A AC-DC power flow calculation method based on learnable switched control mode diagrams

Abstract

The present invention relates to the technical field of AC-DC power transmission, and specifically discloses an AC-DC power flow calculation method based on learnable switched control mode diagrams, including the following steps: Step S1: Construct a power flow model of an AC-DC hybrid system; Step S2: Construct a graph neural network model of an AC-DC power grid; embed AC / DC operation physical information into the GNN learning process; Step S3: Use the violation of the firing / extinction angle constraint as the minimization objective, and construct a physics-guided graph neural network PG-GNN that supports control mode switching based on the power flow model of the AC-DC hybrid system. Use multiple PG-GNNs to master DC control mode switching, experience each control mode to train its specific PG-GNN, and then the PG-GNN that meets the firing / extinction angle constraint makes a final decision. The advantage of the present invention is to propose a gradient transformation algorithm based on the augmented Lagrangian method (ALM), which enables the PG-GNN to strictly follow Kirchhoff's laws and release converter constraint violations as the only signal to trigger control mode switching.

Description

Technical Field

[0001] The present invention relates to the technical field of AC-DC power transmission, and particularly to an AC-DC power flow calculation method based on learnable switched control modes. Background Art

[0002] AC / DC hybrid systems have widely promoted the consumption of renewable energy. Due to the significant uncertainties brought by renewable energy, large-scale interconnected power grids, and the non-linearity of DC lines and converters, faster AC / DC power flow analysis is more necessary than ever. There are rich model-based methods for AC / DC power flow, ranging from basic unified and sequential methods to enhanced parallel methods. However, in large AC-DC hybrid systems, slow convergence speed and diverse control modes still impose a heavy computational burden. Emerging data-driven methods provide faster solutions, but their generalization may be limited by the lack of physical embedding and topological patterns. Physically-guided graph neural networks (PG-GNNs) may be a promising solution to the above problems. For pure AC power flow, two alternative solutions have emerged. One is to simulate the system process of a power flow solver, such as the Newton-Raphson (NR) solver, etc., and the other is to directly regard the power flow equation as a loss function. However, they cannot be directly used for AC / DC power flow analysis because there is no effective method to guide DC control mode switching. Additionally, current GNN architectures are difficult to determine whether the non-convergence of AC-DC power flow stems from inappropriate control modes or violations of Kirchhoff's laws. Summary of the Invention

[0003] The purpose of the present invention is to overcome the drawbacks of the prior art and provide an AC-DC power flow calculation method based on learnable switched control modes.

[0004] The purpose of the present invention is achieved by the following technical solutions: An AC-DC power flow calculation method based on learnable switched control modes, comprising the following steps:

[0005] Step S1: Construct a power flow model of the AC-DC hybrid system;

[0006] Step S2: Construct a graph neural network model of the AC-DC power grid; embed AC / DC operation physical information into the GNN learning process;

[0007] Step S3: Take the violation of the firing / extinction angle constraint as the minimization objective, and construct a physically-guided graph neural network PG-GNN that supports control mode switching based on the power flow model of the AC-DC hybrid system. Master DC control mode switching through multiple physically-guided graph neural networks PG-GNN, experience control mode one and control mode two to train their specific PG-GNNs, and then, the PG-GNN that meets the firing / extinction angle constraint outputs the AC-DC power flow calculation result.

[0008] Specifically, the power flow model of the AC-DC hybrid system in step S1 is as follows:

[0009]

[0010] Among them, and represent the PV bus group, AC bus group, DC bus group, and common coupling point group respectively, is the active power injected into AC bus i, is the reactive power injected into AC bus i, is the active power injected into the common coupling point c, P cj (V c ,V j ,δ ij ) is the sum of the active powers transmitted through the lines connected to c, is the reactive power injected into the common coupling point c, Q cj (V c ,V j ,δ ij ) is the sum of the reactive powers transmitted through the lines connected to c, is the phase angle of the k-th DC line, is the ratio k of the converter transformer on the rectifier side, is the ratio k of the converter transformer on the inverter side; represents the reactance of the converter connected to the inverter / rectifier DC bus k; α k / γ k is the firing / extinguishing angle of the k-th rectifier / inverter; represents the rectification / inversion voltage of the k-th DC line; R dc and I d,k represent the resistance and current of the k-th DC line respectively; j∈i means that the j-th bus is adjacent to the i-th bus through the branch ij; P ij (V i ,V j ,δ ij ) and Q ij (V i ,V j ,δ ij ) are functions for calculating the active and reactive powers of the line ij; δ ij is the voltage phase angle difference between bus i and bus j, V p is the voltage of the PV bus, is the PV bus voltage reference value, is the rectification current of the k-th DC line, is the reference value of the rectification current of the k-th DC line, is the reference value of the inversion voltage of the k-th DC line, is the minimum value of the firing angle of the k-th rectifier / inverter, is the inversion current of the k-th DC line; is the reference value of the inversion current of the k-th DC line.

