An optimization configuration method for monitoring devices used in subsynchronous oscillation monitoring

Abstract

The present invention discloses a method for optimizing the configuration of monitoring devices for subsynchronous oscillation monitoring. The method for optimizing the configuration of monitoring devices includes: S1: obtaining the system parameters of the wind power system; S2: establishing the loop impedance matrix of the wind power system according to the system parameters; S3: calculating the observable index of the nodes of the wind power system according to the loop impedance matrix; S4: obtaining the damping influence index of the nodes of the wind power system; S5: using the entropy weight method to obtain the weights of the observable index and the damping influence index of the nodes of the wind power system according to the observable index and the damping influence index of the nodes of the wind power system; S6: obtaining the criticality of the nodes of the wind power system according to S5 and the probability of the occurrence of the working conditions; S7: constructing a multi-stage optimization configuration model of the monitoring devices for the criticality of the nodes of the wind power system according to the criticality of the nodes of the wind power system; S8: obtaining the optimized configuration result of the monitoring devices according to the multi-stage optimization configuration model of the monitoring devices.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and particularly to an optimization configuration method for a monitoring device used for subsynchronous oscillation monitoring. Background Art

[0002] With the rapid development of renewable energy power generation such as wind power, the proportion of renewable energy grid connection has been continuously increasing. While bringing the benefits of energy conservation and emission reduction, it also brings the risk of subsynchronous oscillation to the power system. There have been multiple subsynchronous oscillation accidents in domestic and foreign wind power systems, seriously threatening the safe and stable operation of the system. In order to monitor oscillations, study the oscillation propagation mechanism, suppress the spread of oscillations, and avoid the recurrence of oscillations, higher requirements are put forward for the observability and controllability of the power grid.

[0003] However, the existing wide-area measurement system is difficult to meet the current research needs for subsynchronous oscillations, and the traditional wide-area measurement system needs to be improved. However, due to time and cost limitations, it is unrealistic to install monitoring devices at all nodes of the wind power system in a short time. At the same time, the research on traditional optimization configuration methods mainly focuses on minimizing the number of monitoring devices on the premise of ensuring the global observability of the system, mostly based on the premise assumption that the key degrees of each node are the same. However, in actual engineering applications, the key degrees of each node for monitoring are different, and the traditional optimization configuration method is difficult to ensure that the monitoring devices are arranged on more critical nodes. In addition, considering that the system often has multiple working conditions where subsynchronous oscillation accidents may occur and the occurrence probabilities of each working condition are different, and that zero-injection nodes can effectively reduce the number of monitoring devices required, how to take into account the occurrence probability of system oscillation working conditions and the influence of zero-injection nodes to improve the economy of the system is an issue worthy of attention. Summary of the Invention

[0004] The purpose of the present invention is to provide an optimization configuration method for a monitoring device used for subsynchronous oscillation monitoring, which can set the monitoring device on more important nodes in the wind power system, and at the same time take into account the occurrence probability of system oscillation working conditions and the influence of zero-injection nodes to further improve the reliability and economy of the monitoring system.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] The present invention provides an optimization configuration method for a monitoring device used for subsynchronous oscillation monitoring, and the optimization configuration method for the monitoring device includes:

[0007] S1: Obtain the system parameters of the wind power system;

[0008] S2: Establish the loop impedance matrix of the wind power system according to the system parameters;

[0009] S3: Calculate the observable index of the nodes of the wind power system according to the loop impedance matrix;

[0010] S4: Obtain the damping influence index of the nodes of the wind power system;

[0011] S5: According to the observable index and the damping influence index of the nodes of the wind power system, use the entropy weight method to obtain the weight of the observable index and the weight of the damping influence index of the nodes of the wind power system;

[0012] S6: According to the observable index, the damping influence index, the weight of the observable index, the weight of the damping influence index and the probability of the working condition of the nodes of the wind power system, obtain the criticality of the nodes of the wind power system;

[0013] S7: According to the criticality of the nodes of the wind power system, construct a multi-stage optimal configuration model of the monitoring device for the criticality of the nodes of the wind power system;

[0014] S8: According to the multi-stage optimal configuration model of the monitoring device, obtain the optimal configuration result of the monitoring device.

