter oscillation is obvious, the forgetting factor λ needs to be adjusted or the initial covariance matrix P(0) needs to be re-set to improve the identification accuracy and convergence speed. Through the above test, it can be ensured that the online identification result of the recursive least squares algorithm with forgetting factor in the method has high reliability.

[0037] Further, based on the pre-set target synchronization stiffness, an adaptive (STR) regulator is constructed to calculate the expected value of the power angle amplification coefficient K s , including:

[0038] Step 1, construct the equivalent model of the controlled object and the target performance index. Taking the power angle-active power relationship of the network type converter as the object, the phase angle disturbance is regarded as the input and the active power disturbance is regarded as the output, and a first-order equivalent model is established:

[0039] ,

[0040] Where, K Pθ (k) represents the equivalent synchronization power coefficient of the system at the kth sampling time, and the target synchronization stiffness K Pθ (k)* is pre-set in combination with the transient stability and damping requirements, and “making the actual synchronization stiffness approach K Pθ (k)*” is taken as the performance target of self-tuning regulation.

[0041] Step 2, obtain the current object parameter estimation. The aforementioned recursive least squares algorithm with forgetting factor is used to perform online identification on the equivalent model parameter K Pθ (k) to obtain its real-time estimation value The estimated value reflects the real synchronization stiffness between the grid-forming converter and the grid under the current working condition, and is the basis for parameter calculation of the self-tuning regulator.

[0042] Step 3, determine the equivalent control law and solve the control parameters. Under the idea of "determining the equivalent control", the target synchronization stiffness K Pθ (k) is regarded as the desired object parameter, and the online estimated K (k) is regarded as the current object parameter, and the control law is constructed to make the actual equivalent synchronization power coefficient tend to the target value. For the power angle amplification coefficient K s (k), the STR adaptive control law is constructed as follows:

[0043] ,

[0044] where K s,des (k) is the expected value of the power angle amplification coefficient given by the self-tuning regulator at the kth sampling time, and ε>0 is a small constant to prevent the denominator from being too small. Then, K s,des (k) is taken as the input of the subsequent projection limiting and smoothing update module to realize online self-tuning of the actual control parameter K s (k). When K (k) converges to within a set threshold, it is considered that the STR method has completed the adaptive tuning of the power angle amplification coefficient, and the system synchronization stiffness reaches the desired level.

[0045] Further, the projection operator is introduced to limit and filter K s , realizing adaptive adjustment of the synchronization power coefficient, including:

[0046] Step 1, set the allowed range of adaptive parameters. In this method, the power angle amplification coefficient K s is calculated by the self-tuning regulator to obtain the expected value K s,des . In order to avoid excessive control caused by too large value of K s , or insufficient system synchronization ability caused by too small value of K s , the allowed value range of the power angle amplification coefficient is first set according to the design experience of the converter controller and the transient stability requirement:

[0047] ,

[0048] where K s,min and K s,max are the lower and upper limits of the power angle amplification coefficient, respectively, used to limit the physical rationality and stability boundary of the adaptive parameter.

[0049] Step 2, construct the projection operator to realize parameter limiting. For the expected value K s,proj(k) introduces a projection operator to perform "projection" processing, forcibly mapping parameters exceeding the allowable interval back to the boundary range, specifically defined as:

[0050] ,

[0051] Through the above projection operator operations, the adaptively adjusted parameter K can be guaranteed. s,proj (k) is always within the preset range, thereby preventing the power angle amplification coefficient from diverging or changing drastically when there is a fault, a large disturbance or a large identification error, which would affect the stable operation of the system.

[0052] Step 3: Improve stability by combining with an adaptive update mechanism. This is done after obtaining the parameters processed by the projection operator. Subsequently, it can be further combined with smoothing updates or filtering stages, such as using first-order filtering to achieve gradual updates:

[0053] ,

[0054] Here, α is the smoothing factor. By combining the "projection operator + smooth update," the projection operator ensures that the parameters remain within a stable range, while suppressing sudden parameter changes, making the adaptive adjustment process more stable and reliable. During parameter operation, K can be monitored... s (k) Assess the rationality of the projection interval setting by checking for frequent boundary breaches and system oscillations, and readjust K if necessary. s,min K s,max Or the smoothing factor α.

