Method, device and medium for judging transient voltage stability of power system based on network energy

Abstract

The present application belongs to the technical field of power system, and discloses a method, device and medium for judging transient voltage stability of power system based on network energy. The method includes: obtaining partial response information of the network, and integrating along the trajectory of the partial response information of the network to construct a network energy function; based on the Pearson correlation coefficient, respectively constructing correlation indicators of active power and reactive power with voltage, quantitatively analyzing the correlation relationship between active power and reactive power and voltage, and extracting the network energy reactive term as the voltage dominant factor; based on the characteristic analysis of the network energy reactive term, converting the transient voltage stability problem into the convergence and divergence problem of the amplitude trajectory curve of the network energy reactive term, and using the maximum Lyapunov exponent to judge the transient voltage stability of the system. The present application not only has a fast calculation speed, but also has strong robustness, and can be applied to new systems with a large amount of new energy access; the present application can realize rapid judgment of transient voltage stability of power system.

Description

Technical Field

[0001] The present application relates to the technical field of power systems, and more specifically, to a method, device and medium for determining transient voltage stability of a power system based on network energy. Background Art

[0002] At present, how to accurately judge transient voltage instability is still a difficult problem that needs to be solved. The reason is that there is a connection between transient voltage stability and power angle stability. In the transient process after a fault, voltage collapse and power angle swing often occur at the same time, and it is difficult to distinguish which is the dominant factor causing system instability. Since the state variables of each component of the system change dramatically during the transient voltage instability process, the static analysis method of voltage stability that does not consider the dynamic process of the component cannot be applied to the judgment of transient voltage stability. To accurately judge transient and medium- and long-term voltage instability, a relatively more complex method that considers the dynamic process of the component is required.

[0003] Energy function method and transient voltage engineering criterion are commonly used methods for judging transient voltage stability of power systems. However, the above methods are not efficient and are not suitable for the current large-scale new energy grid-connected power system. Summary of the invention

[0004] The purpose of this application is to provide a method, device and medium for judging transient voltage stability of power system based on network energy, so as to ensure the correctness of the judgment result of transient voltage stability. For large-scale systems, this application not only has a fast calculation speed but also has a certain mechanism explanation, and is more robust. Compared with the transient voltage stability judgment method based on the maximum Lyapunov exponent of node voltage, this application also shows obvious advantages and can realize the rapid solution of transient voltage stability state of power system, as described below:

[0005] In order to achieve the above purpose, this application adopts the following technical solutions:

[0006] According to a first aspect of the present application, a method for determining transient voltage stability of a power system based on network energy is provided, the method comprising:

[0007] Obtaining partial response information of the network, and integrating along the trajectory of the partial response information of the network to construct a network energy function;

[0008] Based on the Pearson correlation coefficient, the correlation indexes of active power and reactive power with voltage are constructed respectively, the correlation relationship between active power and reactive power and voltage is quantitatively analyzed, and the reactive term of network energy is extracted as the dominant factor of voltage.

[0009] Based on the characteristic analysis of network energy reactive terms, the transient voltage stability problem is converted into the convergence and divergence problem of the amplitude trajectory curve of network energy reactive terms, and the maximum Lyapunov exponent is used to judge the transient voltage stability of the system.

[0010] Furthermore, partial network response information is obtained, and a network energy function is constructed by integrating along the trajectory of the partial network response information, including:

[0011] Extracting partial network response information; wherein the partial network response information includes system node voltage, phase angle, and node injected active power and reactive power;

[0012] By integrating the trajectory after the failure of the partial response information of the network, the network energy function of the system is obtained, which is expressed as:

[0013]

[0014] Where V i and V j are the voltage amplitudes of nodes i and j respectively; θ i and θ j are the voltage phase angles of node i and node j respectively; P ij Inject active power into node i; Q ij and Q ji are the reactive power injected into nodes i and j respectively; l is the number of branches in the power system; W line is the network energy function;

[0015] The network energy function consists of two parts: the network energy active term and the network energy reactive term. The network energy active term and the network energy reactive term are respectively expressed as:

[0016]

[0017] Where W P and W Q They are the network energy active term and network energy reactive term respectively.

[0018] Furthermore, based on the Pearson correlation coefficient, the correlation indicators of active power, reactive power and voltage are constructed respectively, including:

[0019] Extract V from the network i With P ij and Q ij The response trajectory uses covariance to measure V i With P ij and Q ij The overall error of

[0020] Extract V from the network i With P ijand Q ij The response trajectory uses the standard deviation formula to measure V i With P ij and Q ij The standard deviation of

[0021] The covariance is divided by the standard deviation to construct an indicator of the correlation between active power and reactive power and voltage.