[0015] Optionally, in the step S3, the observable index of the nodes of the wind power system is:

[0016]

[0017] where O n is the observable index of node n in the oscillation mode, O l is the relative magnitude of the oscillation current amplitude of branch l and H lp ' is the element in the l-th row and p-th column of matrix H'; H' = B'H; B is the loop-branch incidence matrix; H is the left eigenmatrix and Z(λm) represents the loop impedance matrix of the wind power system in the oscillation mode λm; λm represents the oscillation mode; Λ is the eigenmatrix; W is the right eigenmatrix; N L is the number of branches; Ω ln is the set of adjacent branches of node n.

[0018] Optionally, in the step S4, the damping influence index of the nodes of the wind power system is:

[0019]

[0020] where D j is the damping influence index of node j; h j is the damping sensitivity of node j; N B is the number of system nodes.

[0021] Optionally, the step S5 includes the following sub-steps:

[0022] S51: Standardize the observable index and the damping influence index to obtain the standardized result of the observable index and the standardized result of the damping influence index;

[0023] S52: Calculate the characteristic proportion of the observable index and the characteristic proportion of the damping influence index respectively according to the standardized result of the observable index and the standardized result of the damping influence index;

[0024] S53: Calculate the information entropy of the observable index and the information entropy of the damping influence index respectively according to the characteristic proportion of the observable index and the characteristic proportion of the damping influence index;

[0025] S54: Calculate the observable index weight and the damping influence index weight of the wind power system node respectively by using the information entropy of the observable index and the information entropy of the damping influence index.

[0026] Optionally, in the step S51, the standardization of the observable index and the damping influence index is as follows:

[0027]

[0028] where j = 1 to 2, when j = 1, it represents the observable index; when j = 2, it represents the damping influence index, y nj represents the j-th index of the n-th node after standardization processing, x nj represents the j-th index of the n-th node, x jmax is the maximum value of the j-th index, x jmin is the minimum value of the j-th index;

[0029] In the step S52, the characteristic proportions of the observable index and the damping influence index are as follows:

[0030]

[0031] where Y nj represents the characteristic proportion, j = 1 to 2, when j = 1, it represents the observable index; when j = 2, it represents the damping influence index, y nj represents the j-th index of the n-th node after standardization processing; N B is the number of system nodes;

[0032] In the step S53, the information entropy is as follows:

[0033]

[0034] where e jis the information entropy, j = 1 to 2, when j = 1, it represents the observable index, and when j = 2, it represents the damping influence index, Y nj is the characteristic proportion, N B is the number of system nodes;

[0035] In the step S54, the index weight is:

[0036]

[0037] where t is the number of indexes; W j is the weight of the j-th index, j = 1 to 2, when j = 1, it represents the observable index; when j = 2, it represents the damping influence index.

[0038] Optionally, the criticality of the wind power system node is:

[0039]

[0040] where C n represents the node criticality of the wind power system node n; c n =(W1O n +W2D n )Pσ, and c n represents the node critical value of the node n in the wind power system, that is, the node criticality C n is defined as the relative size of the node critical value c n , O n is the observable index of the oscillation mode node n, D n is the damping influence index of the node n, W1 is the weight of the oscillation observability index; W2 is the weight of the damping influence degree index; P is the probability of the working condition occurring; and P = P k P v ; P k is the disconnection probability of the line k; is the reliability of the line k; N L is the number of system lines, σ is the damping of this working condition; P v is the wind speed probability distribution and v is the wind speed, c and k are the scale parameter and shape parameter respectively; N B is the number of system nodes.

[0041] Optionally, in the step S7, the multi-stage optimal configuration model of the monitoring device for the criticality of the wind power system node includes: the full-cycle configuration model of the monitoring device and the sub-stage configuration model of the monitoring device.

