Online evaluation and control design method for multi-machine grid-connected stability margin of new energy station

Abstract

The invention provides a new energy station multi-machine grid-connected stability margin online evaluation and control design method, which comprises the following steps: step 010, station multi-machine model construction: establishing a control port extension type unit model, and considering an electric network to form a station multi-machine closed-loop model of an explicit retention control loop structure; 020, stability margin online evaluation: putting forward an oscillation mode numerical search algorithm, and respectively substituting local / global information to realize unit and station margin online evaluation; 030, margin feedback control design: constructing a unit autonomous control law to realize zero-difference adjustment and quick response control of margin reference, and constructing a station correction control law to improve the accuracy of unit control; and 040, online refreshing of the margin control parameters: establishing an increment relationship between the damping control quantity and the stability margin based on the time-frequency sensitivity, and performing online calculation of the margin control parameters by controlling discretization. According to the method, the broadband dynamic and multi-machine coupling influence of the new energy field station can be considered, and the accuracy and rapidity of stability margin control can be considered.

Description

Technical Field

[0001] This invention relates to the field of grid connection stability analysis and stability control technology for new energy power plants, specifically, to an online evaluation and control design method for the stability margin of multi-unit grid connection in new energy power plants. Background Technology

[0002] Recent research and practical experience in grid-connected renewable energy projects show that renewable energy sources under converter control exhibit complex control interactions in weak electrical systems, which can easily lead to stability issues such as wideband oscillations. Improving the grid-connected stability of renewable energy sources has become a technical bottleneck restricting their mainstream operation.

[0003] Currently, considerable research has been conducted on this issue, resulting in various control strategies and methods aimed at improving stability characteristics under small disturbances and suppressing oscillations. For the stability control of a single grid-connected converter, there are two main categories: a) impedance reshaping methods based on the RLC characteristics of the circuit impedance; b) stability control methods based on state-space eigenvalue analysis. One approach involves determining key state variables / parameters affecting instability through sensitivity analysis and then making targeted parameter adjustments; the other involves constructing a state feedback controller to shape the eigenvalue positions to achieve a damping effect. Based on these two approaches, suppression methods with adaptive and online capabilities have also been proposed. However, these methods face the risk of control failure in multi-converter scenarios, primarily because the stability control design of a single converter does not consider the interaction between multiple machines. Therefore, some scholars have proposed a single-machine design concept based on "stability no-go zones" or "passivity," ensuring stable operation after interconnection of converters through strong conservative conditions. However, this design method cannot cope with various parameter variations in actual systems, making it essential to research stability control methods that involve multiple machines and possess adaptive capabilities.

[0004] Generally, the formulation of multi-machine stability control strategies requires an analytical model that reflects the interaction between multiple machines, primarily a frequency-domain nodal admittance model and a time-domain state-space model. The main approach to current multi-machine stability control is to identify the "oscillation source" based on multi-machine model analysis and then implement targeted suppression. The advantage of this method is that once the oscillation source (such as the dominant unit or control loop) is determined, single-machine stability control methods can be used for oscillation suppression. However, the ultimate focus remains on the single machine, making it difficult to account for the coupling effects of multiple machines and fully exploit the collaborative stabilization capabilities of multiple units. Summary of the Invention

[0005] To address the shortcomings of existing technologies, the purpose of this invention is to provide an online evaluation and control design method for the stability margin of multi-unit grid connection in new energy power plants, which can take into account the coupling effect of multiple units and meet the requirements of speed and accuracy of stable control.

[0006] According to one aspect of the present invention, an online evaluation method for the stability margin of multiple units in a new energy power station is provided, comprising:

[0007] Step 010: Establish the multi-unit frequency domain model of the power station: Based on the deviation input and output definitions of each controller in the unit, establish the state space model of the unit, and transform it to the frequency domain to obtain the extended unit model of the control port; Based on the branch-node correlation matrix, establish the node impedance matrix of the power station electrical network; Based on the circuit law, connect the unit model and the electrical network model to obtain the power station multi-unit closed-loop model that explicitly retains the control loop structure.

[0008] Step 020, online stability margin assessment: This includes two parts: model numerical search and margin assessment based on numerical search. The model numerical search is used to locate the weakly damped mode by traversing the real and imaginary parts of the Laplace operator in the multi-machine frequency domain model of Step 010. The margin assessment based on numerical search is used to realize the online assessment of the unit and station margins by substituting local / global time-varying parameters respectively.

[0009] Preferably, the multi-unit frequency domain model of the station in step 010 is a model that takes into account the electromagnetic transient broadband characteristics of the transmission network and can display the structure and parameters of all control loops in each unit.

[0010] The steps for constructing the multi-unit frequency domain model of the station include:

[0011] S1.1 Constructing an extended frequency domain model of the unit's control ports, specifically:

[0012] A small-signal state-space model is established, with the controller deviation input, AC filter circuit voltage drop, and DC filter circuit current as the state-space input quantities Δe, and the controller output, AC filter circuit current, and DC filter circuit voltage drop as the state-space output quantities Δγ.

[0013]

[0014] Δγ=CΔx+DΔe

[0015] Transforming the small-signal state-space model to the frequency domain yields the input-output transfer function G(s) between Δγ and Δx, defined as a forward matrix. This forward matrix is ​​presented in block diagonal form, with diagonal elements including the transfer functions K(s) of each controller and the admittance of the AC filter. DC filter impedance

[0016]

[0017] The algebraic equations in the small-signal state-space model of the combined unit and the definition of the state control input Δe are as follows:

[0018] Deviation definition: Δe=A1Δy+B1Δγ

[0019] Algebraic equation: Δy=A²Δγ+B²Δu

[0020] In the formula, Δy represents an intermediate variable and Δu represents the AC side voltage; by eliminating the intermediate variable, the input-output relationship between Δe and Δγ and Δu is obtained, which is called the self-feedback matrix F of the unit.

[0021]

[0022] S1.2 Establish the frequency domain model of the nodal impedance of the power station network, specifically as follows:

[0023] Based on the connection relationship between nodes and branches in the station, and using the branch-node correlation matrix A and the branch admittance matrix Y... br Construct the node admittance matrix Y of the power station network. net :

[0024] Y net =AY br A T

[0025] The nodal impedance matrix Z is obtained by inversion. net It is used to reflect the AC voltage and current relationship at the connection point between the unit and the power network, and is expressed as a function of the state output variable Δγ through the transformation matrix R:

[0026]

[0027] S1.3 Establish a multi-machine closed-loop frequency domain model for the station, specifically as follows:

[0028] The forward matrices of each unit are arranged diagonally in blocks to obtain the forward matrix G(s) of the multi-unit system of the station. This matrix retains all control loops within each unit.