[0011] Specifically, the method for constructing the graph neural network model of the AC-DC power grid in step S2 is as follows:

[0012] By introducing converters into the nodes and DC transformers and DC lines into the edges, the DC power grid is unified in the common graph of the AC power grid. For an AC-DC system with N buses, b branches, and 1 DC line, its corresponding GNN has N + 2a nodes and b + 3a edges:

[0013]

[0014] In the formula, h(·) (0) is the initial hidden layer of the GNN, and are the power generation power and load power respectively, and V i is the voltage; for the inverter, α min is the minimum value of the firing angle, is the rectification reactance, is the reference value of the rectification current of the k-th DC line, is the reference value of the inversion current of the k-th DC line. For the rectifier, or γ min is the minimum value of the extinction angle, is the inversion reference voltage, represents the set of balanced nodes, and are the active power generation power and active load power respectively, and are the reactive power generation power and reactive load power respectively, represents the DC bus group, is the PQ bus group, is the point of common coupling bus group; the input edge features are composed of branches, expressed as Edge feature matrix represents the feature vector of each edge. For the AC branch, x ε =[G ε ,Bε , where G ε and B ε are the ε-line admittance parameters. For rectifiers and inverters, For the DC branch, or is a trainable transformation matrix that embeds edge features into nodes, G ε and B ε are the ε-line conductance and susceptance parameters respectively, and are the minimum and maximum ratios of the commutation transformer, R dc is the resistance of the DC line.

[0015] Specifically, the feed-forward of the graph neural network model is:

[0016]

[0017] In the formula, σ is the activation function, F represents the order of adjacent nodes, is: a filter in the polynomial form of the graph Laplacian operator, which operates according to the graph structure to operate, H(·) (l) represents the response of the l-th layer with the response of the previous layer as the input, x is the input feature matrix of the GNN, representing the initial input, θ j is the trainable weight of the j-th order filter, is the output of the GNN, equal to the response of the last layer M, including the voltage and phase angle of the AC bus, the converter voltage and trigger angle of the DC bus, or [V, α(or γ)], is the forward propagation function of the entire GNN, where θ is the trainable parameter of all layers, [V, δ] is the output feature of the GNN, including the voltage V and phase angle δ of the AC bus.

[0018] Specifically, in step S3, a physical-guided graph neural network PG-GNN that supports control mode switching is constructed with violating the trigger / extinction angle constraint as the minimization objective; and an equivalent optimization problem of the power flow model is formed:

[0019]

[0020] Among them, the state variables to be solved are parameterized by the GNN φ θ (x), f PQV , and are the equality constraint conditions of formulas (1)-

[0021] (3), (4) and (5) respectively;

[0022] The process of simulating the firing / extinguishing angle of the rectifier / inverter transformer is carried out by the following formula, using f DC to represent uniformly and and omitting the subscripts:

[0023]

[0024] In the formula, is the degree of violation of the firing / extinguishing angle constraint, ReLU is the activation function, α min is the minimum firing angle, α is the current firing angle, K re is the ratio of the converter on the rectifier side, K re,max is the maximum allowable value of the converter ratio, K re , min is the minimum allowable value of the converter ratio.

[0025] Specifically, the following ALM-based learning strategy is used in step S3:

[0026] (i ∈ {PQV, DC})

[0027]

[0028] where i is a variable, PQV is the PQ bus and the V bus, DC is all DC buses, is the total loss function, which combines the constraint violation metric, the linear combination term and the quadratic penalty term, ρ > 0, is the penalty parameter, controlling the penalty strength of the constraint violation; θ is the parameter of the GNN, which is updated by minimizing the total loss function, λ i is the dual variable, used to adjust the weight of the constraint term, φ θ (x) is the forward propagation function of the NN, used to calculate the output of the current state. Equation (11a) updates θ by minimizing Equation (10), fixes θ, and then updates the dual variable through Equation (11b). Equation (11) is continuously executed until Equation (10) remains stable.

