Voltage stability critical point acquisition method and system

Abstract

The invention discloses a voltage stability critical point acquisition method and system, and relates to the technical field of power systems. Comprising the steps of obtaining voltage and power of a power grid, and obtaining a power flow equation by setting initial voltage and power injection values of nodes; linearizing a nonlinear problem of the power flow equation, and performing iteration by adopting power series expansion to obtain a power flow initial solution; performing power flow calculation on the power flow equation by using a recursive algorithm, and when power flow calculation divergence is caused by power grid change, supplementing the failure node voltage by using a data fitting compensation method until a voltage stability critical point is obtained, comprising the following steps: marking node voltages which are not calculated to form missing data; constructing a fitting function to reconstruct missing data according to the voltage and power results of the previous calculation; and supplementing the missing node voltage by using the voltage value of the previous load flow calculation and the fitting coefficient of the fitting function.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular, to a method and system for obtaining a voltage stability critical point. Background Art

[0002] Grid voltage stability is one of the key factors to ensure the safe and stable operation of the power system. The voltage stability critical point refers to the critical state of voltage collapse in the power grid. Accurately predicting the voltage stability critical point is crucial for the safe and stable operation of the power grid.

[0003] In the power system, power flow calculation is the basis for ensuring the safe and stable operation of the system. Traditional power flow calculation methods are mostly based on the Newton-Raphson method or the direct power flow method. However, when encountering power grid disturbances or new energy grid connection, these methods are prone to problems such as calculation divergence or non-convergence. With the large-scale grid connection of renewable energy such as wind power and photovoltaic power, the volatility of the system increases, bringing great challenges to power flow calculation. The main difficulty in solving the voltage stability critical point lies in that the Jacobian singular matrix may appear in the power flow equation before reaching the critical point, resulting in non-convergence of the iteration and difficulty in approaching the actual critical point. Especially when there is new energy grid connection in the power grid, the algorithm is difficult to respond to the change of node power flow. Summary of the Invention

[0004] In order to overcome the disadvantage that the above-mentioned existing technology may diverge when calculating close to the SNB (Saddle Node Bifurcation Point, SNB) and the voltage stability critical point cannot be obtained by calculation, the main object of the present invention is to provide a method and system for obtaining a voltage stability critical point.

[0005] To achieve the above object, the present invention adopts the following technical solutions. A method for obtaining a voltage stability critical point includes:

[0006] Obtain the voltage and power of the power grid, and obtain the power flow equation by setting the initial voltage and power injection values of the nodes;

[0007] Linearize the non-linear problem of the power flow equation, perform iteration by using power series expansion to obtain a linear power flow equation, and determine that the solution when the load increase step size in the linear power flow equation is 0 and the reactive power is 0 is the initial power flow solution;

[0008] Perform power flow calculation on the linear power flow equation through the initial power flow solution and the recursive algorithm to obtain the voltage of each node, and output the voltage stability critical point; wherein, when the power grid changes and causes the power flow calculation to diverge, mark the failed nodes for which the node voltages cannot be calculated. For these failed nodes, reconstruct the voltage data of the failed nodes according to the voltage values and power results obtained from the previous power flow calculation.

[0009] Reconstructing the voltage data of the failed node according to the voltage value and power result obtained from the previous power flow calculation includes the following steps:

[0010] Define each non-PV node after each power flow calculation. The V node represents a node with known active power P and voltage amplitude V. The voltage vector is as follows:

[0011] V i =[v i，1 ,...v i,n (7)

[0012] where n is the number of non-PV nodes, i is the number of power flow calculations, and i is an integer greater than or equal to 1;

[0013] Mark the node number of the node whose voltage has not been calculated as loss;

[0014] Use the voltage value and fitting coefficient α from the previous power flow calculation to construct a fitting function g(α), which is expressed by the formula as follows:

[0015]

[0016] where v i-1,n represents the voltage value of the (i - 1)-th power flow calculation of grid node n, and α 1 -α i-1 is the (i - 1)-th fitting coefficient;

[0017] Solve for the fitting coefficient using the fitting function, which is expressed by the formula as follows:

[0018]

[0019] where pinv() is the function for finding the pseudo-inverse matrix;

[0020] Use the fitting coefficient and the voltage value obtained from the previous power flow calculation to calculate the voltage v i,loss of the failed node that was not calculated during the i-th power flow calculation. The formula is as follows:

[0021]

[0022] Output V i =[v i，1 ,...v i,loss-1 ,v i,loss ,v i,loss+1 ,...v i,n as the voltage stability critical point, and the calculation ends.