[0022] Furthermore, extract the V i With P ij and Q ij The response trajectory is calculated by the following formula to calculate the covariance, which measures V i With P ij and Q ij The overall error is:

[0023] Cov(V i ,P ij )=E(V i P ij )-E(V i )E(P ij )

[0024] Cov(V i ,Q ij )=E(V i Q ij )-E(V i )E(Q ij )

[0025] In the formula, E is the mathematical expectation, Cov is the covariance, V i is the voltage amplitude of node i, P ij Inject active power into node i; Cov(V i ,P ij ) is the covariance of the voltage amplitude of node i and the active power injected into node i, Cov(V i ,Q ij ) is the covariance of the voltage amplitude of node i and the reactive power injected into node i, E(V i P ij ) is the mathematical expectation of the product of the voltage amplitude of node i and the active power injected into node i, E(V i ) is the mathematical expectation of the voltage amplitude at node i, E(P ij ) is the mathematical expectation of the active power injected into node i, E(V i Q ij ) is the mathematical expectation of the product of the voltage amplitude at node i and the reactive power injected at node i, E(Q ij ) is the mathematical expectation of the reactive power injected into node i.

[0026] Furthermore, the standard deviation formula is:

[0027]

[0028] In the formula, σ(V i ),σ(P ij ) and σ(Q ij ) are V i , P ij and Q ij The standard deviation of E(V i 2 ) is the mathematical expectation of the square of the voltage amplitude at node i, E 2 (V i ) is the square of the mathematical expectation of the voltage amplitude at node i, E(P ij 2 ) is the mathematical expectation of the square of active power injected into node i, E 2 (P ij ) is the square of the mathematical expectation of the active power injected into node i, E(Q ij 2 ) is the mathematical expectation of the square of reactive power injected into node i, E 2 (Q ij ) is the square of the mathematical expectation of the reactive power injected into node i.

[0029] Furthermore, the correlation index between active power and reactive power and voltage is constructed by dividing the covariance by the standard deviation using the following formula:

[0030]

[0031] In the formula, is the Pearson correlation coefficient between the voltage amplitude of node i and the active power injected into node i, is the Pearson correlation coefficient between the voltage amplitude at node i and the reactive power injected at node i.

[0032] Furthermore, based on the characteristic analysis of the network energy reactive term, the transient voltage stability problem is converted into the convergence and divergence problem of the network energy reactive term amplitude trajectory curve, and the maximum Lyapunov exponent is used to judge the transient voltage stability of the system, including:

[0033] The network energy reactive term W Q The response trajectory is converted to W' Q -ΔW Q plane, by judging ΔW Q Whether the trajectory converges or not is used to evaluate the transient voltage stability of the system; where W' Q W Q Amplitude change rate, ΔW Q W Q The amplitude difference is calculated by the following formula to convert the network energy reactive term WQ The response trajectory is converted to W' Q -ΔW Q flat:

[0034]

[0035] Where W Qn and W Q(n-1) are W of the nth and n-1th time intervals respectively. Q Amplitude, W Q0 is the amplitude at the initial moment, ΔW Qn and ΔW Q(n-1) are W for the nth and n-1th time intervals respectively. Q Amplitude difference, W' Qn is the nth time interval W Q The amplitude change rate, Δt is the sampling time interval;

[0036] Using the maximum Lyapunov exponent to determine ΔW Q Whether the amplitude trajectory curve converges is to determine the system transient voltage stability state. The formula is as follows:

[0037]

[0038] Where N is the initial time interval, k and n are both positive integers, kΔt represents a certain moment, λ(kΔt) is the maximum Lyapunov exponent at kΔt, ΔW Q(k+n) is the network energy reactive term W in the k+nth time interval Q Amplitude difference, ΔW Q(k+n-1) is the network energy reactive term W in the k+n-1th time interval Q Amplitude difference.

[0039] According to a second aspect of the present application, a network energy-based power system transient voltage stability determination device is provided, the device comprising:

[0040] An energy function construction module is configured to obtain partial network response information, and integrate along the trajectory of the partial network response information to construct a network energy function;

[0041] The reactive term extraction module is configured to construct correlation indicators of active power and reactive power with voltage based on the Pearson correlation coefficient, quantitatively analyze the correlation relationship between active power and reactive power and voltage, and extract the network energy reactive term as the voltage dominant factor;

[0042] The stability judgment module is configured to convert the transient voltage stability problem into the convergence and divergence problem of the amplitude trajectory curve of the network energy reactive term based on the characteristic analysis of the network energy reactive term, and use the maximum Lyapunov exponent to judge the transient voltage stability of the system.

[0043] Furthermore, the energy function building module is further configured as follows:

[0044] Extracting partial network response information; wherein the partial network response information includes system node voltage, phase angle, and node injected active power and reactive power;

[0045] By integrating the trajectory after the failure of the partial response information of the network, the network energy function of the system is obtained, which is expressed as:

[0046]

[0047] Where V i and V j are the voltage amplitudes of nodes i and j respectively; θ i and θ j are the voltage phase angles of node i and node j respectively; P ij Inject active power into node i; Q ij and Q ji are the reactive power injected into nodes i and j respectively; l is the number of branches in the power system; W line is the network energy function;

[0048] The network energy function consists of two parts: the network energy active term and the network energy reactive term. The network energy active term and the network energy reactive term are respectively expressed as:

[0049]

[0050] Where W P and W Q They are the network energy active term and network energy reactive term respectively.