[0029]

[0030] Arrange the feedback matrices of each unit diagonally in blocks to obtain the self-feedback matrix F of the multi-unit system at the station.

[0031]

[0032] According to the input-output definition of the self-feedback matrix F, it is divided into:

[0033]

[0034] Substituting the function of the state output variable Δγ into the partition form of the self-feedback matrix F, we obtain the feedback matrix H(s) of the multi-machine system at the station:

[0035]

[0036] Based on the forward matrix G(s) and feedback matrix H(s) of the multi-machine system at the station, a closed-loop frequency domain model of the multi-machine system at the station with explicit representation of the control loop is established:

[0037] L(s) = I + G(s)H(s)

[0038] In the formula, the diagonal elements of the forward matrix G(s) retain the control loop structure and parameters of each unit.

[0039] Preferably, step 020 includes:

[0040] Step 021, perform mode numerical search: Based on the initial search range and step size of the real and imaginary parts, generate frequency domain grid data and substitute it into the multi-machine frequency domain model; calculate the inverse of the determinant of the frequency domain model at each grid point to obtain the grid coordinates corresponding to the peak points; determine a new search range and step size with each peak point as the center to generate new frequency domain grid data; repeat the above peak location and grid generation steps until the grid coordinates at the peak points meet the accuracy requirements to obtain the oscillation mode of the system;

[0041] Step 022: Conduct online assessment of unit and station margins: Considering that the time-varying parameters of the unit itself are consistent with those of other units, substitute them into the multi-unit frequency domain model and calculate the unit margin assessment value through pattern search; Considering the state differences between units, obtain the time-varying parameters of each unit based on communication and substitute them into the multi-unit frequency domain model, and calculate the online assessment value of the station margin through the pattern numerical search.

[0042] Preferably, in step 021, a pattern numerical search is performed, including:

[0043] S21.1 Frequency domain grid generation, specifically:

[0044] Based on the ranges of the real and imaginary parts and the corresponding intervals, a set of frequency domain grid data is generated:

[0045]

[0046] In the formula, σ represents the real part and ω represents the imaginary part; where σ0 is the starting point of the real part search, Δσ is the search interval, k1 represents the number of search points, and σ0+k1Δσ is the ending point of the real part search; ω0 is the starting point of the imaginary part search, Δω is the search interval, k2 represents the number of search points, and ω0+k2Δω is the ending point of the imaginary part search.

[0047] S21.2, Matrix determinant calculation, specifically:

[0048] Substituting the generated frequency domain grid data into the multi-machine frequency domain model, the inverse determinant of the matrix is ​​calculated to obtain the distribution of the inverse determinant under this set of frequency domain grid data:

[0049]

[0050] In the formula, It is an input multi-machine frequency domain model and it incorporates grid data s test and system parameters It belongs to the category of numerical matrices;

[0051] S21.3, Peak point location, specifically:

[0052] Based on the obtained determinant result η, peak points are filtered by combining absolute value and gradient. Specifically, peak points are determined by the following conditions, corresponding to the grid data s. ij Potential oscillation mode:

[0053]

[0054] In the formula, the subscripts i and j represent the real and imaginary part numbers of the grid data, respectively, and ξ abs ξ gra These are the set absolute value threshold and gradient threshold, respectively; after iteration, the peak point grid coordinates within the initial search region are obtained, which represent the oscillation mode of the system.

[0055] λ={s ij}

[0056] Preferably, step 022 involves conducting an online assessment of the unit and station margins, including:

[0057] S22.1, Online assessment of unit margin, specifically:

[0058] Assuming that the status of other units is the same as that of this unit, only the time-varying parameters and damping control information of the unit itself are needed;

[0059] The parameters of the i-th generating unit are:

[0060]

[0061] In the formula, P i Q i V i D i These are the active power, reactive power, grid voltage, and damping control parameters of the i-th generating unit, respectively; the default states of other generating units are the same.

[0062] Substitute the parameters into the multi-machine frequency domain model A numerical search for the model is performed, and the model closest to the imaginary part is selected from the feedback oscillation modes. The real part is then extracted as the online evaluation value of the unit margin; that is, for the i-th unit, the margin is...

[0063] Unit margin: σ 机,i=Re[λ j ]

[0064] In the formula, λ j This is the oscillation mode closest to the imaginary axis;

[0065] S22.2, the online assessment method for station margin, specifically:

[0066] Substituting the time-varying parameters and damping control information of each unit, the parameter input information is as follows:

[0067]

[0068] Substitute the parameters into the multi-machine frequency domain model Furthermore, a numerical search for the mode is performed to filter out the oscillation mode closest to the imaginary part, and the real part is extracted as the online evaluation value of the station margin, such as:

[0069] Station margin: σ 场 =Re[λ j ].

[0070] According to a second aspect of the present invention, a multi-unit stability margin control design method for a new energy power station is provided, including step 030, margin feedback control law;

[0071] Step 030 and step 020 of the online evaluation method for multi-machine stability margin of new energy power stations constitute a closed-loop control structure; step 020 provides margin controlled object information for step 030, and step 030 provides damping control quantity information for step 020.

[0072] Step 030, margin feedback control, includes:

[0073] Step S031, Unit autonomous margin control, that is, to achieve rapid response control of the margin based on the local assessment value of the unit margin;

[0074] Step S032, station margin correction control, that is, generating reference values ​​for the margin control of each unit based on the global assessment value of the station margin, and compensating for the control deviation between the unit's autonomous margin control and the actual station margin.

[0075] Preferably, the autonomous control of the unit uses the deviation between the unit margin assessment value and the margin reference value as input, and achieves error-free tracking of the margin reference value through an integral controller; the output of the integral controller is superimposed with the initial value of the damping control quantity to obtain the current damping control quantity of the unit.

[0076] The aforementioned station correction control calculates a set of deviations by subtracting the margin assessment values ​​of each unit from the station margin assessment values. This deviation vector is then superimposed on the station margin setpoint to obtain the margin reference values ​​for each unit. This can compensate for control deviations caused by the substitution of local information in autonomous control.