[0023] After obtaining the voltage of each node, it further includes:

[0024] Calculate each order coefficient of the node voltage through a recursive algorithm, and verify the voltage accuracy in combination with Padé approximation; among them

[0025] The Padé approximation of order [L / M] is expressed as:

[0026]

[0027] where L and M are the degrees of the numerator and denominator series respectively, so as to maximize the convergence domain.

[0028] By setting the initial voltage and power injection value of the node, obtain the power flow equation; including the following steps:

[0029] Construct the power flow equation, expressed as follows:

[0030]

[0031] where λ is the load increase step, that is, for any node i, when a new energy power source is incorporated, the changes in node current and grid power flow will occur; S i , P i and |V i sp | respectively represent the complex power of the PQ node, P i is the active power of the PV node, represents the given voltage amplitude of the PV node, V i SL represents the given value of the balanced node voltage, Y ik,tr is the line admittance of the (i,k)-th element of the node admittance matrix, Y i,sh is the self-admittance to the ground of node i, N is the total number of columns of the admittance matrix; k is the number of columns of the admittance matrix; Y ik，tr is the line admittance of the (i,k)-th element of the node admittance matrix; V k (λ) is the function of the k-th node voltage with respect to λ; l is the step correction parameter; is the conjugate of the apparent power of the i-th node; Y i,sh is the self-admittance to the ground of node i; V i (λ) is the function of the i-th node voltage with respect to λ; V i * (λ) is the conjugate voltage of the i-th node; j is the imaginary part representation symbol; Q i (λ) is the reactive power of the i-th node embedded with λ; V i * (λ * ) is the conjugate of the i-th node voltage embedded with λ; |V i sp | 2 is the square of the PV node voltage amplitude; V iSL is the given value of the balanced node voltage;

[0032] When λ = 0, the initial power flow equation is obtained, expressed as:

[0033]

[0034] The obtaining of the linear power flow equation includes:

[0035] Linearize the non - linear problem of the power flow equation, expressed as follows:

[0036]

[0037] where V[n] and Q[n] are the n - th order power series coefficients of V and Q with respect to λ respectively;

[0038] The obtaining of the initial power flow solution includes:

[0039] In the initial power flow equation Then when the load increase step is 0 and the reactive power is 0, Q i (0) = Q i [0] = 0, the initial power flow equation has an initial power flow solution, expressed by the formula:

[0040]

[0041] The power flow calculation includes:

[0042] Define W i (λ) = 1 / V i (λ), and substitute of the n - th order into the power flow equation (1) and expand and extract the n - th order coefficients, and the equations are expressed as follows:

[0043]

[0044] In the above formula, the first two equations are the recurrence relations after extracting the n - th order coefficients of the power series expansion of the nodes with known active power P and reactive power Q; the third, fourth, and fifth equations are the recurrence relations after extracting the n - th order coefficients of the power series expansion of the PV nodes; the last two equations are the recurrence relations after extracting the n - th order coefficients of the power series expansion of the balanced nodes.

[0045] where, n is the power series order, l is 0 or 1;

[0046] where G ik,tr is the line conductance with respect to the (i,k) - th element; B ik,tr is the line susceptance with respect to the (i,k) - th element; V k,im [n] is V kThe real part of [n]; V k,re [n] is V k The imaginary part of [n]; P i * is the conjugate of the active power of the i-th node.

[0047] A system for obtaining the critical point of voltage stability, comprising:

[0048] An initial power flow calculation module, configured to obtain the voltage and power of the power grid, and obtain a power flow equation by setting the initial voltage and power injection values of the nodes;

[0049] A recursive power flow calculation module, configured to linearize the non-linear problem of the power flow equation, perform iteration by using power series expansion, obtain a linear power flow equation, and determine that the solution when the load increase step size in the linear power flow equation is 0 and the reactive power is 0 is the initial power flow solution;

[0050] A voltage stability critical point obtaining module, configured to perform power flow calculation on the linear power flow equation through the initial power flow solution and a recursive algorithm, obtain the voltage of each node, and output the voltage stability critical point; wherein, when the power grid changes and causes the power flow calculation to diverge, mark the failed nodes for which the node voltages cannot be calculated, and for the failed nodes, reconstruct the voltage data of the failed nodes according to the voltage values and power results obtained from the previous power flow calculation.