[0042] Optionally, the full-cycle configuration model of the monitoring device is:

[0043]

[0044] Among them, N B is the number of system nodes, and C n represents the criticality of the wind power system node; x n is a variable between 0 and 1, indicating whether the monitoring device is configured for node n; Ω n is the set of adjacent nodes of node n; z m is a variable between 0 and 1, indicating whether node m is a non-zero injection node, and node m is a point in the set of adjacent nodes Ω n of node n; u nm is an auxiliary variable.

[0045] The configuration model of the monitoring device sub-stage is as follows:

[0046]

[0047] Among them, N B is the number of system nodes, and C n represents the criticality of the wind power system node; x n is a variable between 0 and 1, indicating whether the monitoring device is configured for node n; Ω n is the set of adjacent nodes of node n; z m is a variable between 0 and 1, indicating whether node m is a non-zero injection node; u nm is an auxiliary variable; b n is a variable between 0 and 1, indicating whether node n is observable; K represents the number of construction stages divided from the full cycle; g k represents the expected installation number in each stage.

[0048] The present invention has the following beneficial effects:

[0049] 1. Considering the criticality of system nodes, since the sub-synchronous oscillation caused by the large-scale access of new energy to the power system is an electrical oscillation caused by negative damping, the criticality of each node in the oscillation monitoring is defined and calculated from two perspectives of node damping influence and observable controllability. Compared with the traditional method, on the basis of considering the criticality of system nodes, the number of required monitoring devices is not increased, and the monitoring of critical nodes and the economy of configuration are taken into account;

[0050] 2. On this basis, an objective evaluation system for monitoring criticality considering the importance difference between indicators and the occurrence probability of oscillation conditions is established, and an SPMU multi-stage optimal configuration model considering node monitoring criticality and the occurrence probability of oscillation conditions is established. Compared with the full-cycle optimal configuration method, it is more in line with the actual situation of phased installation often existing in engineering applications. Description of the Drawings

[0051] Figure 1Flow chart of the monitoring device optimization configuration method provided for subsynchronous oscillation monitoring in the embodiments of the present invention;

[0052] Figure 2 For Figure 1 Sub - step flow chart of step S5 in;

[0053] Figure 3 Test system for Embodiment 2: ERCOT system topology in Texas, USA;

[0054] Figure 4 Test system for Embodiment 3: Improved New England 39 - bus system topology;

[0055] Figure 5 Comprehensive indicators of each node in the system for Embodiment 3. Specific implementation manners

[0056] The principles and features of the present invention will be described below with reference to the accompanying drawings. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0057] Embodiment 1

[0058] The technical solution of the present invention to solve the above - mentioned technical problems is as follows:

[0059] The present invention provides a monitoring device optimization configuration method for subsynchronous oscillation monitoring. Referring to Figure 1 as shown, the monitoring device optimization configuration method includes:

[0060] S1: Obtain the system parameters of the wind power system;

[0061] Here, the present invention does not specifically limit the system parameters of the wind power system. Those skilled in the art can judge and select the system parameters of the wind power system in combination with this application and actual requirements. In the present invention, the system parameters of the wind power system specifically include topological structure, wind power parameters, and line parameters. The following solutions will describe the present invention in combination with the topological structure, wind power parameters, and line parameters.

[0062] S2: Establish the loop impedance matrix of the wind power system according to the system parameters;

[0063] S3: Calculate the observable indicators of the nodes of the wind power system according to the loop impedance matrix;

[0064] The oscillation observability of a branch is defined as: oscillation mode λ mUnder the condition that a unit impulse disturbance voltage is applied to branch h, the relative current amplitude observed in branch l. The oscillation observability of a node is defined as the relative magnitude of the sum of the observability indices of the branches connected to the node, which characterizes the relative magnitude of the oscillation current amplitude of the branches that can be monitored by the monitoring device at this node. The oscillation observability of a branch can be expressed as:

[0065]

[0066] In the formula: N L is the number of branches; I l is the oscillation current amplitude of branch l; O l is the relative magnitude of the oscillation current amplitude of branch l;

[0067] By establishing the system loop impedance matrix Z, the relationship between the loop voltage and current can be established, and then the rapid calculation of the oscillation observability index can be realized. The relationship between the loop voltage and current can be expressed as:

[0068] Z -1 (s)V(s) = I(s)

[0069] In the formula: Z(s), V(s), and I(s) are the system loop impedance matrix, loop voltage vector, and loop current vector in the s domain, respectively.