[0029] Specifically, the said Equation (11a) is implemented by the following formula:

[0030]

[0031] In the formula, θ is the trainable parameter of the graph neural network GNN, is the loss function, λ is the Lagrange multiplier, φ θ (x) is the output of the GNN.

[0032] The present invention has the following advantages:

[0033] The present invention proposes an AC-DC power flow calculation method based on learnable switched control mode graphs. In the customized graph modeling of AC and DC grids, AC / DC operation physical information is embedded into the GNN learning process. Then, in order to provide a control mode switching signal, the AC / DC power flow model is reconstructed into a parametric optimization problem, where the trigger / extinction angle constraints are relaxed to violation minimization. In addition, the present invention proposes a gradient transformation algorithm based on the augmented Lagrangian method (ALM) to make the PG-GNN strictly follow Kirchhoff's laws and release converter constraint violations as the only signal to trigger control mode switching. Description of the Drawings

[0034] Figure 1 It is a schematic flow chart of the method of the present invention. Detailed Embodiments

[0035] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention, that is, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Usually, the components of the embodiments of the present invention described and illustrated herein can be arranged and designed in various different configurations.

[0036] Therefore, the detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed present invention, but merely represents selected embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0037] It should be noted that relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising one..." does not exclude the presence of additional identical elements in the process, method, article or device including the said element.

[0038] The present invention will be further described below with reference to the accompanying drawings, but the scope of protection of the present invention is not limited to the following. As Figure 1As shown in the figure, a method for calculating AC-DC power flow based on learnable switched control mode diagrams includes the following steps:

[0039] Step S1: Construct a power flow model of the AC-DC hybrid system;

[0040] The power flow model is as follows:

[0041]

[0042] Among them, and represent the PV bus group, AC bus group, DC bus group, and common coupling point group respectively, is the active power injected into AC bus i, is the reactive power injected into AC bus i, is the active power injected into common coupling point c, P cj (V c , V j , δ ij ) is the sum of the active powers transmitted through the lines connected to c, is the reactive power injected into common coupling point c, Q cj (V c , V j , δ ij ) is the sum of the reactive powers transmitted through the lines connected to c, is the phase angle of the k-th DC line, is the ratio k of the converter transformer on the rectifier side, is the ratio k of the converter transformer on the inverter side; represents the reactance of the converter connected to the inverter / rectifier DC bus k; α k / γ k is the firing / extinguishing angle of the k-th rectifier / inverter; represents the rectification / inversion voltage of the k-th DC line; R dc and I d,k represent the resistance and current of the k-th DC line respectively; j ∈ i means that the j-th bus is adjacent to the i-th bus through the branch ij; Pij(V i , V j , δ ij ) and Q ij (V i , V j , δ ij ) are functions for calculating the active and reactive powers of line ij; δ ij is the voltage phase angle difference between bus i and bus j, V p is the voltage of the PV bus, is the PV bus voltage reference value, $I_{d,k}^{r}$ is the rectifier current of the $k$-th DC line, $I_{d,k}^{r,ref}$ is the reference value of the rectifier current of the $k$-th DC line, $V_{i,k}^{r,ref}$ is the reference value of the inverter voltage of the $k$-th DC line, $\alpha_{min,k}$ is the minimum value of the firing angle of the $k$-th rectifier / inverter, $I_{d,k}^{i}$ is the inverter current of the $k$-th DC line; $I_{d,k}^{i,ref}$ is the reference value of the inverter current of the $k$-th DC line. The sign of the third term on the right side in Eqs. (2a) and (2b) is negative if it specifies the rectifier, otherwise it is positive. Eqs. (5a) and (5b) are the control equations of the DC system, and the superscript ref represents the control reference.