[0051] Compared with the prior art, the beneficial effects of the present invention are:

[0052] By introducing a recursive algorithm and Padé approximation, the present invention can improve the calculation accuracy of the voltage stability margin while maintaining the calculation efficiency. Through the data fitting compensation algorithm, the problem of calculation divergence under the conditions of new energy grid connection and power grid disturbance is effectively solved, and the divergence phenomenon of the traditional power flow calculation method is avoided.

[0053] The present invention provides a power flow calculation and divergence compensation method based on a recursive algorithm, which can solve the problem of calculation divergence in traditional power flow calculation under the conditions of new energy grid connection and power grid disturbance. Through power series expansion and data fitting compensation methods, the stability and accuracy of power flow calculation are improved, and it has strong applicability and generalization.

[0054] The present invention has strong generalization in the new energy grid connection environment, and can meet the power flow calculation requirements under the access of renewable energy such as photovoltaic and wind power. It can provide accurate voltage stability margin prediction to assist power system operation and maintenance personnel in making power grid dispatching decisions. Description of the Drawings

[0055] The accompanying drawings described herein are used to provide a further understanding of the present application, and constitute a part of the present application. The schematic embodiments of the present application and their descriptions are used to explain the present application, and do not constitute an improper limitation of the present application.

[0056] Figure 1 is a schematic diagram of the power flow recursive calculation process of the present invention;

[0057] Figure 2 is a schematic diagram of the power flow recursive algorithm in the embodiment of the present invention;

[0058] Figure 3 is a schematic diagram of the IEEE 14-node in the embodiment of the present invention. Detailed implementation manners

[0059] The static voltage stability critical point generally refers to the critical point related to the saddle-node bifurcation (Saddle Node Bifurcation Point, SNB). The main difficulty in solving the SNB currently lies in that the Jacobian singular matrix appears in the power flow equation before reaching the SNB point, the iteration cannot converge, and it is difficult to approach the actual SNB point. Especially when new energy is connected to the grid, the algorithm is difficult to respond to the change of node power flow. In response to this, this patent proposes a calculation method for the fast voltage stability critical point. This method can perform data fitting and compensation on the calculation failure caused by the ill-conditioning of the Jacobian matrix near the SNB, ensure the convergence of the algorithm for solving the SNB, and has the characteristics of fast, efficient and relatively accurate approximation to the target.

[0060] The technical route of the present invention is as Figure 1 shown. First, a dynamic power flow model with the load increment λ as the embedding parameter is established, and by introducing the power series f[x(λ)] and linearizing the power flow model for easy solution, the solution process is a recursive power flow calculation method. Since divergence may occur when calculating close to the SNB and the calculation cannot proceed, an algorithm that fits and reconstructs the current power flow with the calculated power flow data is adopted to make up for the data loss at the SNB.

[0061] The specific technical solutions are as follows:

[0062] 1. Define the initial solution of the power flow calculation

[0063] For PV nodes and PQ nodes, set their power injections and initial voltage values, and initialize the power flow equation. By setting the initial voltage and power injection values of each node in the power grid, the initial solution of the power flow calculation is determined.

[0064] 2. Introduce a recursive algorithm to solve the power flow

[0065] Based on the initial solution of the power flow calculation, a recursive algorithm is used for power flow calculation. Iteration is carried out based on the power flow equation (Equation 5), and the coefficients of each order of voltage and power are gradually calculated. By expanding the power series of the power flow equation, the original non-linear problem is transformed into a linear problem, which is convenient for recursive solution.

[0066] 3. Use Padé approximation for calculation verification

[0067] During the recursive calculation process, Padé approximation (Equation 6) is used to approximate each order of voltage and power to ensure the accuracy of the calculation results. Padé approximation can maximize the convergence domain and ensure the convergence and stability of the calculation process.