[0055] On the other hand, the present invention provides a fault ride-through control device for a grid-type converter based on adaptive adjustment of power angle sensitivity, comprising:

[0056] The synchronous power coefficient calculation unit collects the grid phase angle and the inverter output phase angle, establishes a linearized mathematical model of the power angle, calculates the sensitivity coefficient of active power to the phase angle, and obtains the equivalent synchronous power coefficient.

[0057] The online identification unit uses a recursive least squares algorithm with a forgetting factor to identify the equivalent synchronization power coefficient online.

[0058] The power angle amplification factor calculation unit constructs an adaptive regulator based on a pre-set target synchronization stiffness to calculate the expected value of the power angle amplification factor;

[0059] The adaptive adjustment unit introduces a projection operator to limit and filter the power angle amplification coefficient, thereby achieving adaptive adjustment of the synchronous power coefficient.

[0060] Furthermore, the present invention proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in the present invention.

[0061] Meanwhile, the present invention proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed, it implements the steps of the method described in the present invention.

[0062] Finally, the present invention provides a computer program product comprising a computer program / instructions that, when executed by a processor, implement the steps of the method described in the present invention.

[0063] Compared with the prior art, the present invention, employing the above-described technical solution, has the following beneficial effects:

[0064] This application integrates the fault ride-through control method into the grid-type converter controller, executes the control steps in multi-scenario power grid operation environments, and realizes fault ride-through control of the grid-type converter.

[0065] This application achieves dynamic tracking of the synchronous power coefficient of a grid-connected converter under grid faults and phase angle abrupt changes by introducing online identification and adaptive adjustment of the power angle sensitivity ΔP / Δθ, overcoming the poor adaptability of traditional fixed-parameter control methods. Utilizing a recursive least squares algorithm with a forgetting factor, the equivalent synchronous power coefficient can be quickly and accurately obtained under non-stationary scenarios such as voltage dips and phase angle jumps, providing a reliable basis for subsequent control. Combined with the automatic calculation of the power angle amplification factor K by the self-tuning regulator, this application achieves this. s The expected value of K is obtained, and K is updated using the projection operator and smooth update mechanism. s Implement bounded adaptive adjustment to avoid system oscillations caused by parameter divergence and abrupt changes.

[0066] The technical solution of this application actively suppresses the excessive shift of the power angle to the unstable region during the fault period, reduces the risk of loss of synchronization, and significantly improves the low voltage ride-through capability and transient stability level of the grid-type converter, which has high engineering applicability and robustness. Attached Figure Description

[0067] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0068] Figure 1 This is a schematic diagram of the steps of the fault ride-through control method for grid-type converters based on adaptive adjustment of power angle sensitivity in this invention.

[0069] Figure 2 This is a schematic diagram of the grid-type converter and its grid-connected control system in this invention.

[0070] Figure 3 This is a schematic diagram of the adaptive power angle adjustment system of the grid-type converter in this invention. Detailed Implementation

[0071] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be described in detail below. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other implementation methods obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0072] The following describes, with reference to the accompanying drawings, a fault ride-through control method for a grid-type converter based on adaptive adjustment of power angle sensitivity, provided in an embodiment of this application.

[0073] Example 1: As Figure 1 As shown, this embodiment provides a fault ride-through control method for grid-type converters based on adaptive adjustment of power angle sensitivity, including:

[0074] S101, collect the grid phase angle and inverter output phase angle, establish a linearized mathematical model of the power angle, calculate the sensitivity coefficient ΔP / Δθ of active power to the phase angle, and obtain the equivalent synchronous power coefficient, including:

[0075] Neglecting line resistance, the equivalent circuit between the grid-connected inverter and the grid can be represented as a voltage source E. v ∠θ v Through reactance X g Connected grid voltage U g ∠θ g The active power of the grid-type converter is calculated using the following method.

[0076] ,

[0077] Where, θ v θ is the phase angle of the converter output voltage. g δ is the phase angle of the grid voltage, and δ is the power angle. In this implementation case, E v =380V, θ v =6°, X g =0.12pu, U g =380V, θ g =0°, P=25kW. Applying small-signal linearization to the active power near the steady-state power angle δ0, we have:

[0078] ,

[0079] in, Let be the synchronization power coefficient, then we have:

[0080] ,

[0081] When |δ0| < 90°, K Pδ When |δ0| > 90°, the system has synchronous stiffness; when K > 0, the system has synchronous stiffness. Pδ If the value is less than 0, the system enters the desynchronization region.