[0077] Preferably, the station margin correction control and the unit autonomous margin control are cascaded, with the following input-output relationship:

[0078] Based on the margin assessment value σ of each unit 机,1～n Station margin assessment value σ 场 and station margin setting value The station correction control outputs margin reference values ​​for each unit. Each unit's self-margin σ is based on the assessment. 机,1～n and the margin reference values ​​issued by the station The output damping control quantity D is controlled by autonomous margin control. 机,1～n .

[0079] Preferably, in step S031, the unit implements autonomous control, specifically as follows:

[0080] The key controller is an integrator in the continuous domain, with a control gain of K. dr,i The input is the deviation Δσ between the unit margin reference value and the assessed value. 机,i The damping control change ΔD is generated by the integrator; the change is then superimposed with the initial damping value D0 to obtain the damping control quantity D that the unit needs to match. 机,i The damping control quantity is limited according to the set upper and lower damping limits.

[0081] The time-domain expression for the unit's autonomous control is:

[0082]

[0083] Preferably, step 032, station margin correction control, specifically includes:

[0084] Calculate the deviation Δσ between the site margin assessment value and the individual unit margin assessment values. ε,i This refers to the error in assessing the unit margin caused by substituting local information; this deviation is then added to the set station margin reference value. Obtain the margin reference values ​​for each unit. In the autonomous margin control of the units, margin assessment deviations are compensated, and the overall field margin is made to follow the set reference value through the autonomous control of the margin of each unit.

[0085] Preferably, it also includes: step 040, updating the margin control parameters online;

[0086] Based on the modeling obtained in step S010, step 040 establishes an incremental relationship between the damping control quantity and the unit's autonomous assessment margin based on time-frequency sensitivity; calculates the real-axis component of the damping control quantity sensitivity and obtains the integrator gain in the continuous domain by inversion; discretizes the control period to obtain the integrator gain in the discrete domain; and updates the integrator parameters in the autonomous margin control in step 020 online.

[0087] Preferably, step 040, the margin control parameters are updated online, which is the dynamic adjustment of the margin control parameters, including two parts: the parameter sensitivity calculation of the damping control quantity D and the discretization of the integrator;

[0088] The parameter sensitivity calculation of the damping control quantity D includes:

[0089] The parameter sensitivity of the damping control quantity D relative to the unit margin σ is calculated based on the time-frequency modal factor, specifically as follows:

[0090]

[0091] In the formula, σ represents the unit margin, which is the real part of the oscillation mode λ; ξ(λ,Λ) represents the conversion factor from frequency domain characteristic sensitivity to time domain characteristic sensitivity; Λ represents the weakest frequency domain eigenvalue Λ=0, obtained by logging the numerical matrix. Obtained by feature decomposition;

[0092] Based on the derivative chain rule, all multi-machine frequency domain matrix elements L containing damped control quantity D are... ij Perform partial derivative calculations and sum them, then multiply by the time-frequency conversion factor ξ(λ,Λ) to obtain the parameter sensitivity of the damping control quantity relative to the unit margin.

[0093] Based on this sensitivity result, the incremental relationship is determined as follows:

[0094]

[0095] Discretization of the integrator includes:

[0096] Discretizing the integral controller of the time-domain expression for the unit's autonomous control, we obtain:

[0097]

[0098] In the formula, T step Represents the control cycle;

[0099] By analogy with the incremental relationship and the structure of the discretized form of the integral controller, the gain of the integral controller is obtained:

[0100]

[0101] Based on the online assessment results of the unit's margin, the integrator gain is adjusted in real time, and the integrator in the unit's autonomous margin control is dynamically adjusted.

[0102] According to a third aspect of the present invention, a terminal is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, can be used to perform the method described herein.

[0103] According to a fourth aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, can be used to perform the method described thereon.

[0104] Compared with the prior art, the embodiments of the present invention have at least one of the following beneficial effects:

[0105] The online stability margin assessment method for multiple generator units in the embodiments of this invention takes into account the system's wideband dynamics and the electromagnetic transient characteristics of the power grid because its multi-generator frequency domain model considers the multi-bandwidth control dynamics of the generator units and the electromagnetic transient characteristics of the power grid.

[0106] The online stability margin assessment method for multiple generators in new energy power plants in this embodiment of the invention adopts a hierarchical architecture of "generator autonomy - power plant correction" for the margin. The former is mainly based on the local measurement information of the generator and does not rely on the power plant communication to ensure the speed of margin response. The latter is mainly based on the global information of the power plant and can compensate for the error-free control brought by autonomous control and improve the accuracy of margin response. Therefore, this method can take into account both the accuracy and speed of stability margin control. Attached Figure Description

[0107] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0108] Figure 1 This is a schematic diagram of the framework structure of the online evaluation and control design method for the stability margin of multi-unit grid connection of new energy power stations in an embodiment of the present invention;

[0109] Figure 2 This is a schematic diagram of the framework structure of the online stability margin evaluation method in one embodiment of the present invention;

[0110] Figure 3 This is a flowchart of the pattern numerical search in a preferred embodiment of the present invention;

[0111] Figure 4 This is a schematic diagram illustrating the cascade relationship between station margin correction control and unit autonomous margin control in a preferred embodiment of the present invention.

[0112] Figure 5This is a schematic diagram of the unit autonomous control structure in a preferred embodiment of the present invention;

[0113] Figure 6 This is a schematic diagram of the station margin correction control structure in a preferred embodiment of the present invention;

[0114] Figure 7 The test system topology and main electrical parameters are shown in a specific embodiment of the present invention;

[0115] Figure 8 This is a schematic diagram of the pattern numerical search effect in a specific embodiment of the present invention;

[0116] Figure 9 This serves as a verification of the unit and station margin assessment effect in a specific embodiment of the present invention;

[0117] Figure 10 This is a verification of the unit's autonomous margin control effect in a specific embodiment of the present invention;

[0118] Figure 11 This is to verify the effect of the station margin correction control in a specific embodiment of the present invention. Detailed Implementation

[0119] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention. These all fall within the scope of protection of the present invention.

[0120] like Figure 1 As shown in the figure, one embodiment of the present invention provides an online evaluation method for the stability margin of multiple units in a new energy power station, the main steps of which are as follows:

[0121] Step 010: Establish the multi-unit frequency domain model of the power station: Based on the deviation input and output definitions of each controller in the unit, establish the state space model of the unit, and transform it to the frequency domain to obtain the extended unit model of the control port; Based on the branch-node correlation matrix, establish the node impedance matrix of the power station electrical network; Based on the circuit law, connect the unit model and the electrical network model to obtain the power station multi-unit closed-loop model that explicitly retains the control loop structure.