[0043] Step S2: Construct a graph neural network model for the AC-DC power grid; embed the AC / DC operation physical information into the GNN learning process; the AC / DC operation physical information includes the AC power balance equation, the DC power balance equation, and the voltage and current relationship equation;

[0044] The power system can be represented as an undirected graph $G=(V, E, A)$, where $V$, $E$, and $A$ are solved on the bus set, branch set, and adjacency matrix respectively. For an AC / DC hybrid system, it is not easy to create a GNN because the necessary characteristics for representing the AC grid, DC grid, and their bridges do not always match each other. Therefore, the present invention proposes an alternative to unify these characteristics, specifically:

[0045] By introducing converters in the nodes and DC transformers and DC lines in the edges, the DC grid is unified in the common graph of the AC grid. For an AC-DC system with $N$ buses, $b$ branches, and $a$ DC lines, its corresponding GNN has $N + 2a$ nodes and $b + 3a$ edges:

[0046]

[0047] where $H(\cdot)$ (0) is the initial hidden layer of the GNN and is equal to the input feature matrix $x$, and $P_{g}$ and i $P_{l}$ are the generation power and load power respectively, and $V$ or $\alpha$ min is the minimum value of the firing angle, $X_{r}$ is the rectifier reactance, $I_{d,k}^{r,ref}$ is the reference value of the rectifier current of the $k$-th DC line, $I_{d,k}^{i,ref}$ is the reference value of the inverter current of the $k$-th DC line. For the rectifier, or $\gamma$min is the minimum extinction angle, is the inverter reference voltage, represents the set of balancing nodes, and are the active power generation and active load power respectively, and are the reactive power generation and reactive load power respectively, represents the DC bus group, is the PQ bus group, is the point of common coupling bus group; the input edge features are composed of branches, denoted as edge feature matrix represents the eigenvector of each edge. For AC branches, x ε =[G ε , B ε , where G ε and B ε are the ε-line admittance parameters. For rectifiers and inverters, For DC branches, or is a trainable transformation matrix that embeds edge features into nodes. G ε and B ε are the ε-line conductance and susceptance parameters respectively, and are the minimum and maximum ratios of the commutation transformer, and R dc is the resistance of the DC line.

[0048] ChebNet is a graph filter because ChebNet can aggregate high-order neighbor information to obtain better performance. The feedforward of ChebNet is given by the following formula (L = D - A):

[0049]

[0050] In the formula, F represents the order of adjacent nodes, and H(·) (l) represents the response of the l-th layer with the response of the previous layer as the input. The output of GNN is the response of the last layer M, including the voltage and phase angle of the AC bus, the converter voltage and trigger angle of the DC bus, or [V, α(or γ)]; for simplicity, the present invention uses φ θ (x) to represent ChebNet;

[0051] Step S3: Taking the violation of the trigger / extinction angle constraint as the minimization objective, a physical-guided graph neural network PG-GNN that supports control mode switching is constructed based on the power flow model of the AC-DC hybrid system. The DC control mode switching is mastered through multiple physical-guided graph neural networks PG-GNN. Each control mode is experienced to train its specific PG-GNN. Then, the PG-GNN that meets the trigger / extinction angle constraint outputs the AC-DC power flow calculation results.

[0052] To achieve control mode switching, a physical-guided graph neural network PG-GNN that supports control mode switching is constructed with the violation of the trigger / extinction angle constraint as the minimization objective; and an equivalent optimization problem of the power flow model is formed:

[0053]

[0054] where the state variables to be solved are parameterized by the GNN φ θ (x), and f PQV , and are the equality constraint conditions of formulas (1)-

[0055] (3), (4), and (5) respectively; formula (1) includes (1a) and (1b), formula (2) includes (2a) and (2b), formula (4) includes (4a) and (4b), and formula (5) includes (5a) and (5b).

[0056] To simulate the process of tuning the trigger / extinction angle of the rectifier / inverter transformer, the following formula is given. Using f DC to represent uniformly and and omitting the subscripts:

[0057]

[0058] In the formula, represents the degree of violation of the trigger angle / extinction angle constraint, ReLU is the activation function, α min is the minimum trigger angle, α is the current trigger angle, K re is the ratio of the converter on the rectifier side, K re,max is the maximum allowable value of the converter ratio, and K re,min is the minimum allowable value of the converter ratio.