[0082] Considering the power-frequency-phase control structure of a virtual synchronous generator (VSG), its small-signal model can be simplified as follows:

[0083] ,

[0084] Where J is the virtual moment of inertia, D is the virtual damping coefficient, and ΔP c For active power feedback, K s Where J is the power angle amplification factor, and Δθ is the inverter port phase angle disturbance. In this implementation case, J = 0.45 kg·m. 2 D=18, voltage drop of 20%, Δθ=-10°, ΔP=-12kW.

[0085] Within the small signal range, Δδ≈K can be approximated. s Δθ, therefore ΔP≈K Pδ K s The equivalent synchronization power factor is calculated using the following method for Δθ.

[0086] ,

[0087] S102, the equivalent synchronization power coefficient is identified online using a recursive least squares algorithm with a forgetting factor. The specific steps are as follows:

[0088] Step 1, construct a linear identification model. Assume ΔP≈K Pδ K s Δθ is discretized. Considering the relationship between the power angle and active power in this method, the phase angle disturbance is selected as the input and the grid-connected active power disturbance as the output. A first-order linear identification model is constructed at sampling time k:

[0089] ,

[0090] ,

[0091] Where y(k) is the active power disturbance at the k-th sampling time, K Pθ(k) represents the vector of parameters to be identified, and e(k) represents the modeling error.

[0092] Step 2: Establish the recursive least squares update formula with a forgetting factor. Given initial parameter estimates... Based on (0) and the covariance matrix P(0), a forgetting factor λ∈(0,1) is introduced and updated for each sampling time k, k=1,2,…K, according to the following recursive formula. In this embodiment, λ=0.96:

[0093] ,

[0094] Where K(k) is the gain vector, K Pθ P(k) represents the parameter estimation result after the kth update, P(k) is the covariance matrix, and the forgetting factor λ is used to apply exponential decay weights to historical data to improve the algorithm's ability to track changes in operating conditions such as faults.

[0095] Step 3, identification effect and parameter convergence test. First, calculate the identification residual at each sampling time:

[0096] ,

[0097] Constructing weighted mean square performance indices based on residuals:

[0098] ,

[0099] When J(k) is less than a preset threshold, and the parameter change satisfies |K Pθ (k)- K Pθ When (k-1) |<ε, the current forgetting factor setting is considered reasonable, and the online identification result of the power angle sensitivity ΔP / Δθ is reliable. If J(k) is too large for a long time or the parameter oscillation is obvious, the forgetting factor λ needs to be adjusted appropriately or the initial covariance matrix P(0) needs to be reset to improve the identification accuracy and convergence speed. Through the above tests, it can be guaranteed that the online identification result of the recursive least squares algorithm with forgetting factor in this method has high reliability.

[0100] S103, based on the pre-set target synchronization stiffness, construct an adaptive regulator to calculate the power angle amplification factor K. s The expected value, K in this embodiment s The allowed range is [0.5, 3.0]. The specific steps are as follows:

[0101] Step 1: Construct the equivalent model and target performance indicators of the controlled object. Taking the power angle-active power relationship of a grid-type converter as the object, and considering phase angle disturbance as input and active power disturbance as output, a first-order equivalent model is established:

[0102] ,

[0103] Among them, K Pθ (k) Characterize the equivalent synchronization power coefficient of the system at the k-th sampling time. Based on transient stability and damping requirements, the target synchronization stiffness K is preset. Pθ (k)*, K in this embodiment Pθ (k)*=120kW / rad, which will "make the actual synchronous stiffness approximate K". Pθ (k)*” is used as the performance target for self-tuning regulation.

[0104] Step 2: Obtain the current object parameter estimate. Using the aforementioned recursive least squares algorithm with a forgetting factor, estimate the equivalent model parameters K. Pθ (k) Perform online identification to obtain its real-time estimate. This estimate reflects the true synchronization stiffness between the grid-type converter and the power grid under current operating conditions, and serves as the basis for parameter calculations by the self-tuning regulator.