[0068] 4. Handle the problem of calculation divergence

[0069] When the power flow calculation diverges or the voltage does not converge, the data fitting compensation method is used for correction. The specific steps are as follows:

[0070] Mark missing data: After each calculation, mark the voltage of the failed nodes that cannot be calculated to form missing data.

[0071] Construct a fitting function: According to the voltage and power results of the previous calculation, construct a fitting function to reconstruct the missing data.

[0072] Supplement missing data: Supplement the missing node voltages through the coefficients of the fitting function and the voltage of the previous calculation to ensure the continuation of the calculation process and perform data supplementation.

[0073] 5. Output the prediction result of the voltage stability margin

[0074] Through multiple recursive iterations and data compensation, the predicted value of the voltage stability margin (SNB) is finally output, and the voltage stability of the system is analyzed. This method can accurately predict the voltage stability of the system under different load and new energy grid-connected conditions.

[0075] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0076] Embodiment:

[0077] First, construct a power flow model based on the power series of power changes

[0078] Suppose there are N nodes in the power grid with a high proportion of new energy. For any node i, when new energy sources are incorporated, the node current and power grid power flow will change, and this change is represented by the parameter λ, that is, the load increase step. Traditional power flow iterative algorithms may have problems such as calculation failure, no solution of the algorithm, or non-convergence. Therefore, a holomorphic classical power flow model for the new energy power grid is established:

[0079]

[0080] Among them, S i , P i and |V i sp | respectively represent the complex power of the PQ node, P i is the active power of the PV node, represents the given voltage amplitude of the PV node, V i SL represents the given value of the balanced node voltage, Y ik,tr and Y i,sh are the line admittance and the shunt admittance with respect to the (i, k)-th element of the nodal admittance matrix.

[0081] It can be seen from equation (1) that in addition to the voltage being a function of λ, the reactive power of the PV node is also a function of λ. When λ = 0, it is regarded as the initial state of the equation, and the initial solution can be expressed as:

[0082]

[0083] To obtain the initial solution in equation (2), the voltage of the node to be solved and the injection of generator reactive power are expressed as variables x. Then the power balance equation is f(x) = 0, that is, it satisfies the PQ node balance equation, the PV node balance equation, and the voltage equation of the balanced node. Express x as a function of the power series of λ:

[0084]

[0085] Then f(x) = 0 becomes a function of α containing a power series The purpose of introducing x(λ) is to convert the solution of the equation f(x) = 0 into the coefficients of x(λ), and then convert the problem of solving f(x) = 0 into the recursive solution of x[n]. This conversion is based on the convenience of solving the power flow. The original solution of the power flow f(x) = 0 is a non-linear problem, while the recursive solution is a linear problem. Then the solution of equation (1) expressed by the power series includes two parts:

[0086]

[0087] Among them, V[n] and Q[n] are the n-th order power series coefficients of V and Q with respect to λ respectively.

[0088] For equation (2), since If we let Q i (0) = Q i [0] = 0, then the power flow initial solution in the initial state is:

[0089]

[0090] Establish the recursive solution equation of the power flow.

[0091] After the initial power flow solution is determined, the continuation power flow calculation needs to be carried out. For the convenience of calculation, define W i (λ) = 1 / V i (λ), and substitute the nth-order expansion of into Equation (1). After extracting the nth-order coefficients and arranging them, the recursive equation is obtained:

[0092]

[0093] The above process transforms the non-linear solution of the power flow equation into a linear solution problem. Through the given initial power flow value and Equation (5), the recursive algorithm can solve any-order coefficient of the node voltage, and then until the power series expression of the voltage is obtained. For the target solution of the recursive algorithm, the Padé approximation value calculated by Equation (6) can be used as a verification to obtain the ideal voltage accuracy, so as to achieve the purpose of predicting the voltage stability margin.

[0094] The [L / M]-order Padé approximation can be expressed as:

[0095]

[0096] Among them, L and M are the degrees of the numerator and denominator series respectively. Usually, the approximate calculation of Equation (6) can ensure the maximization of the convergence domain.