[0059] The solution of formula (10) can be approximated by the Lagrangian dual method. Considering that formula (10) is non-convex, the present invention introduces ALM to alleviate the difficulty of solving. ALM can usually alleviate the strong convexity condition and can obtain the tolerable performance of PG-GNN. The following learning strategy based on ALM is used:

[0060] (i ∈ {PQV, DC})

[0061]

[0062] Among them, \(i\) is a variable, \(PQV\) is the \(PQ\) busbar and the \(V\) busbar, and \(DC\) is all DC busbars. is the total loss function, which combines the constraint violation metric, the linear combination term, and the quadratic penalty term. \(\rho>0\) is the penalty parameter that controls the penalty strength for constraint violation; \(\lambda\) is a parameter of the GNN and is updated by minimizing the total loss function. \(\lambda\) i is the dual variable used to adjust the weight of the constraint term, \(\varphi\) θ \(\varphi(x)\) is the forward propagation function of the NN, which is used to calculate the output of the current state. Equation (11a) updates \(\theta\) by minimizing Equation (10). By fixing \(\theta\), the dual variable is then updated by Equation (11b). Equation (11) is continuously executed until Equation (10) remains stable. In Equation (10), As the loss function, it guides the optimization of the GNN model in the AC-DC power flow calculation. By combining the objective function and the constraint conditions, it realizes the comprehensive optimization of the system state, ensuring that the system reaches the best operating state while satisfying various constraint conditions.

[0063] Furthermore, Equation (11a) is implemented by the following formula:

[0064]

[0065] In the formula, \(\theta\) is the trainable parameter of the graph neural network GNN. is the loss function, \(\lambda\) is the Lagrange multiplier, \(\varphi\) θ \(\varphi(x)\) is the output of the GNN.

[0066] Finally, multiple PG-GNNs are used to master the DC control mode switching. Each control mode is experienced to train its specific PG-GNN. Then, the PG-GNN that meets the trigger / extinction angle constraint outputs the AC-DC power flow calculation result.

[0067] The above is only the preferred embodiment of the present invention and does not impose any form of limitation on the present invention. Any person skilled in the art can make many possible changes and modifications to the technical solution of the present invention or modify it into an equivalent embodiment with equivalent changes without departing from the scope of the technical solution of the present invention. Therefore, any changes, modifications, equivalent changes, and modifications made to the above embodiments based on the technical solution of the present invention without departing from the content of the technical solution of the present invention all fall within the protection scope of this technical solution.

Claims

1. A method for calculating AC and DC power flows based on switchable control mode graph learning, characterized in that: The following steps are involved: Step S1: construct a power flow model of an AC / DC hybrid system; The power flow model of the AC / DC hybrid system is: Control model 1: Control model 2: in, and They represent PV busbar group, AC busbar group, DC busbar group and common axis point group respectively. is the active power injected into the AC bus i, is the reactive power injected into the AC bus i, is the active power injected into the common axis point c, P cj (V c ,V j ,δ ij ) is the sum of the active powers transmitted through the lines connected to c, is the reactive power injected into the common axis point c, Q cj (V c ,V j ,δ ij ) is the sum of reactive powers transmitted through the lines connected to c, is the phase angle of the kth DC line, is the ratio k of the converter transformer on the rectifier side, is the ratio k of the converter transformer on the inverter side; represents the reactance of the converter connected to the inverter / rectifier DC bus k; α k / γ k is the triggering / extinguishing angle of the kth rectifier / inverter; represents the rectifier / inverter voltage of the kth DC line; R dc and I d,k represent the resistance and current of the kth DC line respectively; j∈i means that the jth bus is adjacent to the ith bus through the branch ij; Pij(V i ,V j ,δ ij ) and Q ij (V i ,V j ,δ ij ) is the function for calculating the active and reactive power of line ij; δ ij is the voltage phase difference between busbars i and j, V p is the voltage of the PV bus, is the PV bus voltage reference value, is the rectified current of the kth DC line, is the reference value of the rectified current of the kth DC line, is the reference value of the inverter voltage of the kth DC line, is the minimum value of the trigger angle of the kth rectifier / inverter, is the inverter current of the kth DC line; is the reference value of the inverter current of the kth DC line; Step S2: construct a graph neural network model of AC / DC power grid; embed AC / DC operation physical information into the GNN learning process; Step S3: Taking violation of the trigger / extinguishing angle constraint as the minimization target, and constructing a physical guidance graph neural network PG-GNN that supports control mode switching based on the power flow model of the AC / DC hybrid system, the DC control mode switching is mastered through multiple physical guidance graph neural networks PG-GNNs, and each control mode is experienced to train its specific PG-GNN. Then, the PG-GNN that meets the trigger / extinguishing angle constraint outputs the AC / DC power flow calculation results.