[0105] Step 3: Determine the equivalent control law and solve for the control parameters. Under the concept of "determining the equivalent control," the target synchronization stiffness K is... Pθ (k)* is considered as the desired object parameter, and will be estimated online. Treating these as parameters of the current object, a control law is constructed to make the actual equivalent synchronous power coefficient approach the target value. For the power angle amplification factor K... s (k), construct the STR adaptive control law:

[0106] ,

[0107] Where K s,des (k) represents the expected value of the power angle amplification coefficient given by the self-tuning regulator at the k-th sampling time, and ε>0 is a small constant to prevent the denominator from being too small. Subsequently, K... s,des (k) serves as the input to the subsequent projection limiting and smoothing update modules, thereby controlling the actual parameter K. s Online self-tuning of (k). When the convergence reaches within the set threshold, it is considered that the STR method has completed the adaptive tuning of the power angle amplification factor, and the system synchronous stiffness has reached the desired level.

[0108] S104, Introducing the projection operator pair K s Limiting and filtering are performed to achieve adaptive adjustment of the synchronization power coefficient. The specific steps are as follows:

[0109] Step 1: Set the allowable range for the adaptive parameters. In this method, the power angle amplification factor K... s The expected value K is calculated by the self-tuning regulator. s,des In order to avoid K sIf the value is too large, the control will be too aggressive; if the value is too small, the system synchronization capability will be insufficient. Therefore, based on the design experience of the converter controller and the transient stability requirements, the allowable range of the power angle amplification factor should be preset.

[0110] ,

[0111] Among them, K s,min and K s,max These are the lower and upper limits of the power angle amplification factor, respectively, used to define the physical rationality and stability boundaries of the adaptive parameters.

[0112] Step 2: Construct a projection operator to implement parameter limiting. This is for the desired value K of the self-tuning regulator output. s,proj (k) introduces a projection operator to perform "projection" processing, forcibly mapping parameters exceeding the allowable interval back to the boundary range, specifically defined as:

[0113] ,

[0114] Through the above projection operator operations, the adaptively adjusted parameter K can be guaranteed. s,proj (k) is always within the preset range, thereby preventing the power angle amplification coefficient from diverging or changing drastically when there is a fault, a large disturbance or a large identification error, which would affect the stable operation of the system.

[0115] Step 3: Improve stability by combining with an adaptive update mechanism. This is done after obtaining the parameters processed by the projection operator. Subsequently, it can be further combined with smoothing updates or filtering stages, such as using first-order filtering to achieve gradual updates:

[0116] ,

[0117] Here, α is the smoothing factor. By combining the "projection operator + smooth update," the projection operator ensures that the parameters remain within a stable range, while suppressing sudden parameter changes, making the adaptive adjustment process more stable and reliable. During parameter operation, K can be monitored... s (k) Assess the rationality of the projection interval setting by checking for frequent boundary breaches and system oscillations, and readjust K if necessary. s,min K s,max Or the smoothing factor α.

[0118] like Figure 2 The diagram shown illustrates a grid-connected converter and its grid-connected control system according to an embodiment of the present invention. The grid-connected converter includes a DC-side capacitor C. d Converter bridge, filter inductor L and its series resistor R, parallel filter capacitor C, grid-connected inductor L g and grid-connected resistor Rg Main circuit components, DC side voltage is u dc The AC output voltage of the converter is V. c The grid connection point voltage is u c The voltage connected to the power grid is u g The angular frequency of the power grid is ω g The converter output current is denoted as i. L The grid-connected current is denoted as i l .

[0119] At the control level, the AC side voltage and current of the converter are transformed into I by the abc / dq coordinate transformation module. dq E dq The signal is received and sent to the power calculation module to calculate the active power P and reactive power Q. The active power loop adopts a virtual synchronous machine control structure, and the active power command P is used to calculate the active power P and reactive power Q. ref The difference between the actual active power P and the actual active power P is processed by the virtual damping element D and the moment of inertia element 1 / Js in the virtual synchronous machine structure to obtain the frequency deviation Δω. This frequency deviation is then integrated by the element 1 / s to generate the phase angle deviation Δθ, which is then compared with the angular velocity ω passing through the power grid. n The integrated reference phase angles are superimposed to form the phase angle reference value θ. ref The reactive power loop uses reactive power command Q. ref The deviation from the actual reactive power Q is the input, which is then processed by the reactive power regulator G. Q (s) Generates voltage parameters, superimposed with voltage reference U ref With operating voltage U n The difference is used to form a voltage control reference signal.