[0070] The mode corresponding to the eigenvalue of Z being zero or very close to zero is the oscillation mode λ m = σ m ±jω m , σ m is the oscillation damping, and ω m is the oscillation angular frequency.

[0071] Due to the symmetry of the loop impedance matrix Z, it can be obtained that:

[0072] Z(λ m ) = HΛW

[0073] In the formula: Λ is the eigenmatrix; H and W are the left eigenmatrix and right eigenmatrix, respectively.

[0074] Furthermore, Z -1 (λ m ) can be rewritten as:

[0075]

[0076] In the formula: b is the number of independent loops; μ i is the i-th eigenvalue of matrix Z; H lp and W ph are the elements of the l-th row and p-th column, and the p-th row and h-th column of matrices H and W, respectively.

[0077] The relationship between the branch current and voltage can be obtained as follows:

[0078]

[0079] H' = B T H

[0080] W' = WB

[0081] Where: H′ lp and W′ ph are the elements of the l-th row and p-th column, and the p-th row and h-th column of the matrices H' and W' respectively. B is the loop-branch incidence matrix.

[0082] Optionally, in the step S3, the observable index of the wind power system node is:

[0083]

[0084] Where, O n is the observable index of branch l and node n in the oscillation mode, O l is the relative magnitude of the oscillation current amplitude of branch l and H lp ' is the element of the l-th row and p-th column of the matrix H'; H' = B'H; B is the loop-branch incidence matrix; H is the left eigenmatrix and Z(λm) represents the loop impedance matrix of the wind power system in the oscillation mode λm; λm represents the oscillation mode; Λ is the eigenmatrix; W is the right eigenmatrix; N L is the number of branches; Ω ln is the set of adjacent branches of node n.

[0085] S4: Obtain the damping influence index of the wind power system node;

[0086] Use the vector fitting technique to calculate the zeros of the system aggregated impedance to obtain the oscillation frequency and damping, and on this basis, obtain the node damping influence index

[0087] The stability of the system depends on the system aggregated impedance. By calculating the zeros of the aggregated impedance, the oscillation mode of the system can be obtained, including the oscillation damping σ m and the oscillation frequency f m .

[0088] Considering that in engineering applications, it is difficult to establish the impedance model of new energy units such as wind power, so it is regarded as a black box model, and the impedance of new energy units such as wind power is measured by injecting perturbations. Further, by finding a frequency-domain discrete function f(s) to match the measured impedance:

[0089]

[0090] The vector fitting method sets a group of initial poles a 0,k and constructs an auxiliary function σ(s):

[0091]

[0092]

[0093] The auxiliary function σ(s) can be understood as the fitting error, c 0,k represents the initial residue; s represents the complex frequency; N represents the fitting order; k = 1, 2,..., N; e, d represent real coefficients.

[0094] can be rewritten as:[[]] rewritten as:

[0095]

[0096] where: z k and z 0,k are the corresponding function zeros, z 0,k = eig(a - bc T ), where a, b, c can be solved using σ(s).

[0097] It can be seen from the above formula that after the poles on both sides of the equation are cancelled out, the zeros on the right side will become new poles. Repeating this process can obtain the poles a k of the transfer function f(s), and then d and e can be obtained by the least squares method, and the output impedance model of the black box model such as the wind turbine can be obtained.

[0098] On this basis, the system aggregated impedance model can be calculated according to the known information such as the system topology and line parameters. The real part of the zeros of the system aggregated impedance is the oscillation damping σ m ; the imaginary part of the zeros of the system aggregated impedance is the oscillation angular frequency ω m ; the oscillation frequency f m can be obtained from the oscillation angular frequency:

[0099]

[0100] The node damping sensitivity is defined as: the change in system damping caused by a 10% change in the generator impedance.

[0101] The node damping influence index is the ratio of the absolute value of the corresponding node damping sensitivity to the sum of the absolute values of all node damping sensitivities, that is:

[0102] In step S4, the damping influence index of the nodes of the wind power system is:

[0103]

[0104] where, Dn is the damping influence index of node n; h n is the damping sensitivity of node n; N B is the number of system nodes.