[0051] According to a third aspect of the present application, a readable storage medium is provided, wherein the readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the method as described above.

[0052] This application has at least the following beneficial effects:

[0053] The present application can realize accurate judgment of transient voltage stability. Compared with transient voltage engineering judgment criteria and voltage-based MLE methods, the present application is more timely. The present application improves the network energy function and extracts the voltage-dominant term in the network energy function through a network energy-voltage correlation index based on the correlation coefficient. The network energy model is simplified while ensuring the accuracy of the model, thereby improving the efficiency of stability judgment. Compared with traditional model-driven and data-driven methods, the transient voltage stability judgment method of the present invention that integrates the model-driven and data-driven methods combines the mechanism interpretability of the model-driven method and the scenario adaptability of the data-driven method, has a clear physical model and improves the speed of stability judgment. The transient voltage stability judgment method of the present invention has good adaptability to the current large-scale new energy grid-connected power system, and provides a feasible idea for the identification of transient voltage stability states of new power system flows. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 A flow chart of a method for determining transient voltage stability of a power system based on network energy according to an embodiment of the present application is shown.

[0055] Figure 2 A correlation coefficient diagram of the active power and reactive power of a power system branch and the node voltage in a method for determining transient voltage stability of a power system based on network energy according to an embodiment of the present application is shown;

[0056] Figure 3 The network energy reactive term W of a method for judging transient voltage stability of a power system based on network energy according to an embodiment of the present application is shown. Q Response trajectory plot;

[0057] Figure 4 FIG. 1 shows a method for determining transient voltage stability of a power system based on network energy according to an embodiment of the present application. Q -ΔW Q Plane curve graph;

[0058] Figure 5 A modified New England 10-machine 39-node example network topology diagram according to an embodiment of the present application is shown;

[0059] Figure 6 A correlation coefficient result diagram under stable conditions according to an embodiment of the present application is shown;

[0060] Figure 7 The maximum Lyapunov exponent MLE image under stable conditions according to an embodiment of the present application is shown;

[0061] Figure 8 A correlation coefficient result diagram under instability conditions according to an embodiment of the present application is shown;

[0062] Fig. 9 The maximum Lyapunov exponent MLE image under instability according to an embodiment of the present application is shown;

[0063] Fig.10 A structural diagram of a power system transient voltage stability determination device based on network energy according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0064] In order to enable those skilled in the art to better understand the technical solution of the present application, the present application is described in detail below in conjunction with the accompanying drawings and specific implementation methods. The embodiments of the present application are further described in detail below in conjunction with the accompanying drawings and specific embodiments, but are not intended to limit the present application. For the various steps described herein, if there is no necessity for a contextual relationship between each other, the order in which they are described as examples herein should not be regarded as a limitation, and those skilled in the art should know that they can be adjusted in order, as long as the logic between them is not destroyed, resulting in the inability to implement the entire process.

[0065] The embodiment of the present application provides a method for determining transient voltage stability of a power system based on network energy. Figure 1 The flowchart of the method is shown in FIG. The method comprises the following steps:

[0066] S1. Obtain partial network response information, and integrate along the trajectory of the partial network response information to construct a network energy function.

[0067] In some embodiments, step S1 specifically includes:

[0068] Step S11, extracting partial network response information; wherein the partial network response information includes system node voltage, phase angle, and node injected active power and reactive power.

[0069] Step S12: Integrate the trajectory after the partial network response information fails to obtain the network energy function of the system, which is expressed as:

[0070]

[0071] Where V i and V j are the voltage amplitudes of nodes i and j respectively; θ i and θ j are the voltage phase angles of node i and node j respectively; P ij Inject active power into node i; Q ij and Q ji are the reactive power injected into nodes i and j respectively; l is the number of branches in the power system; W line is the network energy function.

[0072] The network energy function consists of two parts: the network energy active term and the network energy reactive term. The network energy active term and the network energy reactive term are respectively expressed as:

[0073]

[0074] Where W P and W Q They are the network energy active term and network energy reactive term respectively.

[0075] In this embodiment, the network energy function constructed based on the partial response information of the system does not need to perform complex modeling of the system generator. It relies on branch information, and the modeling is simple and fast. At the same time, it ensures the interpretability of the system transient voltage instability, laying a model foundation for using MLE to judge the transient voltage stability in the following text.

[0076] S2. Based on the Pearson correlation coefficient, the correlation indicators of active power, reactive power and voltage are constructed respectively, the correlation relationship between active power, reactive power and voltage is quantitatively analyzed, and the reactive term of network energy is extracted as the dominant factor of voltage.