[0120] The voltage control section includes the outer voltage loop controller G. v (s), Current inner loop controller G c (s) and a voltage synthesis module, the voltage synthesis module converting the voltage command u cref This is converted into a modulation reference voltage vector, which, combined with the PWM module, drives the converter bridge switch to operate, thereby controlling the converter output voltage v. cv and grid connection point voltage u c This enables grid-connected operation of grid-type converters and the aforementioned fault ride-through control method.

[0121] like Figure 3 The diagram shown is a schematic of a grid-type converter power angle adaptive adjustment system according to an embodiment of the present invention. The system takes the active power signal P of the grid-type converter as input and generates a phase angle reference value θ through virtual inertia, virtual damping, and power angle amplification factors. ref This is combined with the aforementioned fault ride-through control method based on adaptive adjustment of power angle sensitivity.

[0122] The difference between the active power command and the actual active power is processed in the pre-processing stage to form the active power deviation, which is then fed into the virtual moment of inertia stage 1 / Js to obtain the frequency deviation Δω. The frequency deviation is related to the grid reference angular velocity ω. n The instantaneous angular velocity signal of the virtual synchronizer is obtained by superposition and passing through the adder node. This angular velocity signal is input to the virtual damping element D, and its output is fed back to the aforementioned adder node to form a damped frequency regulation channel, thereby achieving damping suppression of power disturbances. On the other hand, the frequency deviation Δω is integrated by the 1 / s element to obtain the phase angle deviation Δθ.

[0123] Phase angle deviation Δθ enters the power angle amplification factor module K s The success angle adjustment amount δ is generated. In this invention, the K... s The adaptive adjustment algorithm for power angle sensitivity proposed in this invention, which combines "recursive least squares with forgetting factor + self-tuning regulator with projection operator," updates in real time, enabling online tuning and bounded constraints of the power angle amplification factor. (Power grid reference angular velocity ω) n The reference phase angle is obtained through the integrator 1 / s. This reference phase angle is added to the power angle adjustment at the final adder node, and the output is the phase angle reference value θ of the grid-type converter. ref It is used to drive the subsequent voltage synthesis and PWM modulation unit, thereby realizing the power angle control and fault ride-through control of the grid-type converter based on the adaptive adjustment of power angle sensitivity.

[0124] Example 2: This example provides a fault ride-through control device for a grid-type converter based on adaptive adjustment of power angle sensitivity, comprising:

[0125] The synchronous power coefficient calculation unit collects the grid phase angle and the inverter output phase angle, establishes a linearized mathematical model of the power angle, calculates the sensitivity coefficient of active power to the phase angle, and obtains the equivalent synchronous power coefficient.

[0126] The online identification unit uses a recursive least squares algorithm with a forgetting factor to identify the equivalent synchronization power coefficient online.

[0127] The power angle amplification factor calculation unit constructs an adaptive regulator based on a pre-set target synchronization stiffness to calculate the expected value of the power angle amplification factor;

[0128] The adaptive adjustment unit introduces a projection operator to limit and filter the power angle amplification coefficient, thereby achieving adaptive adjustment of the synchronous power coefficient.

[0129] Example 3: This example proposes a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the control method as described in this invention.

[0130] Example 4: This example proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed, it implements the steps of the method described in this invention.

[0131] Example 5: The present invention proposes a computer program product, including a computer program / instructions, which, when executed by a processor, implement the steps of the method described in the present invention.

[0132] It should be noted that the processing flow of embodiments 2-5 corresponds to the specific steps of the method provided in embodiment 1 of the present invention, and has the corresponding functional modules and beneficial effects of the method. Technical details not described in detail in this embodiment can be found in the method provided in embodiment 1 of the present invention.

[0133] The program code used to implement the methods of this application may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0134] The specific implementation schemes described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific implementation schemes of the present invention and are not intended to limit the scope of the present invention. Any equivalent changes and modifications made by those skilled in the art without departing from the concept and principles of the present invention should fall within the scope of protection of the present invention.