[0097] Divergence compensation method for recursive power flow calculation

[0098] Since the recursive calculation process of Equation (5) has certain limitations, that is, the acquisition of the power series in the recursive algorithm is not infinite, and the calculation ends with accuracy as the index rather than the exact solution. When substituting into Equation (1), there is a deviation compared with the actual power flow result. As Figure 2 shown, for the curve s 1 it cannot finally converge to SNB, and for the curve s 2 it will approach SNB. Therefore, the algorithm will increase the probability of divergence and be unable to continue the calculation due to parameter changes caused by new energy grid connection or power grid disturbance. At this time, for s 1 the situation that appears needs to be compensated by data fitting.

[0099] The idea of the data compensation algorithm is to mark the data that cannot be obtained at the current level of the recursive algorithm, construct a fitting function, and the fitting coefficients in the function are generated by the determined solutions in the previous recursive power flow calculation. The missing data is fitted and reconstructed through the fitting function, and the reconstructed data is supplemented into the power flow equation so that the calculation can continue. The specific process is as follows:

[0100] Step1: Define i as the number of power flow calculation times, and let i = 1

[0101] Step 2: Define the voltage vectors of each non-PV node after each power flow calculation as shown in Equation (7):

[0102] V i =[v i，1 ,...v i,n (7)

[0103] where n is the number of non-PV nodes and i is the number of calculations.

[0104] Step 3: Check if there are any node voltages in V i that have not been calculated. If not, go to Step 7. If so, mark the node numbers with uncalculated node voltages as loss.

[0105] Step 4: Construct a function as shown in Equation (8).

[0106]

[0107] where v i-1,n represents the voltage value of the i-1th power flow calculation of grid node n. α 1 -α i-1 is the i-1th fitting coefficient.

[0108] Step 5: Solve for α 1 -α i-1 .

[0109]

[0110] where pinv() is the function for finding the pseudo-inverse matrix.

[0111] Step 6: Calculate v i,loss of the node voltage that has not been calculated in the i-th power flow calculation according to Equation (9).

[0112]

[0113] Step 7: Check if it reaches the voltage critical point. If not, set i = i + 1 and go back to Step 1. If so, directly output V i =[v i，1 ,...v i,loss-1 ,v i,loss ,v i,loss+1 ,...v i,n as SNB, and the calculation ends.

[0114] Example 2

[0115] Taking the IEEE 14-node system as an example, as Figure 3 shown, the grid parameters are shown in Table 1.

[0116] Table 1 IEEE 14 - node parameters

[0117]

[0118] Implementation steps:

[0119] 0): Define \(i\) as the number of power flow calculations, and let \(i = 1\).

[0120] 1): Define the voltage vectors of each non - PV node after each power flow calculation as shown in Equation ①:

[0121] \(\mathbf{V}\) i \(= [v\) i，1 ,...v\) i,n ①

[0122] Where \(n = 9\), that is, Figure 1 the number of non - PV nodes in \(\mathbf{V}\), \(i\) is the number of calculations, and the node voltage values after each power flow calculation are shown in Table 1.

[0123] 2): When \(i = 302\), the voltage \(v\) of the i node in \(\mathbf{V}\) i，7 is not calculated. Denote the node number with the uncalculated node voltage as \(loss = 7\).

[0124] 3): Construct a function as shown in Equation ②. The constructed matrix is 8 rows and 301 columns, that is, the power flow is recursively calculated up to 301 times without the voltage data of the 7th node.

[0125]

[0126] Where \(v\) i-1,n represents the voltage value of the \(n\)th power grid node in the \((i - 1)\)th power flow calculation. \(\alpha\) 1 -\(\alpha\) i-1 are \((i - 1)\) fitting coefficients.

[0127] 5): Use Equation ③ to solve for \(\alpha\) 1 -\(\alpha\) 302-1 .

[0128]

[0129] Where \(pinv()\) is the function to find the pseudo - inverse matrix.

[0130] 6): According to the recursive power flow calculation results of the first 301 times of the 7th node, the voltage \(v\) of the node with the uncalculated node voltage in the 302nd power flow calculation can be calculated according to Equation ④ 302,7 , as shown in Table 2.

[0131]

[0132] 7): Judge whether it reaches the voltage critical point. If not, then \(i = i + 1\) and go back to step 1). If yes, directly output \(\mathbf{V}\)​i = [v i，1 ,...v i,6 ,v i,7 ,v i,7 ,...v i,9 is the SNB critical point.