[0105] S5: According to the observable index and damping influence index of the wind power system nodes, using the entropy weight method, obtain the observable index weight and damping influence index weight of the wind power system nodes;

[0106] The entropy weight method is a commonly used objective weighting method that does not rely on human subjective judgment. The evaluation result has more mathematical basis and can objectively evaluate each index of the node monitoring criticality.

[0107] Thus, referring to Figure 2 as shown, the evaluation steps of the entropy weight method, that is, the sub-steps of step S5, are as follows:

[0108] S51: Standardize the observable index and the damping influence index to obtain the standardized result of the observable index and the standardized processing result of the damping influence index;

[0109] The standardization of the observable index and the damping influence index is:

[0110]

[0111] where j = 1 to 2, when j = 1, it represents the observable index; when j = 2, it represents the damping influence index, y nj represents the j-th index of the n-th node after standardization processing, x nj represents the j-th index of the n-th node, x jmax is the maximum value of the j-th index, x jmin is the minimum value of the j-th index.

[0112] S52: According to the standardized result of the observable index and the standardized processing result of the damping influence index, calculate the characteristic proportion of the observable index and the characteristic proportion of the damping influence index respectively;

[0113] In step S52, the characteristic proportions of the observable index and the damping influence index are:

[0114]

[0115] where, Y nj represents the characteristic proportion, j = 1 to 2, when j = 1, it represents the observable index; when j = 2, it represents the damping influence index, y nj represents the j-th index of the n-th node after standardization processing; N B is the number of system nodes.

[0116] S53: Calculate the information entropy of the observable index and the information entropy of the damping influence index respectively according to the characteristic proportion of the observable index and the characteristic proportion of the damping influence index;

[0117] The information entropy is:

[0118]

[0119] where, e j is the information entropy, j = 1 to 2 and when j = 1 it represents the observable index, when j = 2 it represents the damping influence index, Y nj is the characteristic proportion, N B is the number of system nodes.

[0120] S54: Calculate the weight of the observable index and the weight of the damping influence index of the wind power system node respectively by using the information entropy of the observable index and the information entropy of the damping influence index.

[0121] In the step S54, the index weight is:

[0122]

[0123] where, t is the number of indexes; W j is the weight of the j-th index, j = 1 to 2 and when j = 1 it represents the observable index; when j = 2 it represents the damping influence index.

[0124] S6: Obtain the criticality of the wind power system node according to the observable index, the damping influence index, the weight of the observable index, the weight of the damping influence index and the probability of the operating condition;

[0125] Determine the probability of the oscillation operating condition according to the system wind speed and the N-1 line disconnection probability.

[0126] The wind speed probability distribution can be expressed as:

[0127]

[0128] In the formula: P v is the wind speed probability distribution and v is the wind speed, c and k are the scale parameter and the shape parameter respectively.

[0129] Given the wind power resource characteristics (average wind speed v0, wind speed standard deviation r), the parameters c and k can be calculated by the following algorithm:

[0130]

[0131] To calculate the oscillation operating condition probability, the system N-1 line disconnection probability P k should also be determined:

[0132]

[0133] Wherein: P k is the line break probability of line k; R Lk is the reliability of line k; N L is the number of system lines, and σ is the damping of this working condition.

[0134] After obtaining the wind speed and the N-1 line break probability model, the occurrence probability P of a certain oscillation working condition of the system can be calculated:

[0135] P = P k P v

[0136] Thus, according to the node observable and controllable index, damping influence index, index weight, and working condition occurrence probability, the nodal criticality (Nodal Critical Index, NCI) of the system nodes is comprehensively evaluated.