Claims

1. A fault ride-through control method for a grid-type converter based on adaptive adjustment of power angle sensitivity, characterized in that, include: Collect the phase angle of the power grid and the output phase angle of the inverter, establish a linearized mathematical model of the power angle, calculate the sensitivity coefficient of active power to the phase angle, and obtain the equivalent synchronous power coefficient. The equivalent synchronization power coefficient is identified online using a recursive least squares algorithm with a forgetting factor. Based on the pre-set target synchronization stiffness, an adaptive regulator is constructed to calculate the expected value of the power angle amplification factor; By introducing a projection operator to limit and filter the power angle amplification factor, adaptive adjustment of the synchronous power factor is achieved.

2. The method according to claim 1, characterized in that, The process involves collecting the grid phase angle and the inverter output phase angle, establishing a linearized mathematical model of the power angle, calculating the sensitivity coefficient of active power to the phase angle, and obtaining the equivalent synchronous power coefficient, including: The active power P of the grid-type converter is calculated using the following method. ， Among them, U g E is the grid voltage. v As a voltage source, X g For reactance, θ v θ is the phase angle of the converter output voltage. g δ is the phase angle of the grid voltage, and δ is the power angle; Small-signal linearization of the active power near the steady-state power angle δ0 yields: ， Among them, let Let be the synchronization power coefficient, then we have: ， When |δ0| < 90°, K Pδ When |δ0| > 90°, the system has synchronous stiffness; when K > 0, the system has synchronous stiffness. Pδ If the value is less than 0, the system enters the desynchronization region; Considering the power-frequency-phase angle control structure of the virtual synchronous generator (VSG), its small-signal model simplifies to: ， Where J is the virtual moment of inertia, D is the virtual damping coefficient, and ΔP c For active power feedback, K s Δθ is the power angle amplification factor, and Δθ is the inverter port phase angle disturbance. Let Δδ≈K within the small signal range. s Δθ, therefore ΔP≈K Pδ K s The equivalent synchronization power factor is calculated using the following method for Δθ: 。 3. The method according to claim 1, characterized in that, The online identification of equivalent synchronization power coefficients using a recursive least squares algorithm with a forgetting factor includes: Step 1: Construct a linear identification model: ΔP≈K Pδ K s Δθ is discretized, and considering the relationship between the power angle and active power, the phase angle disturbance is selected as the input and the grid-connected active power disturbance is selected as the output. A first-order linear identification model is constructed at sampling time k: ， ， Where y(k) is the active power disturbance at the k-th sampling time, K Pθ (k) is the vector of parameters to be identified, and e(k) is the modeling error; Step 2: Establish the recursive least squares update formula with forgetting factor: Given initial parameter estimates Based on (0) and the covariance matrix P(0), a forgetting factor λ∈(0,1) is introduced, and the data is updated for each sampling time k according to the following recursive formula: ， Where K(k) is the gain vector, P(k) represents the parameter estimation result after the k-th update, where P(k) is the covariance matrix and the forgetting factor λ is used to apply exponential decay weights to the historical data. Step 3: Identification effect and parameter convergence test: First, calculate the identification residual at each sampling time: ， Constructing weighted mean square performance indices based on residuals: ， When J(k) is less than a preset threshold, and the parameter change satisfies |K Pθ (k)- K Pθ When (k-1) |<ε, the current forgetting factor setting is considered reasonable, and the online identification result of the power angle sensitivity ΔP / Δθ is reliable; if J(k) is too large for a long time or the parameter oscillation is obvious, the forgetting factor λ should be adjusted appropriately or the initial covariance matrix P(0) should be reset to improve the identification accuracy and convergence speed.