[0133] 8): The above cycle was carried out 442 times until the voltage reached the collapse point. The actual node corresponding to PQ node 7 in the compensation calculation is Figure 3 Node 12 in it. The calculated values of the deduced voltage values for each time are shown in Tables 2 and 3. The calculation results show that the algorithm of the present invention can compensate and calculate data under the condition of data loss, quickly solve the power flow, and has a small error.

[0134] Table 2 Results of recursive power flow calculation

[0135]

[0136]

[0137]

[0138]

[0139]

[0140]

[0141]

[0142] To prove the effectiveness of the compensated algorithm data, based on the first 301 data of the recursive power flow algorithm, starting from the 302nd calculation, the compensated algorithm and the recursive algorithm are used in parallel for comparative calculation. The results show that the compensated algorithm can effectively fill the gap of data loss, and the deviation between the voltage values fitted by its algorithm and the recursive algorithm is less than 1×10 -4 , and the calculation process is faster and more efficient, and it is more suitable for approaching the solution of SNB.

[0143] Table 3 Comparison of voltage (P.U.) results between recursive algorithm and compensated algorithm (compensating data of node 12)

[0144]

[0145]

[0146]

[0147] Photovoltaic and wind power sources with the same output as a traditional motor are connected to Node 3 and Node 6 respectively. The power output range of the wind turbine is 30% - 100% of the rated value, and the power output range of the photovoltaic generator is 0% - 100% of the rated value. The results are shown in Table 4. The results indicate that the algorithm of this patent can reasonably predict the SNB and load margin, and also has certain generalization and applicability for new energy grid-connected systems.

[0148] Table 4 Comparison of data for different power source connections

[0149]

[0150] It should be noted that in the present invention, 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 such actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device.

[0151] The above embodiments are only illustrative examples of the present invention and do not constitute a limitation on the protection scope of the present invention. Any design identical or similar to the present invention falls within the protection scope of the present invention.

Claims

1. A method for obtaining a voltage stability critical point, characterized in that: include: Obtain the voltage and power of the power grid, and obtain the power flow equation by setting the initial voltage and power injection value of the node; The nonlinear problem of the power flow equation is linearized, and it is iterated by power series expansion to obtain a linear power flow equation, and the solution when the load increase step is 0 and the reactive power is 0 in the linear power flow equation is determined as the initial power flow solution; The linear power flow equation is calculated by the initial power flow solution and the recursive algorithm to obtain the voltage of each node and the critical point of voltage stability. When the power grid changes and causes the power flow calculation to diverge, the failed nodes whose node voltages cannot be calculated are marked, and for the failed nodes, the voltage data of the failed nodes are reconstructed according to the voltage value and power result obtained from the previous power flow calculation.

2. The voltage stability critical point acquisition method according to claim 1, characterized in that: The voltage data of the failed node is reconstructed according to the voltage value and power result obtained by the previous power flow calculation, comprising the following steps: Define the voltage vector of each non-PV node after each power flow calculation. The PV node represents a node whose known quantities are active power P and voltage amplitude V. The formula of the voltage vector of each non-PV node is as follows: V i =[v i，1 ,...v i,n ] Where n is the number of non-PV nodes, i is the number of power flow calculations, and i is an integer greater than or equal to 1; The node numbers whose node voltages are not calculated are marked as loss; Using the voltage value and fitting coefficient α of the previous power flow calculation, the fitting function g(α) is constructed. The formula is as follows: Among them, v i-1,n represents the voltage value of the power flow calculation of the grid node n in the i-1th time, α1-α i-1 is the i-1th fitting coefficient; The fitting function is used to solve the fitting coefficient. The formula is as follows: Among them, pinv() is a function for finding the pseudo-inverse matrix; Using the fitting coefficient and the voltage value obtained from the previous power flow calculation, calculate the voltage v of the failed node that was not calculated during the i-th power flow calculation. i,loss , the formula is as follows: Output V i =[v i，1 ,...v i,loss-1 ,v i,loss ,v i,loss+1 ,...v i,n ] is the critical point of voltage stability.

3. The voltage stability critical point acquisition method according to claim 1, characterized in that: After obtaining the voltage of each node, the method further includes: The coefficients of each order of node voltage are calculated by recursive algorithm, and the voltage accuracy is verified in combination with Padé approximation; The [L / M]-order Padé approximation is expressed as: Among them, L and M are the degrees of the numerator and denominator series respectively, so as to maximize the convergence domain.