[0122] Step 020, online stability margin assessment: This includes two parts: model numerical search and margin assessment based on numerical search. The model numerical search is used to locate the weakly damped mode by traversing the real and imaginary parts of the Laplace operator in the multi-machine frequency domain model of Step 010. The margin assessment based on numerical search is used to realize the online assessment of the unit and station margins by substituting local / global time-varying parameters respectively.

[0123] The multi-unit frequency domain model for power stations of this invention needs to be able to display the structure and parameters of all control loops in each unit, while also considering the electromagnetic transient broadband characteristics of the transmission network. Therefore, in a preferred embodiment of this invention, the streamlined construction steps are as follows:

[0124] S1.1 Unit Control Port Extended Frequency Domain Model

[0125] First, using the deviation input of each controller (the difference between the control reference and the control feedback), the voltage drop of the AC filter circuit, and the current of the DC filter circuit as the state space input quantity Δe, and the controller output, the current of the AC filter circuit, and the voltage drop of the DC filter circuit as the state space output quantity Δγ, a small-signal state-space model is established as shown below:

[0126]

[0127] Further transforming the above equation to the frequency domain yields the input-output transfer function G(s) between Δγ and Δx, defined as the forward matrix. This matrix is ​​in block diagonal form, with diagonal elements primarily consisting of the transfer functions K(s) of each controller and the admittance of the AC filter. DC filter impedance composition.

[0128]

[0129] The algebraic equations in the small-signal model of the combined unit and the definition of the state control input Δe mentioned above:

[0130]

[0131] In the formula, Δy represents intermediate variables (such as modulation ratio, power, etc.), and Δu represents AC side voltage. Eliminating the intermediate variables, the input-output relationship between Δe and Δγ, Δu can be obtained, which is called the self-feedback matrix F of the unit.

[0132]

[0133] At this point, the construction of the extended frequency domain model of the control port is complete (the extended model of the unit control port includes two parts: a forward matrix and a self-feedback matrix. The forward matrix retains the input and output ports of all controllers (such as the synchronization loop, the inner current loop, the outer voltage loop, etc.) and includes controller parameters, etc.; the self-feedback matrix includes operating point parameters, etc.).

[0134] S1.2 Frequency domain model of the station's electrical network's nodal impedance

[0135] Based on the connection relationship between nodes and branches in the station, and using the branch-node correlation matrix A and the branch admittance matrix Y... br Construct the node admittance matrix Ynet of the power station network:

[0136] Y net =AY br A T (7)

[0137] Y net This is a frequency domain model of the node impedance of the power grid at the substation, including parameters such as the line impedance of the transmission network, the transformer impedance, and the impedance within the power grid.

[0138] The nodal impedance matrix Z is obtained by inversion. net This reflects the relationship between AC voltage and current at the connection point between the unit and the power network, and can be further expressed as a function of the state output variable Δγ through the transformation matrix R.

[0139]

[0140] S1.3 Station Multi-Machine Closed-Loop Frequency Domain Model

[0141] Arrange the forward matrices of each unit diagonally in blocks to obtain the forward matrix G(s) of the multi-unit system at the station. This matrix retains all control loops within each unit, which facilitates subsequent damping design and sensitivity calculation.

[0142]

[0143] Arrange the feedback matrices of each unit diagonally in blocks to obtain the self-feedback matrix F of the multi-unit system at the station.

[0144]

[0145] Based on the input-output definition of the self-feedback matrix, it can be divided into the following forms:

[0146]

[0147] Substituting equation (8) into equation (1), we obtain the feedback matrix H(s) of the multi-machine system at the station:

[0148]

[0149] Based on equations (9) and (12), a station multi-machine closed-loop frequency domain model with explicit representation of the control loop can be established:

[0150] L(s)=I+G(s)H(s) (13)

[0151] In the formula, the diagonal elements of the forward matrix G(s) retain the control loop structure and parameters of each unit.

[0152] Based on the constructed multi-machine closed-loop frequency domain model of the site, an online stability margin assessment is performed. To obtain a more accurate assessment, in a preferred embodiment of the present invention, as follows... Figure 2 As shown, the evaluation method is divided into the following two steps:

[0153] Step 021: Propose a numerical search method for the mode. Based on the initial search range and step size of the real and imaginary parts, generate frequency domain grid data and substitute it into the multi-machine frequency domain model; calculate the inverse of the determinant of the frequency domain model at each grid point to obtain the grid coordinates corresponding to the peak points; determine a new search range and step size centered on each peak point to generate more accurate frequency domain grid data; repeat the above peak location and grid generation steps until the grid coordinates at the peak points meet the accuracy requirements, thus obtaining the oscillation mode of the system.

[0154] Step 022: Propose an online assessment method for unit and station margins. Considering that the time-varying parameters (such as operating point, damping control variables, etc.) of the unit are consistent with those of other units, these parameters are substituted into a multi-unit frequency domain model, and the unit margin assessment value is calculated through pattern search. Considering the state differences between units, the time-varying parameters of each unit are obtained based on communication and substituted into the multi-unit frequency domain model. The online assessment value of the station margin is calculated through pattern search.

[0155] Furthermore, such as Figure 3 As shown, the input to the model numerical search is a multi-machine frequency domain model and search conditions, and the output is the system oscillation mode. It mainly includes steps such as frequency domain grid generation, determinant calculation, and peak value determination. The input to the unit and station margin calculation module is time-varying parameter information and the system oscillation mode, and the output is the multi-machine frequency domain model after parameter substitution and the unit / station margin. It mainly includes steps such as parameter substitution and mode selection.

[0156] In some preferred embodiments, a preferred pattern numerical search process is provided, as follows:

[0157] S21.1 Frequency Domain Mesh Generation

[0158] Based on the ranges of the real and imaginary parts and the corresponding intervals, a set of frequency domain grid data is generated, as follows:

[0159]

[0160] In the formula, σ represents the real part and ω represents the imaginary part. σ0 is the starting point for the real part search, Δσ is the search interval, k1 represents the number of search points, and σ0 + k1Δσ is the ending point for the real part search; ω0 is the starting point for the imaginary part search, Δω is the search interval, k2 represents the number of search points, and ω0 + k2Δω is the ending point for the imaginary part search.

[0161] S21.2 Matrix Determinant Calculation

[0162] Substituting the frequency domain grid data generated above into the multi-machine frequency domain model and calculating the inverse of the determinant of the matrix, we can obtain the distribution of the inverse of the determinant under this set of grid data.