4. The method according to claim 3, characterized in that, Based on a pre-defined target synchronization stiffness, an adaptive regulator is constructed to calculate the expected value of the power angle amplification factor, including: Step 1: Construct the equivalent model of the controlled object and the target performance index: Taking the power angle-active power relationship of a grid-type converter as the object, and considering phase angle disturbance as input and active power disturbance as output, a first-order equivalent model is established: ， Among them, K Pθ (k) Characterizes the equivalent synchronization power coefficient of the system at the k-th sampling time, and pre-sets the target synchronization stiffness K based on transient stability and damping requirements. Pθ (k)*, and will make the actual synchronous stiffness approximate K. Pθ (k)* serves as the performance target for self-tuning regulation; Step 2: Obtain the current object parameter estimate: Using the aforementioned recursive least squares algorithm with a forgetting factor, the equivalent model parameter K is... Pθ (k) Perform online identification to obtain its real-time estimate. ; Step 3: Determine the equivalent control law and solve for the control parameters: Target synchronization stiffness K Pθ (k)* is considered as the desired object parameter, and will be estimated online. Treating the current object parameters, a control law is constructed to make the actual equivalent synchronous power coefficient approach the target value, for the power angle amplification factor K. s (k), construct the STR adaptive control law: ， Where K s,des (k) represents the expected value of the power angle amplification coefficient given by the self-tuning regulator at the k-th sampling time, and ε>0 is a small constant to prevent the denominator from being too small; subsequently, K s,des (k) serves as the input to the subsequent projection limiting and smoothing update modules, thereby controlling the actual parameter K. s (k) online self-tuning, when When the convergence reaches within the set threshold, it is considered that the STR method has completed the adaptive tuning of the power angle amplification factor, and the system synchronous stiffness has reached the desired level.

5. The method according to claim 4, characterized in that, By introducing a projection operator to limit and filter the power angle amplification factor, adaptive adjustment of the synchronization power factor is achieved, including: Step 1: Set the allowable range for the adaptive parameters: First, based on the design experience of converter controllers and transient stability requirements, the allowable range of power angle amplification factor is pre-defined: ， Among them, K s,min and K s,max These are the lower and upper limits of the power angle amplification factor, respectively, used to limit the physical rationality and stability boundaries of the adaptive parameters; Step 2: Construct a projection operator to implement parameter limiting: The desired value K of the self-tuning regulator output s,proj (k) introduces a projection operator to perform projection processing, forcibly mapping parameters exceeding the allowable interval back to the boundary range, specifically defined as: ， Step 3: Implement gradual updates using first-order filtering. ， Where α is the smoothing factor; During parameter operation, K is monitored. s (k) Assess the rationality of the projection interval setting by checking whether the boundary is frequently touched and whether the system oscillates, and readjust K if necessary. s,min K s,max Or the smoothing factor α.

6. A fault ride-through control device for a grid-type converter based on adaptive adjustment of power angle sensitivity, characterized in that, include: The synchronous power coefficient calculation unit collects the grid phase angle and the inverter output phase angle, establishes a linearized mathematical model of the power angle, calculates the sensitivity coefficient of active power to the phase angle, and obtains the equivalent synchronous power coefficient. The online identification unit uses a recursive least squares algorithm with a forgetting factor to identify the equivalent synchronization power coefficient online. The power angle amplification factor calculation unit constructs an adaptive regulator based on a pre-set target synchronization stiffness to calculate the expected value of the power angle amplification factor; The adaptive adjustment unit introduces a projection operator to limit and filter the power angle amplification coefficient, thereby achieving adaptive adjustment of the synchronous power coefficient.

7. A fault ride-through control device for a grid-type converter based on adaptive adjustment of power angle sensitivity according to claim 6, characterized in that, The adaptive adjustment unit introduces a projection operator to limit and filter the power angle amplification factor, thereby achieving adaptive adjustment of the synchronous power factor. It is configured to perform the following actions: Setting the allowable range of adaptive parameters: First, based on the design experience of the converter controller and the transient stability requirements, pre-set the allowable range of the power angle amplification factor: ， Among them, K s,min and K s,max These are the power angle magnification factors K s The lower and upper limits are used to define the physical rationality and stability boundaries of the adaptive parameters; Step 2: Construct a projection operator to implement parameter limiting: The desired value K of the self-tuning regulator output s,proj (k) introduces a projection operator to perform projection processing, forcibly mapping parameters exceeding the allowable interval back to the boundary range, specifically defined as: ， Step 3: Implement gradual updates using first-order filtering. ， Where α is the smoothing factor; During parameter operation, K is monitored. s (k) Assess the rationality of the projection interval setting by checking whether the boundary is frequently touched and whether the system oscillates, and readjust K if necessary. s,min K s,max Or the smoothing factor α.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the method as described in any one of claims 1 to 5.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the computer program is executed, it implements the steps of the method as described in any one of claims 1 to 5.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method according to any one of claims 1 to 5.

Images