[0137] The nodal criticality (Nodal Critical Index, NCI) of the system nodes can be determined by the following formula:

[0138]

[0139] Wherein, C n represents the nodal criticality of node n in the wind power system; c n =(W1O n +W2D n )Pσ, and c n represents the nodal critical value of node n in the wind power system, that is, the nodal criticality C n is defined as the relative magnitude of the nodal critical value c n , O n is the observable index of oscillation mode node n, D n is the damping influence index of node n, W1 is the weight of the oscillation observability index; W2 is the weight of the damping influence degree index; P is the working condition occurrence probability; and P = P k P v ; P k is the line break probability of line k; is the reliability of line k; N L is the number of system lines, and σ is the damping of this working condition; P v is the wind speed probability distribution and v is the wind speed, c and k are the scale parameter and shape parameter respectively; N B is the number of system nodes.

[0140] S7: Construct a multi-stage optimal configuration model for the monitoring device of the wind power system node criticality according to the wind power system node criticality;

[0141] In the full-cycle installation plan, it should be ensured that the monitoring device is finally installed on the node with higher monitoring criticality; when making the installation plan for each sub-stage, it should be ensured that the installation plan of this stage can make the observability of the monitoring system optimal while configuring the monitoring device on a more critical node. Therefore, for a system with N B nodes and N L lines, first, taking the node monitoring criticality index as the node weight and global observability as the constraint condition, establish a full-cycle configuration model for the monitoring device.

[0142] Optionally, in step S7, the multi-stage optimal configuration model for the monitoring device of the wind power system node criticality includes: a full-cycle configuration model for the monitoring device and a sub-stage configuration model for the monitoring device.

[0143] Optionally, the full-cycle configuration model for the monitoring device is:

[0144]

[0145] where N B is the number of system nodes, C n represents the wind power system node criticality; x n is a 0-1 variable indicating whether a monitoring device is configured at node n; Ω n is the set of adjacent nodes of node n; z m is a 0-1 variable indicating whether node m is a non-zero injection node and node m is a point in the set of adjacent nodes Ω n of node n; u nm is an auxiliary variable.

[0146] Considering that in actual engineering applications, the monitoring devices often cannot be configured in place at one time, there are generally multiple construction cycles, and there are expected installation numbers for each cycle. Therefore, the full cycle is divided into K construction stages, and the expected installation number for each stage is g k . Taking the maximum observability of the system in the construction stage (i.e., the largest number of observable nodes in the system) and the highest node monitoring criticality as the objectives, and taking the expected installation number of the monitoring device in the sub-stage as the constraint, establish a sub-stage optimal configuration model:

[0147]

[0148] where N B is the number of system nodes, C n represents the wind power system node criticality; x n is a 0-1 variable indicating whether a monitoring device is configured at node n; Ωn is the set of adjacent nodes of node n; z m is a binary variable indicating whether node m is a non-zero injection node; u nm is a auxiliary variable; b n is a binary variable indicating whether node n is observable; K represents the number of construction stages divided from the full cycle; g k represents the expected installation number in each stage.

[0149] S8: Obtain the optimized configuration result of the monitoring device according to the multi-stage optimized configuration model of the monitoring device.

[0150] Specifically in the present invention, the multi-stage optimized configuration scheme of the monitoring device is obtained by solving the model through integer linear programming.

[0151] The present invention has the following beneficial effects:

[0152] 1. Considering the criticality of system nodes, since the sub-synchronous oscillation caused by the large-scale access of new energy to the power system is an electrical oscillation caused by negative damping, the criticality of each node in oscillation monitoring is defined and calculated from two perspectives of node damping influence and observability and controllability. Compared with the traditional method, without increasing the number of monitoring devices required, the economy of monitoring and configuration of critical nodes is taken into account while considering the criticality of system nodes;

[0153] 2. On this basis, an objective evaluation system for monitoring criticality is established, taking into account the differences in the importance of indicators and the occurrence probability of oscillation conditions. A multi-stage optimized configuration model of SPMU considering node monitoring criticality and the occurrence probability of oscillation conditions is established. Compared with the full-cycle optimized configuration method, it is more in line with the actual situation of phased installation often existing in engineering applications.