[0163]

[0164] In the formula, It is an input multi-machine frequency domain model and it incorporates grid data s test and system parameters It belongs to the category of numerical matrices. Determinant calculations on this matrix are fast, and the reciprocal values ​​of the determinant η for each grid coordinate can be obtained in a short time.

[0165] S21.3 Peak Point Location

[0166] Based on the determinant result η obtained above, peak points are filtered by combining absolute value and gradient. Specifically, peak points are determined by the following conditions, corresponding to the grid data s. ij This is a potential oscillation mode.

[0167]

[0168] In the formula, the subscripts i and j represent the real and imaginary part numbers of the grid data, respectively, and ξ abs ξ gra These are the set absolute value threshold and gradient threshold, respectively. After iteration, the peak point grid coordinates within the initial search region are obtained, which represent the oscillation mode of the system.

[0169] λ={s ij} (17)

[0170] In other preferred embodiments, a preferred online assessment process for unit and station margins is provided, specifically as follows:

[0171] S22.1 Unit Margin Online Assessment

[0172] The core step of online assessment of unit margin is to assume that the state of other units is the same as that of this unit, meaning that only the unit's own time-varying parameters and damping control information (information obtained locally online) are needed. Taking the i-th unit as an example, the parameters are substituted as follows:

[0173]

[0174] In the formula, P i Q i V i D i These represent the active power, reactive power, grid-connected voltage, and damping control parameters of the i-th generating unit, respectively; the states of other generating units are assumed to be the same. Substitute these parameters into the multi-unit frequency domain model. The data is then transmitted to the mode numerical search module, which selects the mode closest to the imaginary part from the feedback oscillation modes and extracts its real part as the online evaluation value of the unit's margin. Taking the i-th unit as an example, its margin is...

[0175] Unit margin: σ 机,i =Re[λ j (19)

[0176] In the formula, λ j This is the oscillation mode closest to the imaginary axis.

[0177] S22.2 Station Margin Online Assessment

[0178] The core steps of online assessment of power station margin are to input the time-varying parameters and damping control information (global information) of each unit. The parameter input information is as follows:

[0179]

[0180] Substitute the above parameters into the multi-machine frequency domain model The data is then transmitted to the mode numerical search module, which filters out the oscillation mode closest to the imaginary part and extracts its real part as an online assessment value of the station margin, such as:

[0181] Station margin: σ 场 =Re[λ j ] (twenty one)

[0182] Based on the same inventive concept, other embodiments of the present invention also provide a multi-unit stability margin control design method for new energy power plants, including step 030, margin feedback control, such as... Figure 3 As shown, step 030 and step 020 of the online evaluation method for the stability margin of multiple units in new energy power plants constitute a closed-loop control structure; step 020 provides margin control object information for step 030, and step 030 provides real-time damping control quantity information for step 020.

[0183] Step 030, margin feedback control, includes:

[0184] Step S031, Unit autonomous margin control, that is, to achieve rapid response control of the margin based on the local assessment value of the unit margin;

[0185] Step S032, station margin correction control, that is, generating reference values ​​for the margin control of each unit based on the global assessment value of the station margin, and compensating for the control deviation between the unit's autonomous margin control and the actual station margin.

[0186] The margin feedback control design method proposed in the above embodiments has a hierarchical architecture of "unit autonomy" and "station correction", which can ensure the speed and accuracy of margin control and improve the margin disturbance resistance and stable operation capability of new energy power plants under time-varying operating conditions.

[0187] The unit autonomous control in the above embodiments mainly uses the deviation between the unit margin assessment value and the margin reference value (issued by the station correction control) as input. An integral controller achieves error-free tracking of the margin reference value. The output of the integral controller is superimposed with the initial value of the damping control quantity to obtain the current damping control quantity of the unit. The station correction control mainly calculates a deviation by subtracting the margin assessment value of each unit from the station margin assessment value. This deviation vector is superimposed with the station margin setpoint to obtain the margin reference value of each unit, which can compensate for control deviations caused by substituting local information during autonomous control.

[0188] Furthermore, in a preferred implementation, multi-unit margin online feedback control consists of cascaded station margin correction control and unit autonomous margin control, with the input-output relationship as follows: Figure 4 As shown. Based on the margin assessment value σ of each unit. 机,1～n Station margin assessment value σ 场 and station margin setting value The station correction control outputs margin reference values ​​for each unit. Each unit's self-margin v based on the assessment 机,1～n and the margin reference values ​​issued by the station The output damping control quantity D is controlled by autonomous margin control. 机,1～n .

[0189] The unit is under autonomous control, and its control structure Figure 5 As shown (taking the i-th unit as an example). The key controller is an integrator in the continuous domain, with a control gain of K. dr,i That is, the integrator gain as described in claim 1; the input is the deviation Δσ between the unit margin reference value and the evaluated value. 机,i The damping control variation ΔD can be generated through the integrator; the variation is then superimposed with the initial damping value D0 to obtain the damping control quantity D that the unit currently needs to match. 机,i The damping control quantity is limited according to the set upper and lower damping limits. The time-domain expression of this control law is shown below:

[0190]

[0191] The station margin correction control law has the following control structure: Figure 6 As shown. Calculate the deviation Δσ between the estimated site margin and the estimated margin of each generating unit. ε,i This refers to the error in assessing the unit margin caused by substituting local information. This deviation is then added to the set station margin reference value. The margin reference values ​​for each unit can be obtained. This compensates for margin assessment deviations in the autonomous margin control of the units, thereby enabling the overall field margin to follow the set reference value through the autonomous control of the margins of each unit.

[0192] To further ensure the accuracy of online margin assessment, another preferred embodiment of the present invention adds step 040, which involves online updating of margin control parameters. Specifically, step 040, based on the modeling obtained in step S010, establishes an incremental relationship between the damping control quantity and the unit's autonomous assessment margin based on time-frequency sensitivity; calculates the real-axis component of the damping control quantity sensitivity, and obtains the integrator gain in the continuous domain by inversion; discretizes the control cycle to obtain the integrator gain in the discrete domain; and updates the integrator parameters in the autonomous margin control in step 020 online.