[0077] In some embodiments, step S2 specifically includes:

[0078] S21. Extract V from the network i With P ij and Q ij The response trajectory uses covariance to measure V i With P ij and Q ij The overall error of the two variables is shown as follows:

[0079] Cov(V i ,P ij )=E(V i P ij )-E(V i )E(P ij )

[0080] Cov(V i ,Q ij )=E(V i Q ij )-E(V i )E(Q ij )

[0081] In the formula, E is the mathematical expectation, Cov is the covariance, V i is the voltage amplitude of node i, P ij Inject active power into node i; Cov(V i ,P ij) is the covariance of the voltage amplitude of node i and the active power injected into node i, Cov(V i ,Q ij ) is the covariance of the voltage amplitude of node i and the reactive power injected into node i, E(V i P ij ) is the mathematical expectation of the product of the voltage amplitude of node i and the active power injected into node i, E(V i ) is the mathematical expectation of the voltage amplitude at node i, E(P ij ) is the mathematical expectation of the active power injected into node i, E(V i Q ij ) is the mathematical expectation of the product of the voltage amplitude at node i and the reactive power injected at node i, E(Q ij ) is the mathematical expectation of the reactive power injected into node i.

[0082] S22, extract V in the network i With P ij and Q ij The response trajectory uses the standard deviation formula to measure V i With P ij and Q ij The standard deviation of two variables is shown as follows:

[0083]

[0084] In the formula, σ(V i ),σ(P ij ) and σ(Q ij ) are V i , P ij and Q ij The standard deviation of E(V i 2 ) is the mathematical expectation of the square of the voltage amplitude at node i, E 2 (V i ) is the square of the mathematical expectation of the voltage amplitude at node i, E(P ij 2 ) is the mathematical expectation of the square of active power injected into node i, E 2 (P ij ) is the square of the mathematical expectation of the active power injected into node i, E(Q ij 2 ) is the mathematical expectation of the square of reactive power injected into node i, E 2 (Q ij ) is the square of the mathematical expectation of the reactive power injected into node i.

[0085] S23. Divide the obtained covariance by the standard deviation to construct a voltage correlation index, and the formula is as follows:

[0086]

[0087] In the formula, is the Pearson correlation coefficient between the voltage amplitude of node i and the active power injected into node i, is the Pearson correlation coefficient between the voltage amplitude at node i and the reactive power injected at node i.

[0088] like Figure 2 As shown in FIG. 1 , it is a correlation coefficient diagram of the active power and reactive power of the power system branch and the node voltage in the transient voltage stability judgment of the power system based on network energy provided by the embodiment of the present application. As can be seen from the figure, in the stable condition, after the fault is removed, as the voltage recovers to the vicinity of the initial value, V i With P ij Correlation coefficient Much smaller than V i With Q ij Correlation coefficient In the case of instability, after the fault is removed, as the voltage continues to drop, V i With P ij Correlation coefficient Still much smaller than V i With Q ij Correlation coefficient That is, reactive power has a stronger correlation with voltage than active power. P and W Q Formula, extract W Q It is the voltage dominant term in the network energy function.

[0089] S3. Based on the characteristic analysis of the network energy reactive term, the transient voltage stability problem is converted into the convergence and divergence problem of the amplitude trajectory curve of the network energy reactive term, and the maximum Lyapunov exponent is used to judge the transient voltage stability of the system.

[0090] Figure 3 The W extracted based on the voltage correlation index is shown line Medium voltage dominant term W Q The response curve of . It can be found from the figure that after the fault is removed and the transient voltage is stable, W Q The response trajectory of the voltage track returns to the initial voltage value and converges to a stable value; when the transient voltage is unstable, the voltage track drops and oscillates violently, and W Q The response curve also drops in the initial period and then oscillates and diverges.

[0091] Further, according to the transient voltage stability and W Q The coupling between the response trajectory characteristics is Q The response trajectory is converted to W' Q -ΔW Q plane, thus transforming the transient voltage stability problem into W'Q -ΔW Q The convergence and divergence of the plane response curve can be solved by judging ΔW Q Whether the trajectories converge or not, and then evaluate the transient voltage stability of the system, the formula is as follows:

[0092]

[0093] Where W Qn and W Q(n-1) are W at the nth and (n-1th)th time intervals respectively. Q Amplitude, W Q0 W Qn Initial amplitude 0, ΔW Qn and ΔW Q(n-1) The nth and (n-1)th time intervals W Q Amplitude difference, W' Q is the nth time interval W Q The amplitude change rate, Δt is the sampling time interval.