2. The method for calculating AC and DC power flows based on switchable control mode graph learning according to claim 1, characterized in that: The method for constructing the graph neural network model of the AC / DC power grid in step S2 is: By introducing converters in nodes and DC transformers and DC lines in edges, the DC grid is unified in the common graph of the AC grid. For an AC / DC system with N buses, b branches and 1 DC line, the corresponding GNN has N+2a nodes and b+3a edges: In the formula, H(·) (0) is the initial hidden layer of GNN, and are the power generation and load power, V i is the voltage; for the inverter, or α min is the minimum value of the firing angle, is the rectifying reactance, is the reference value of the rectified current of the kth DC line, is the reference value of the inverter current of the kth DC line. For the rectifier, or γ min is the minimum value of the extinction angle, is the inverter reference voltage, represents the set of balanced nodes, and are active generation power and active load power respectively, and are reactive power generation and reactive load power, respectively. It represents the DC bus group. For the PQ busbar group, is the common coupling point busbar group; the input edge feature is composed of branches, expressed as Edge feature matrix represents the eigenvector of each edge. For the AC branch, x ε =[G ε ,B ε ], where G ε and B ε is the ε line admittance parameter, for rectifier and inverter, For DC branch, or is a trainable transformation matrix that embeds edge features into nodes, G ε and B ε are the ε line conductance and susceptance parameters, and is the ratio of the minimum and maximum values ​​of the converter transformer, R dc is the resistance of the DC line.

3. The method for calculating AC and DC power flows based on switchable control mode graph learning according to claim 2, characterized in that: The feedforward of the graph neural network model is: In the formula, σ is the activation function, F represents the order of adjacent nodes, For: The polynomial filter of the graph Laplacian operator, according to the graph structure To operate, H(·) (l) represents the lth layer response with the previous layer response as input, x is the input feature matrix of GNN, represents the initial input, θ j is the trainable weight of the j-th order filter, is the output of GNN, which is equal to the response of the last layer M, including the voltage and phase angle of the AC bus, the converter voltage and trigger angle of the DC bus, Or [V, α (or γ)], φ θ (x) is the forward propagation function of the entire GNN, where θ is the trainable parameter of all layers, and [V,δ] is the output feature of the GNN, including the voltage V and phase angle δ of the AC bus.

4. The method for calculating AC and DC power flows based on switchable control mode graph learning according to claim 3 is characterized in that: In step S3, a physical guided graph neural network PG-GNN supporting control mode switching is constructed with violation of the trigger / extinguishing angle constraint as the minimization target; and an equivalent optimization problem of the power flow model is formed: Among them, the state variables to be solved are represented by GNNφ θ (x) parameterization, f PQV , and are the equality constraints of formulas (1)-(3), (4) and (5) respectively; The process of tuning the trigger / extinguishing angle of the rectifier / inverter transformer is simulated by the following formula, using f DC Unified Representation and and omit the subscript: In the formula, To indicate the degree of violation of the trigger angle / extinguishing angle constraint, ReLU is the activation function, α min is the minimum trigger angle, α is the current trigger angle, K re is the ratio of the rectifier side converter, K re,max is the maximum allowable value of the converter ratio, K re,min is the minimum allowable value of the converter ratio.

5. The method for calculating AC and DC power flows based on switchable control mode graph learning according to claim 4, characterized in that: In step S3, the physical guided graph neural network PG-GNN is optimized using the following ALM-based learning strategy: (i∈{PQV,DC}) Where i is a variable, PQV is the PQ bus and V bus, DC is all DC buses, is the total loss function, · combines the constraint violation measure, linear combination term and quadratic penalty term, ρ>0, is the penalty parameter, which controls the penalty intensity of constraint violation; θ is the parameter of GNN, which is updated by minimizing the total loss function, λ i is the dual variable used to adjust the weight of the constraint term, φ θ (x) is the forward propagation function of GNN, which is used to calculate the output of the current state. Formula (11a) updates θ by minimizing formula (10). By fixing θ, the dual variable is updated by formula (11b). Formula (11) is executed continuously until formula (10) remains stable.

6. The method for calculating AC and DC power flows based on switchable control mode graph learning according to claim 5, characterized in that: The formula (11a) is realized by the following formula: In the formula, θ is the trainable parameter of the graph neural network GNN, is the loss function, λ is the Lagrange multiplier, φ θ (x) is the output of GNN.

Images