[0193] The dynamic adjustment of the margin control parameters mainly includes two parts: the calculation of the parameter sensitivity of the damping control quantity D and the discretization of the integrator. First, the parameter sensitivity of the damping control quantity D relative to the unit margin σ is calculated based on the time-frequency mode factor, as follows:

[0194]

[0195] In the formula, σ represents the unit margin, which is the real part of the oscillation mode λ. ξ(λ,Λ) represents the conversion factor from frequency domain characteristic sensitivity to time domain characteristic sensitivity; Λ represents the weakest frequency domain eigenvalue (Λ=0), which can be obtained by logging the numerical matrix. Eigenvalue decomposition is performed. Based on the derivative chain rule, all elements L of the multi-machine frequency domain matrix containing damped control quantity D are obtained. ij By performing partial derivative calculations and summing the results, and then multiplying by the time-frequency conversion factor ξ(λ,Λ), the parameter sensitivity of the damping control quantity relative to the unit margin can be obtained. Based on this sensitivity result, we have:

[0196]

[0197] Further discretizing the integral controller in equation (22), we can obtain (taking the forward Euler as an example):

[0198]

[0199] In the formula, T step This represents the control cycle. Analogous to the structures of equations (25) and (24), the integral controller gain can be obtained:

[0200]

[0201] Based on the online evaluation results of the unit margin, the integrator gain shown in equation (26) is calculated in real time, and the integrator in the autonomous margin control of the unit is dynamically adjusted.

[0202] The above embodiments mainly include multi-unit modeling of the power plant, online margin assessment, margin feedback control law, and parameter updating. Specifically, multi-unit modeling explicitly retains the control loop structure, facilitating control design; online margin assessment includes model numerical search and unit / power plant margin assessment; the margin feedback control law includes a "unit autonomous control law" and a "power plant correction control law," the former using an integrator to achieve error-free control of the unit's autonomous margin assessment, and the latter used to compensate for control deviations between the unit's autonomous margin control and the power plant's actual margin; parameter updating dynamically adjusts the integrator gain based on time-frequency sensitivity. The margin feedback control method proposed in this invention has a layered architecture of "unit autonomy" and "power plant correction," ensuring the speed and accuracy of margin control and improving the power plant's margin immunity and stable operation capability under time-varying conditions.

[0203] To verify the online evaluation and control design method for the stability margin of multi-unit grid connection in new energy power plants as described above, a specific embodiment of the present invention is used to verify feasibility and effectiveness.

[0204] Select Figure 7 The test system shown employs a multi-unit aggregation model of a wind farm, considering four feeders, each with an aggregated unit capacity of 24MW, under different wind conditions (4.5m / s, 5.5m / s, 6.5m / s, and 7.2m / s). The site is equipped with 12.5MW of centralized energy storage (approximately 13% of the total capacity). The grid uses a Thevenin equivalent model, adjusting the short-circuit ratio by modifying the internal impedance. The wind turbines are direct-drive type, employing a grid-following control mode (maximum power point tracking mode for the turbine-side converter and constant DC voltage mode for the grid-side converter), considering control loops such as the inner current loop, constant DC voltage / reactive power outer loop, torque loop, phase-locked loop, and voltage feedforward. The grid-connected energy storage uses a power synchronization grid-connected control mode, considering control loops such as the inner current loop, virtual admittance loop, reactive power-voltage droop outer loop, and virtual synchronization loop. For the above system, based on online margin assessment, online margin control, and dynamic adjustment of integrator gain, four unit autonomous margin control modules and one station margin correction control module are constructed, thereby forming the grid-connected stability margin online control system of the target system.

[0205] Pattern numerical search results are as follows Figure 8 As shown. Within the selected search range, such as a real part range of [-30, 0] and an imaginary part range of [50 rad / s, 150 rad / s] and [500 rad / s, 650 rad / s], by calculating the reciprocal of the determinant of the aforementioned multi-machine frequency domain model, two relatively obvious resonance peaks can be observed. By continuously narrowing the grid range and improving the grid search accuracy, a more accurate oscillation mode can be obtained, thus verifying the effectiveness of the mode numerical search.

[0206] The effectiveness of the unit and station margin assessment and simulation verification are as follows: Figure 9 As shown, from top to bottom, these represent the short-circuit ratio, stability margin, and synchronization frequency. It can be seen that when margin control is not enabled, the unit margin assessment values ​​and the station margin assessment values ​​respond naturally. As the short-circuit ratio gradually decreases from 4 to 2, the margins of each unit and the overall station also show a decreasing trend. Around 23 seconds, a margin assessment value greater than 0 indicates system instability, and the synchronization frequency exhibits divergent instability, thus verifying the effectiveness of the margin assessment.

[0207] The unit's autonomous margin control effect is as follows Figure 10 As shown, from top to bottom, the figures represent the short-circuit ratio, stability margin, and damping control values ​​for each unit. It can be seen that after enabling autonomous margin control, even with changes in the short-circuit ratio, the margins of each unit can follow the set unit margin reference values ​​(here, -5 and -4). However, a magnified view reveals that although the margins of each unit can track the reference values ​​well, there is a certain difference between the actual margin and the unit margin reference values ​​for the entire operation.

[0208] The effect of station margin correction control is as follows Figure 11 As shown, from top to bottom, these represent the short-circuit ratio, stability margin, and damping control values ​​for each unit. It can be seen that after enabling station correction control, deviations occur between the unit's margin and the station reference value. This is due to the information gap between the unit's self-assessment and the overall station assessment. Station control correction ensures that the actual margin in the station assessment follows the station margin setpoint.

[0209] Based on the same inventive concept, in other embodiments of the present invention, a terminal is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it can be used to perform the above-described method or to run the above-described system.

[0210] Optionally, the memory is used to store programs; the memory may include volatile memory, such as random-access memory (RAM), such as static random-access memory (SRAM), double data rate synchronous dynamic random-access memory (DDR SDRAM), etc.; the memory may also include non-volatile memory, such as flash memory. The memory is used to store computer programs (such as application programs, functional modules, etc. that implement the above methods), computer instructions, etc., and the above-mentioned computer programs, computer instructions, etc., may be partitioned and stored in one or more memories.

[0211] A processor is used to execute a computer program stored in memory to implement the various steps of the methods involved in the above embodiments. For details, please refer to the relevant descriptions in the preceding method embodiments.

[0212] The processor and memory can be separate structures or integrated structures. When the processor and memory are separate structures, they can be coupled together via a bus.

[0213] Based on the same inventive concept, in other embodiments of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, can be used to perform the above-described method or to run the above-described system.