[0094] Figure 4 W' Q -ΔW Q Planar response curve diagram, from the figure, we can see that under the condition of transient voltage stability, W' Q -ΔW Q The trajectory will converge asymptotically to a stable fixed point, where W' Q Approaching 0, ΔW Q Approaches a fixed value; in the case of transient voltage instability, W' Q -ΔW Q The trajectory will continue to oscillate and cannot converge to a fixed value, where W' Q Wave divergence, ΔW Q It is impossible to approach a fixed value. In summary, by judging ΔW Q Whether the transient voltage is unstable can be determined by whether it can converge, and then the system voltage stability state can be determined.

[0095] Furthermore, the maximum Lyapunov exponent is used to determine ΔW Q Whether the amplitude trajectory curve converges, and then judge the system transient voltage stability state, the formula is as follows:

[0096]

[0097] Where N is the initial time interval, k and n are both positive integers, kΔt represents a certain moment, λ(kΔt) is the maximum Lyapunov exponent at kΔt, ΔW Q(k+n) is the network energy reactive term W in the k+nth time interval Q Amplitude difference, ΔW Q(k+n-1)is the network energy reactive term W in the k+n-1th time interval Q Amplitude difference.

[0098] Lyapunov index is one of the important concepts for evaluating the stability of nonlinear dynamic systems. It is often used as an important parameter to describe the average exponential divergence rate of adjacent trajectories in phase space. Among them, MLE can be used to judge the overall stability of the system after being disturbed. If MLE>0, it indicates that adjacent trajectories will separate, and the system tends to be unstable; if MLE<0, it indicates that adjacent trajectories will converge, and the system tends to be stable. That is, if ΔW Q If MLE>0, the system transient voltage is unstable; if ΔW Q When MLE<0, the system transient voltage will tend to be stable.

[0099] In an exemplary embodiment, taking the New England 10-machine 39-node system as an example, a simulation analysis of the system is performed based on the method described above.

[0100] Figure 5 This is the modified New England 10-machine 39-node example network topology diagram. Figure 5 The numbers 1 to 39 represent 39 nodes respectively. This example includes 10 synchronous generators, all of which adopt the classic second-order model, with a total active power of 6520MW, including 900MW of wind power, and a total load of 5814MW. The wind power is connected to the system at nodes 20, 25 and 29 through step-up transformers. A three-phase short circuit fault occurs at node 5 at 0.2s, and the fault is removed at 0.25s. At this time, the system voltage is stable. Applying the transient voltage stability judgment method of the power system based on network energy proposed in this application, first identify the most severely disturbed node, that is, the node with the largest voltage drop, and determine its branch. Then extract the P of this branch. ij and Q ij response data, and then use the voltage correlation index to calculate ρ Vi,Pij and ρ Vi,Qij , compare the sizes of the two. Figure 6 For stable conditions, the correlation coefficient ρ Vi,Pij and ρ Vi,Qij Calculation results: From the figure, we can see that ρ Vi,Pij is 0.93, much larger than ρ Vi,Qij is -0.15, so W is determined Q is the voltage dominant term in the network energy function. ij , Q ji and V i 、V j Calculate W Q , and then W Q Amplitude trace converted to W' Q -ΔW QPlane, use MLE to determine ΔW Q The convergence and divergence can be used to determine the transient voltage stability of the system. Figure 7 To use ΔW Q The MLE curve obtained is compared with the MLE curve obtained by using voltage. It can be seen from the figure that ΔW Q The first swing vertex of the MLE curve is less than zero, so ΔW can be determined at 0.48s. Q The trajectory converges, that is, the system transient voltage is stable; however, the first swing vertex of the voltage MLE curve is greater than zero, and the subsequent swing vertices are all greater than zero, which can easily lead to incorrect identification of the transient voltage stability state.

[0101] It is set that a three-phase short circuit fault occurs at node 5 at 0.2s, and the fault is removed at 0.39s. At this time, the system voltage is unstable. The present invention is used to determine the branch where the node with the most serious voltage drop is located, and then extract the P ij and Q ij Response data, voltage correlation index is used to calculate ρ Vi,Pij and ρ Vi,Qij , compare the sizes of the two. Figure 8 In the case of instability, the correlation coefficient ρ Vi,Pij and ρ Vi,Qij Calculation results: From the figure, we can see that ρ Vi,Pij is 0.89, much larger than ρ Vi,Qij is -0.24, so W is determined Q is the voltage dominant term in the network energy function. ij , Q ji and V i 、V j Calculate W Q , and then W Q Amplitude trace converted to W' Q -ΔW Q Plane, use MLE to determine ΔW Q The convergence and divergence can be used to determine the transient voltage stability of the system. Fig. 9 To use ΔW Q The MLE curve obtained is compared with the MLE curve obtained by using voltage. It can be seen from the figure that ΔW Q The MLE curve of the first swing back crosses the zero point at 0.48s, so ΔW can be judged at 0.48s. Q The trajectory diverges, which means that the system transient voltage is unstable; and the voltage MLE curve first swings back and crosses the zero point at 0.82s. It can be seen that this application has better timeliness than using the voltage MLE curve to judge the system transient voltage stability.