4. The voltage stability critical point acquisition method according to claim 1, characterized in that: The method of obtaining the power flow equation by setting the initial voltage and power injection value of the node comprises the following steps: By setting the initial voltage and power injection value of the node, the power flow equation is constructed as follows: Among them, λ is the load increase step size, that is, at any node i, when the new energy source is integrated, the node current and grid flow will change; S i , P i and V i sp Represents the complex power of PQ node, P i is the active power of the PV node, representing the given PV node voltage amplitude, V i SL represents the given value of the equilibrium node voltage, Y ik,tr is the line admittance of the node admittance matrix about the (i,k)th element, Y i,sh is the self-admittance of node i to the ground, N is the total number of columns in the admittance matrix; k is the number of columns in the admittance matrix; Y ik，tr is the line admittance of the (i, k)th element of the node admittance matrix; V k (λ) is the function of the kth node voltage with respect to λ; l is the step correction parameter; is the conjugate of the apparent power of the ith node; Y i,sh is the self-admittance of node i to the ground; V i (λ) is the function of the voltage of the ith node with respect to λ; V i * (λ) is the conjugate voltage of the ith node; j is the symbol of the imaginary part; Q i (λ) is the reactive power of the ith node embedded in λ; V i * (λ * ) is the conjugate of the voltage of the ith node embedded in λ; |V i sp | 2 is the square of the PV node voltage amplitude; V i SL is the given value of the equilibrium node voltage; When the load increase step λ = 0, the initial flow equation is obtained, which is expressed as:

5. The voltage stability critical point acquisition method according to claim 4, characterized in that: The obtaining of the linear power flow equation comprises: The nonlinear problem of the power flow equation is linearized and expressed as follows: Where V[n] and Q[n] are the n-order power series coefficients of V and Q with respect to λ, respectively; The obtaining of the initial solution of the power flow comprises: The initial power flow equation Then when the load increase step is 0 and the load increase step with embedding 0 is 0, the reactive power is 0, Q i (0) = Q i [0]=0, the initial flow equation has the initial flow solution, which is expressed as:

6. The voltage stability critical point acquisition method according to claim 1, characterized in that: The power flow calculation includes: Define W i (λ)=1 / V i (λ), and Substitute the nth order into the power flow equation (1) and expand it to extract the nth order coefficients. The formulas of each equation are expressed as follows: Among them, the first two equations in the formula are the recursive relations after extracting the n-order coefficients from the power series expansion of the nodes with known quantities of active power P and reactive power Q; the third, fourth, and fifth equations are the recursive relations after extracting the n-order coefficients from the power series expansion of the PV nodes; the last two equations are the recursive relations after extracting the n-order coefficients from the power series expansion of the equilibrium nodes; G ik,tr is the line conductance about the (i,k)th element; B ik,tr is the line susceptance about the (i,k)th element; V k,im [n] is V k The real part of [n]; V k,re [n] is V k The imaginary part of [n]; P i * is the conjugate of the active power of the ith node.

7. A voltage stability critical point acquisition system, characterized in that: include: The initial power flow calculation module is used to obtain the voltage and power of the power grid and obtain the power flow equation by setting the initial voltage and power injection value of the node; The nonlinear problem of the power flow equation is linearized, and it is iterated by power series expansion to obtain a linear power flow equation. When the load increase step in the linear power flow equation is 0 and the load increase step where the node embedding is 0 makes the reactive power 0, the initial power flow solution is obtained; A recursive power flow calculation module is used to perform power flow calculation by using the recursive algorithm of the power flow initial solution and the linear power flow equation to obtain the voltage of each node, wherein when the power grid changes, causing the power flow calculation to diverge, the voltage of the failed node that cannot be calculated is marked to form the voltage data of the missing node; The voltage stability critical point acquisition module is used to construct a fitting function and reconstruct the missing node voltage data based on the voltage and power results of the previous power flow calculation; the missing node voltage is supplemented by using the voltage value of the previous power flow calculation and the fitting coefficient of the fitting function, the missing node voltage is obtained, and the voltage is output as the voltage stability critical point.

8. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 2 to 6 is implemented.

9. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method described in any one of claims 2 to 6 is implemented.

Images