[0214] Computer-readable media include computer storage media and communication media, wherein communication media include any medium that facilitates the transfer of computer programs from one place to another. Storage media can be any available medium accessible to a general-purpose or special-purpose computer. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. Alternatively, the ASIC can reside in a user device. Of course, the processor and storage medium can also exist as discrete components in a communication device.

[0215] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0216] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0217] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0218] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0219] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for online evaluation of the stability margin of multiple units in a new energy power station, characterized in that, include: Step 010: Establish the multi-unit frequency domain model of the power station: Based on the deviation input and output definitions of each controller in the unit, establish the state space model of the unit, and transform it to the frequency domain to obtain the extended unit model of the control port; Based on the branch-node correlation matrix, establish the node impedance matrix of the power station electrical network; Based on the circuit law, connect the unit model and the electrical network model to obtain the power station multi-unit closed-loop model that explicitly retains the control loop structure. Step 020, online stability margin assessment: This includes two parts: model numerical search and margin assessment based on numerical search. The model numerical search is used to locate the weakly damped mode by traversing the real and imaginary parts of the Laplace operator in the multi-machine frequency domain model of Step 010. The margin assessment based on numerical search is used to realize the online assessment of the unit and station margins by substituting local / global time-varying parameters respectively.

2. The online evaluation method for multi-unit stability margin of a new energy power station according to claim 1, characterized in that, The multi-unit frequency domain model of the station in step 010 is a model that takes into account the electromagnetic transient broadband characteristics of the transmission network and can display the structure and parameters of all control loops in each unit. The steps for constructing the multi-unit frequency domain model of the station include: S1.1 Constructing an extended frequency domain model of the unit's control ports, specifically: A small-signal state-space model is established, with the controller deviation input, AC filter circuit voltage drop, and DC filter circuit current as the state-space input quantities Δe, and the controller output, AC filter circuit current, and DC filter circuit voltage drop as the state-space output quantities Δγ. Δγ=CΔx+DΔe Transforming the small-signal state-space model to the frequency domain yields the input-output transfer function G(s) between Δγ and Δx, defined as a forward matrix. This forward matrix is ​​presented in block diagonal form, with diagonal elements including the transfer functions K(s) of each controller and the admittance of the AC filter. DC filter impedance The algebraic equations in the small-signal state-space model of the combined unit and the definition of the state control input Δe are as follows: Deviation definition: Δe=A1Δy+B1Δγ Algebraic equation: Δy=A²Δγ+B²Δu In the formula, Δy represents an intermediate variable and Δu represents the AC side voltage; by eliminating the intermediate variable, the input-output relationship between Δe and Δγ and Δu is obtained, which is called the self-feedback matrix F of the unit. S1.2 Establish the frequency domain model of the nodal impedance of the power station network, specifically as follows: Based on the connection relationship between nodes and branches in the station, and using the branch-node correlation matrix A and the branch admittance matrix Y... br Construct the node admittance matrix Y of the power station network. net : AND net =AY br TO T The nodal impedance matrix Z is obtained by inversion. net It is used to reflect the AC voltage and current relationship at the connection point between the unit and the power network, and is expressed as a function of the state output variable Δγ through the transformation matrix R: S1.3 Establish a multi-machine closed-loop frequency domain model for the station, specifically as follows: The forward matrices of each unit are arranged diagonally in blocks to obtain the forward matrix G(s) of the multi-unit system of the station. This matrix retains all control loops within each unit. Arrange the feedback matrices of each unit diagonally in blocks to obtain the self-feedback matrix F of the multi-unit system at the station. According to the input-output definition of the self-feedback matrix F, it is divided into: Substituting the function of the state output variable Δγ into the partition form of the self-feedback matrix F, we obtain the feedback matrix H(s) of the multi-machine system at the station: Based on the forward matrix G(s) and feedback matrix H(s) of the multi-machine system at the station, a closed-loop frequency domain model of the multi-machine system at the station with explicit representation of the control loop is established: L(s) = I + G(s)H(s) In the formula, the diagonal elements of the forward matrix G(s) retain the control loop structure and parameters of each unit.

3. The online evaluation method for the stability margin of multiple units in a new energy power station according to claim 1, characterized in that, Step 020 includes: Step 021, perform mode numerical search: Based on the initial search range and step size of the real and imaginary parts, generate frequency domain grid data and substitute it into the multi-machine frequency domain model; calculate the inverse of the determinant of the frequency domain model at each grid point to obtain the grid coordinates corresponding to the peak points; determine a new search range and step size with each peak point as the center to generate new frequency domain grid data; repeat the above peak location and grid generation steps until the grid coordinates at the peak points meet the accuracy requirements to obtain the oscillation mode of the system; Step 022: Conduct online assessment of unit and station margins: Considering that the time-varying parameters of the unit itself are consistent with those of other units, substitute them into the multi-unit frequency domain model and calculate the unit margin assessment value through pattern search; Considering the state differences between units, obtain the time-varying parameters of each unit based on communication and substitute them into the multi-unit frequency domain model, and calculate the online assessment value of the station margin through the pattern numerical search.

4. The online evaluation method for multi-unit stability margin of a new energy power station according to claim 3, characterized in that, In step 021, a pattern numerical search is performed, including: S21.1 Frequency domain grid generation, specifically: Based on the ranges of the real and imaginary parts and the corresponding intervals, a set of frequency domain grid data is generated: In the formula, σ represents the real part and ω represents the imaginary part; where σ0 is the starting point of the real part search, Δσ is the search interval, k1 represents the number of search points, and σ0+k1Δσ is the ending point of the real part search; ω0 is the starting point of the imaginary part search, Δω is the search interval, k2 represents the number of search points, and ω0+k2Δω is the ending point of the imaginary part search. S21.2, Matrix determinant calculation, specifically: Substituting the generated frequency domain grid data into the multi-machine frequency domain model, the inverse determinant of the matrix is ​​calculated to obtain the distribution of the inverse determinant under this set of frequency domain grid data: In the formula, It is an input multi-machine frequency domain model and it incorporates grid data s test and system parameters It belongs to the category of numerical matrices; S21.3, Peak point location, specifically: Based on the obtained determinant result η, peak points are filtered by combining absolute value and gradient. Specifically, peak points are determined by the following conditions, corresponding to the grid data s. ij Potential oscillation mode: In the formula, the subscripts i and j represent the real and imaginary part numbers of the grid data, respectively, and ξ abs ξ gra These are the set absolute value threshold and gradient threshold, respectively; after iteration, the peak point grid coordinates within the initial search region are obtained, which represent the oscillation mode of the system. λ={s ij }。 5. The online evaluation method for the stability margin of multiple units in a new energy power station according to claim 3, characterized in that, Step 022, conduct online assessment of unit and station margins, including: S22.1, Online assessment of unit margin, specifically: Assuming that the status of other units is the same as that of this unit, only the time-varying parameters and damping control information of the unit itself are needed; The parameters of the i-th generating unit are: In the formula, P i Q i V i D i These are the active power, reactive power, grid voltage, and damping control parameters of the i-th generating unit, respectively; the default states of other generating units are the same. Substitute the parameters into the multi-machine frequency domain model A numerical search for the model is performed, and the model closest to the imaginary part is selected from the feedback oscillation modes. The real part is then extracted as the online evaluation value of the unit margin; that is, for the i-th unit, the margin is... Unit margin: σ 机,i =Re[λ j ] In the formula, λ j This is the oscillation mode closest to the imaginary axis; S22.2, the online assessment method for station margin, specifically: Substituting the time-varying parameters and damping control information of each unit, the parameter input information is as follows: Substitute the parameters into the multi-machine frequency domain model Furthermore, a numerical search for the mode is performed to filter out the oscillation mode closest to the imaginary part, and the real part is extracted as the online evaluation value of the station margin, such as: Station margin: σ 场 =Re[λ j ].