[0102] The present application also provides a device for determining transient voltage stability of a power system based on network energy, such as Fig.10 As shown, the device comprises:

[0103] The energy function construction module 1001 is configured to obtain partial network response information, and integrate along the trajectory of the partial network response information to construct a network energy function;

[0104] The reactive term extraction module 1002 is configured to construct correlation indices of active power and reactive power with voltage respectively based on the Pearson correlation coefficient, quantitatively analyze the correlation relationship between active power and reactive power and voltage, and extract the network energy reactive term as the voltage dominant factor;

[0105] The stability determination module 1003 is configured to convert the transient voltage stability problem into the convergence and divergence problem of the amplitude trajectory curve of the network energy reactive term based on the characteristic analysis of the network energy reactive term, and use the maximum Lyapunov exponent to determine the transient voltage stability of the system.

[0106] In some embodiments, the energy function building module is further configured to:

[0107] Extracting partial network response information; wherein the partial network response information includes system node voltage, phase angle, and node injected active power and reactive power;

[0108] By integrating the trajectory after the failure of the partial response information of the network, the network energy function of the system is obtained, which is expressed as:

[0109]

[0110] Where V i and V j are the voltage amplitudes of nodes i and j respectively; θ i and θ j are the voltage phase angles of node i and node j respectively; P ij Inject active power into node i; Q ij and Q ji are the reactive power injected into nodes i and j respectively; l is the number of branches in the power system; W line is the network energy function;

[0111] The network energy function consists of two parts: the network energy active term and the network energy reactive term. The network energy active term and the network energy reactive term are respectively expressed as:

[0112]

[0113] Where W P and W Q They are the network energy active term and network energy reactive term respectively.

[0114] In some embodiments, the reactive term extraction module is further configured to extract V i With P ij and Q ij The response trajectory uses covariance to measure V i With P ij and Q ij The total error of the extraction network V i With P ij and Q ij The response trajectory uses the standard deviation formula to measure V i With P ij and Q ij Standard deviation of: Divide the covariance by the standard deviation to construct an indicator of the correlation between active power and reactive power and voltage.

[0115] In some embodiments, the reactive term extraction module is further configured to extract V i With P ij and Q ij The response trajectory is calculated by the following formula to calculate the covariance, which measures V i With P ij and Q ij The overall error is:

[0116] Cov(V i ,P ij )=E(V i P ij )-E(V i )E(P ij )

[0117] Cov(V i ,Q ij )=E(V i Q ij )-E(V i )E(Q ij )

[0118] In the formula, E is the mathematical expectation, Cov is the covariance, V i is the voltage amplitude of node i, P ij Inject active power into node i; Cov(V i ,P ij ) is the covariance of the voltage amplitude of node i and the active power injected into node i, Cov(V i ,Q ij ) is the covariance of the voltage amplitude of node i and the reactive power injected into node i, E(V i P ij ) is the mathematical expectation of the product of the voltage amplitude of node i and the active power injected into node i, E(V i ) is the mathematical expectation of the voltage amplitude at node i, E(Pij ) is the mathematical expectation of the active power injected into node i, E(V i Q ij ) is the mathematical expectation of the product of the voltage amplitude at node i and the reactive power injected at node i, E(Q ij ) is the mathematical expectation of the reactive power injected into node i.

[0119] In some embodiments, the standard deviation formula is:

[0120]

[0121] In the formula, σ(V i ),σ(P ij ) and σ(Q ij ) are V i , P ij and Q ij The standard deviation of E(V i 2 ) is the mathematical expectation of the square of the voltage amplitude at node i, E 2 (V i ) is the square of the mathematical expectation of the voltage amplitude at node i, E(P ij 2 ) is the mathematical expectation of the square of active power injected into node i, E 2 (P ij ) is the square of the mathematical expectation of the active power injected into node i, E(Q ij 2 ) is the mathematical expectation of the square of reactive power injected into node i, E 2 (Q ij ) is the square of the mathematical expectation of the reactive power injected into node i.

[0122] In some embodiments, the reactive term extraction module is further configured to construct a correlation index of active power and reactive power with voltage by dividing the covariance by the standard deviation using the following formula:

[0123]

[0124] In the formula, is the Pearson correlation coefficient between the voltage amplitude of node i and the active power injected into node i, is the Pearson correlation coefficient between the voltage amplitude at node i and the reactive power injected at node i.