[0154] Embodiment 2

[0155] The effectiveness of the proposed algorithm is verified by using the ERCOT system in Texas, USA. The topological structure of the system is as Figure 3 shown. The number of wind turbines in wind farms WF1 - WF5 is shown in Table 1. The wind speed v of each wind farm is 8 m / s, and the average wind speed v0 and wind speed standard deviation r are taken as 6.42812 and 3.04964 respectively. The compensation capacity of the fixed series compensator between nodes 5 and 6 is 25%, and the reliability of each line is shown in Table 2.

[0156] Table 1 Number of wind turbines in each wind farm

[0157]

[0158]

[0159] Table 2 Reliability of each line

[0160]

[0161] In the first step, by establishing the system loop impedance matrix Z, the node observability index is calculated.

[0162] In this simulation case, the damping and frequency corresponding to each line break situation are shown in Table 3.

[0163] Table 3 Damping and Frequency under Each Line Break Situation in the ERCOT System

[0164]

[0165] It can be found that only when the line L5-8 is broken will subsynchronous oscillation be triggered. Therefore, only the node indexes under the break of L5-8 need to be calculated. The node observability indexes are shown in Table 4.

[0166] Table 4 Observability Indexes

[0167]

[0168]

[0169] In the second step, the vector fitting technique is used to calculate the zeros of the system aggregated impedance to obtain the oscillation frequency and damping, and on this basis, the node damping influence index is obtained.

[0170] The calculation results of the node damping indexes are shown in Table 5.

[0171] Table 5 Node Damping Indexes

[0172]

[0173] In the third step, the entropy weight method is used to comprehensively evaluate the weights of the node observability indexes and damping influence indexes.

[0174] The weights of each index obtained by the entropy weight method are: W1 = 0.3995, W2 = 0.6005.

[0175] In the fourth step, the oscillation condition occurrence probability is determined according to the system wind speed and N-1 line break probability.

[0176] Given the average wind speed v0, wind speed standard deviation r, wind speed v, and line reliability, the oscillation condition occurrence probability can be obtained. The oscillation condition occurrence probability in this example is: 0.45626%

[0177] In the fifth step, the system node criticality (Nodal Critical Index, NCI) is comprehensively evaluated according to the node observable and controllable indexes, damping influence indexes, index weights, and condition occurrence probability.

[0178] The system node criticality obtained is shown in Table 6.

[0179] Table 6 Node Criticality

[0180]

[0181] Step 6: Considering the influence of zero-injection nodes, construct a multi-stage optimal configuration model of monitoring devices based on the criticality of system nodes.

[0182] Step 7: Solve the model through integer linear programming to obtain a multi-stage optimal configuration plan for monitoring devices. Compare it with the results obtained by the traditional algorithm without considering the criticality of system nodes. The comparison of the configuration results of the two algorithms is shown in Table 7.

[0183] Table 7 Configuration of Monitoring Devices

[0184]

[0185] Compared with the traditional method, the configuration method considering node criticality proposed in the present invention does not increase the number of required monitoring devices on the premise of considering the criticality of system nodes, so that the monitoring devices are arranged on more important nodes, taking into account both the monitoring of critical nodes and the economy of configuration.

[0186] Example 3

[0187] Use the improved New England 39-node system to verify the effectiveness of the proposed algorithm. The topological structure of the system is as Figure 4 shown. The system has 39 nodes and 46 branches. Wind farms WF1, WF2, WF3, WF4, WF5, WF6, WF7, WF8, and WF9 are respectively set at nine nodes of 4, 5, 7, 8, 16, 17, 18, 26, and 27. Series compensation capacitors are set on lines L4-5 and L17-27 respectively. Analyze the sub-synchronous oscillation risk of the system under typical working conditions shown in Table 8.

[0188] Table 8 Settings of Typical Working Conditions

[0189]

[0190]

[0191] In this example, the average wind speed v0 and the wind speed standard deviation r are respectively taken as 6.42812 and 3.04964. For the working conditions with sub-synchronous oscillation risk, the system oscillation observability index, damping influence degree index, and monitoring criticality index are obtained as Figure 5 shown. The reliability Rk of each line of the improved New England 39-node system is shown in Table 9.