6. A multi-unit stability margin control design method for new energy power plants, characterized in that, Including step 030, margin feedback control law; Step 030 and step 020 of the online evaluation method for multi-unit stability margin of new energy power stations according to any one of claims 1-5 constitute a closed-loop control structure; step 020 provides margin controlled object information for step 030, and step 030 provides damping control quantity information for step 02; Step 030, margin feedback control, includes: Step S031, Unit autonomous margin control, that is, to achieve rapid response control of the margin based on the local assessment value of the unit margin; Step S032, station margin correction control, that is, generating reference values ​​for the margin control of each unit based on the global assessment value of the station margin, and compensating for the control deviation between the unit's autonomous margin control and the actual station margin.

7. The multi-unit stability margin control design method for new energy power plants according to claim 6, characterized in that, The unit's autonomous control uses the deviation between the unit's margin assessment value and the margin reference value as input, and achieves error-free tracking of the margin reference value through an integral controller; the output of the integral controller is superimposed with the initial value of the damping control quantity to obtain the unit's current damping control quantity; The aforementioned station correction control calculates a set of deviations by subtracting the margin assessment values ​​of each unit from the station margin assessment values. This deviation vector is then superimposed on the station margin setpoint to obtain the margin reference values ​​for each unit. This can compensate for control deviations caused by the substitution of local information in autonomous control.

8. The multi-unit stability margin control design method for a new energy power station according to claim 7, characterized in that: The station margin correction control and the unit autonomous margin control are cascaded, and their input-output relationship is as follows: Based on the margin assessment value σ of each unit 机,1～n Station margin assessment value σ 场 and station margin setting value The station correction control outputs margin reference values ​​for each unit. Each unit's self-margin σ is based on the assessment. 机,1～n and the margin reference values ​​issued by the station The output damping control quantity D is controlled by autonomous margin control. 机,1～n .

9. The multi-unit stability margin control design method for a new energy power station according to claim 8, characterized in that: Step S031, Unit Autonomous Control, specifically: The key controller is an integrator in the continuous domain, with a control gain of K. dr,i The input is the deviation Δσ between the unit margin reference value and the assessed value. 机,i The damping control change ΔD is generated by the integrator; the change is then superimposed with the initial damping value D0 to obtain the damping control quantity D that the unit needs to match. 机,i The damping control quantity is limited according to the set upper and lower damping limits. The time-domain expression for the unit's autonomous control is: 。 10. The multi-unit stability margin control design method for a new energy power station according to claim 8, characterized in that: Step 032, station margin correction control, specifically: Calculate the deviation Δσ between the site margin assessment value and the individual unit margin assessment values. ε,i This refers to the error in assessing the unit margin caused by substituting local information; this deviation is then added to the set station margin reference value. Obtain the margin reference values ​​for each unit. In the autonomous margin control of the units, margin assessment deviations are compensated, and the overall field margin is made to follow the set reference value through the autonomous control of the margin of each unit.

11. The multi-unit stability margin control design method for a new energy power station according to claim 6, characterized in that, It also includes: Step 040, online update of margin control parameters; Based on the modeling obtained in step S010, step 040 establishes an incremental relationship between the damping control quantity and the unit's autonomous assessment margin based on time-frequency sensitivity; calculates the real-axis component of the damping control quantity sensitivity and obtains the integrator gain in the continuous domain by inversion; discretizes the control period to obtain the integrator gain in the discrete domain; and updates the integrator parameters in the autonomous margin control in step 020 online.

12. The multi-unit stability margin control design method for a new energy power station according to claim 11, characterized in that, Step 040, online update of margin control parameters, which is the dynamic adjustment of margin control parameters, including two parts: parameter sensitivity calculation of damping control quantity D and integrator discretization; The parameter sensitivity calculation of the damping control quantity D includes: The parameter sensitivity of the damping control quantity D relative to the unit margin σ is calculated based on the time-frequency modal factor, specifically as follows: In the formula, σ represents the unit margin, which is the real part of the oscillation mode λ; ξ(λ,Λ) represents the conversion factor from frequency domain characteristic sensitivity to time domain characteristic sensitivity; Λ represents the weakest frequency domain eigenvalue Λ=0, obtained by logging the numerical matrix. Obtained by feature decomposition; Based on the derivative chain rule, all multi-machine frequency domain matrix elements L containing damped control quantity D are... ij Perform partial derivative calculations and sum them, then multiply by the time-frequency conversion factor ξ(λ,Λ) to obtain the parameter sensitivity of the damping control quantity relative to the unit margin. Based on this sensitivity result, the incremental relationship is determined as follows: Discretization of the integrator includes: Discretizing the integral controller of the time-domain expression for the unit's autonomous control, we obtain: In the formula, T step Represents the control cycle; By analogy with the incremental relationship and the structure of the discretized form of the integral controller, the gain of the integral controller is obtained: Based on the online assessment results of the unit's margin, the integrator gain is adjusted in real time, and the integrator in the unit's autonomous margin control is dynamically adjusted.

13. A terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it can be used to perform the method of any one of claims 1-12.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, this program can be used to perform the method of any one of claims 1-12.

Images