[0125] In some embodiments, the stability determination module is further configured to:

[0126] The network energy reactive term W Q The response trajectory is converted to W' Q -ΔW Q plane, by judging ΔWQ Whether the trajectory converges or not is used to evaluate the transient voltage stability of the system; where W' Q W Q Amplitude change rate, ΔW Q W Q The amplitude difference is calculated by the following formula to convert the network energy reactive term W Q The response trajectory is converted to W' Q -ΔW Q flat:

[0127]

[0128] Where W Qn and W Q(n-1) are W of the nth and n-1th time intervals respectively. Q Amplitude, W Q0 is the amplitude at the initial moment, ΔW Qn and ΔW Q(n-1) are W for the nth and n-1th time intervals respectively. Q Amplitude difference, W' Qn is the nth time interval W Q The amplitude change rate, Δt is the sampling time interval;

[0129] Using the maximum Lyapunov exponent to determine ΔW Q Whether the amplitude trajectory curve converges is to determine the system transient voltage stability state. The formula is as follows:

[0130]

[0131] Where N is the initial time interval, k and n are both positive integers, kΔt represents a certain moment, λ(kΔt) is the maximum Lyapunov exponent at kΔt, ΔW Q(k+n) is the network energy reactive term W in the k+nth time interval Q Amplitude difference, ΔW Q(k+n-1) is the network energy reactive term W in the k+n-1th time interval Q Amplitude difference.

[0132] It should be noted that the various device structures described in this embodiment belong to the same technical concept as the previously described method, and achieve the same technical effect through the same principle, which will not be repeated here.

[0133] An embodiment of the present application further provides a readable storage medium, which stores one or more programs. The one or more programs can be executed by one or more processors to implement the method described in any of the above embodiments.

[0134] In addition, although exemplary embodiments have been described herein, the scope includes any and all embodiments based on the present application with equivalent elements, modifications, omissions, combinations (e.g., various embodiments intersecting schemes), adaptations or changes. The elements in the claims will be interpreted broadly based on the language adopted in the claims, and are not limited to the examples described in this specification or during the implementation of the application, and the examples will be interpreted as non-exclusive. Therefore, this specification and examples are intended to be considered as examples only, and the true scope and spirit are indicated by the following claims and the full scope of their equivalents.

[0135] The above description is intended to be illustrative rather than restrictive. For example, the above examples (or one or more of them) can be used in combination with each other. For example, those of ordinary skill in the art can use other embodiments when reading the above description. In addition, in the above-mentioned specific embodiments, various features can be grouped together to simplify the application. This should not be interpreted as a feature of an application that does not require protection being necessary for any claim. On the contrary, the subject matter of the present application may be less than all the features of the embodiments of a specific application. Thus, the following claims are incorporated into the specific embodiments as examples or embodiments, wherein each claim is independently used as a separate embodiment, and it is considered that these embodiments can be combined with each other in various combinations or arrangements. The scope of the present application should be determined with reference to the full scope of the equivalent forms of the attached claims and these claims.

Claims

1. A method for judging transient voltage stability of a power system based on network energy, characterized in that: The method comprises: Obtaining partial response information of the network, and integrating along the trajectory of the partial response information of the network to construct a network energy function; Based on the Pearson correlation coefficient, the correlation indexes of active power and reactive power with voltage are constructed respectively, the correlation relationship between active power and reactive power and voltage is quantitatively analyzed, and the network energy reactive term is extracted as the dominant factor of voltage. Based on the characteristic analysis of network energy reactive term, the transient voltage stability problem is converted into the convergence and divergence problem of the amplitude trajectory curve of network energy reactive term, and the maximum Lyapunov exponent is used to judge the transient voltage stability of the system. Obtaining partial network response information, and integrating along the trajectory of the partial network response information to construct a network energy function, including: Extracting partial network response information; wherein the partial network response information includes system node voltage, phase angle, and node injected active power and reactive power; By integrating the trajectory after the failure of the partial response information of the network, the network energy function of the system is obtained, which is expressed as: Where V i and V j are the voltage amplitudes of node i and node j respectively; P ij Inject active power into node i; Q ij and Q ji are the reactive power injected into nodes i and j respectively; l is the number of branches in the power system; W line is the network energy function; The network energy function consists of two parts: the network energy active term and the network energy reactive term. The network energy active term and the network energy reactive term are respectively expressed as: Where W P and W Q They are the network energy active term and network energy reactive term respectively.

2. The method according to claim 1, characterized in that Based on the Pearson correlation coefficient, the correlation indicators of active power, reactive power and voltage are constructed respectively, including: Extract V from the network i With P ij and Q ij The response trajectory uses covariance to measure V i With P ij and Q ij The overall error of Extract V from the network i With P ij and Q ij The response trajectory uses the standard deviation formula to measure V i With P ij and Q ij The standard deviation of The covariance is divided by the standard deviation to construct an indicator of the correlation between active power and reactive power and voltage.

3. The method according to claim 2, characterized in that Extract V from the network i With P ij and Q ij The response trajectory is calculated by the following formula to calculate the covariance, which measures V i With P ij and Q ij The overall error is: The (V i ,P ij )=E(V i P ij )-E(V i )E(P ij ) The (V i ,Q ij )=E(V i Q ij )-E(V i )E(Q ij ) In the formula, E is the mathematical expectation, Cov is the covariance, V i is the voltage amplitude of node i, P ij Inject active power into node i; Cov(V i ,P ij ) is the covariance of the voltage amplitude of node i and the active power injected into node i, Cov(V i ,Q ij ) is the covariance of the voltage amplitude of node i and the reactive power injected into node i, E(V i P ij ) is the mathematical expectation of the product of the voltage amplitude of node i and the active power injected into node i, E(V i ) is the mathematical expectation of the voltage amplitude at node i, E(P ij ) is the mathematical expectation of the active power injected into node i, E(V i Q ij ) is the mathematical expectation of the product of the voltage amplitude at node i and the reactive power injected at node i, E(Q ij ) is the mathematical expectation of the reactive power injected into node i.

4. The method according to claim 3, characterized in that The standard deviation formula is: In the formula, σ(V i ),σ(P ij ) and σ(Q ij ) are V i , P ij and Q ij The standard deviation of E(V i 2 ) is the mathematical expectation of the square of the voltage amplitude at node i, E 2 (V i ) is the square of the mathematical expectation of the voltage amplitude at node i, E(P ij 2 ) is the mathematical expectation of the square of active power injected into node i, E 2 (P ij ) is the square of the mathematical expectation of the active power injected into node i, E(Q ij 2 ) is the mathematical expectation of the square of reactive power injected into node i, E 2 (Q ij ) is the square of the mathematical expectation of the reactive power injected into node i.

5. The method according to claim 4, characterized in that The correlation index of active power and reactive power with voltage is constructed by dividing the covariance by the standard deviation using the following formula: In the formula, is the Pearson correlation coefficient between the voltage amplitude of node i and the active power injected into node i, is the Pearson correlation coefficient between the voltage amplitude at node i and the reactive power injected at node i.

6. The method according to claim 2, characterized in that Based on the characteristic analysis of network energy reactive items, the transient voltage stability problem is converted into the convergence and divergence problem of the amplitude trajectory curve of network energy reactive items. The maximum Lyapunov index is used to judge the transient voltage stability of the system, including: The network energy reactive term W Q The response trajectory is converted to W' Q -ΔW Q plane, by judging ΔW Q Whether the trajectory converges or not is used to evaluate the transient voltage stability of the system; where W' Q W Q Amplitude change rate, ΔW Q W Q The amplitude difference is calculated by the following formula to convert the network energy reactive term W Q The response trajectory is converted to W' Q -ΔW Q flat: Where W Qn and W Q(n-1) are W of the nth and n-1th time intervals respectively. Q Amplitude, W Q0 is the amplitude at the initial moment, ΔW Qn and ΔW Q(n-1) are W for the nth and n-1th time intervals respectively. Q Amplitude difference, W' Qn is the nth time interval W Q The amplitude change rate, Δt is the sampling time interval; Using the maximum Lyapunov exponent to determine ΔW Q Whether the amplitude trajectory curve converges is to determine the system transient voltage stability state. The formula is as follows: Where N is the initial time interval, k and n are both positive integers, kΔt represents a certain moment, λ(kΔt) is the maximum Lyapunov exponent at kΔt, ΔW Q(k+n) is the network energy reactive term W in the k+nth time interval Q Amplitude difference, ΔW Q(k+n-1) is the network energy reactive term W in the k+n-1th time interval Q Amplitude difference.

7. A device for determining transient voltage stability of a power system based on network energy, characterized in that: The device comprises: An energy function construction module is configured to obtain partial network response information, and integrate along the trajectory of the partial network response information to construct a network energy function; The reactive term extraction module is configured to construct correlation indicators of active power and reactive power with voltage based on the Pearson correlation coefficient, quantitatively analyze the correlation relationship between active power and reactive power and voltage, and extract the network energy reactive term as the voltage dominant factor; The stability judgment module is configured to convert the transient voltage stability problem into the convergence and divergence problem of the amplitude trajectory curve of the network energy reactive term based on the characteristic analysis of the network energy reactive term, and use the maximum Lyapunov exponent to judge the transient voltage stability of the system; The energy function building block is further configured as: Extracting partial network response information; wherein the partial network response information includes system node voltage, phase angle, and node injected active power and reactive power; By integrating the trajectory after the failure of the partial response information of the network, the network energy function of the system is obtained, which is expressed as: Where V i and V j are the voltage amplitudes of node i and node j respectively; P ij Inject active power into node i; Q ij and Q ji are the reactive power injected into nodes i and j respectively; l is the number of branches in the power system; W line is the network energy function; The network energy function consists of two parts: the network energy active term and the network energy reactive term. The network energy active term and the network energy reactive term are respectively expressed as: Where W P and W Q They are the network energy active term and network energy reactive term respectively. 8 . A non-transitory computer-readable storage medium storing instructions, which, when executed by a processor, executes the method according to claim 1 .

Images