[0192] Table 9 Reliability of Each Line of the System

[0193]

[0194]

[0195] Let the number of construction stages K = 3, and the expected installation numbers of SPMUs in each stage be m1 = m2 = m3 = 3 respectively. A multi-stage optimal configuration model of the monitoring device based on the criticality of the system nodes is constructed, and the optimal configuration scheme of the monitoring device is obtained by solving the model through integer linear programming. Comparing with the traditional multi-stage optimal configuration algorithm (considering zero-injection nodes and not considering node criticality), the results are shown in Table 10.

[0196] Table 10 SPMU configuration under different algorithms

[0197]

[0198] The configuration scheme of SPMU proposed by the present invention can make the node criticality of the SPMU configuration scheme higher, that is, it can ensure that the SPMU is configured on more critical nodes (such as node 27), and the required number of SPMUs is not increased. It can be seen that the proposed method ensures that the SPMU is configured on more critical nodes and makes the observable degree of the monitoring system in each stage the highest on the premise of ensuring the economy of the configuration scheme.

[0199] It can be seen that compared with the traditional method, the configuration method considering the criticality of system nodes proposed by the present invention does not increase the required number of monitoring devices on the basis of considering the criticality of system nodes, and takes into account the monitoring of critical nodes and the economy of configuration.

[0200] The above are only the preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An optimization configuration method for a monitoring device used for subsynchronous oscillation monitoring, characterized in that, The method for optimizing the configuration of the monitoring device includes: S1: Obtain the system parameters of the wind power system; S2: Establish the loop impedance matrix of the wind power system according to the system parameters; S3: Calculate the observable index of the nodes of the wind power system according to the loop impedance matrix; S4: Obtain the damping influence index of the nodes of the wind power system; S5: According to the observable index and the damping influence index of the nodes of the wind power system, use the entropy weight method to obtain the weight of the observable index and the weight of the damping influence index of the nodes of the wind power system; S6: According to the observable index, the damping influence index, the weight of the observable index, the weight of the damping influence index and the probability of the working condition of the nodes of the wind power system, obtain the criticality of the nodes of the wind power system; S7: According to the criticality of the nodes of the wind power system, construct a multi-stage optimization configuration model of the monitoring device for the criticality of the nodes of the wind power system; S8: According to the multi-stage optimization configuration model of the monitoring device, obtain the optimized configuration result of the monitoring device; In the step S3, the observable index of the nodes of the wind power system is: Among them, O n is the observable index of node n in the oscillation mode, O l is the relative magnitude of the oscillation current amplitude of branch l and H lp ' is the element in the l-th row and p-th column of matrix H'; H' = B'H; B is the loop-branch incidence matrix; H is the left eigenmatrix and Z(λm) represents the loop impedance matrix of the wind power system in the oscillation mode λm; λm represents the oscillation mode; Λ is the eigenmatrix; W is the right eigenmatrix; N L is the number of branches; Ω ln is the set of adjacent branches of node n; In the step S4, the damping influence index of the nodes of the wind power system is: Among them, D n is the damping influence index of node n; h n is the damping sensitivity of node n; N B is the number of system nodes; In the step S7, the multi-stage optimization configuration model of the monitoring device for the criticality of the nodes of the wind power system includes: a full-cycle configuration model of the monitoring device and a sub-stage configuration model of the monitoring device; The full-cycle configuration model of the monitoring device is: Among them, N B is the number of system nodes, and C n represents the criticality of the wind power system node; x n is a variable from 0 to 1, indicating whether a monitoring device is configured at node n; Ω n is the set of adjacent nodes of node n; z m is a variable from 0 to 1, indicating whether node m is a non-zero injection node and node m is a point in the set of adjacent nodes Ω n of node n; u nm is an auxiliary variable; The sub-stage configuration model of the monitoring device is: Among them, N B is the number of system nodes, and C n represents the criticality of the wind power system nodes; x n is a variable from 0 to 1, indicating whether a monitoring device is configured at node n; Ω n is the set of adjacent nodes of node n; z m is a variable from 0 to 1, indicating whether node m is a non-zero injection node; u nm is an auxiliary variable; b n is a variable from 0 to 1, indicating whether node n is observable; K represents the number of construction stages divided from the full cycle; g k represents the expected installation number in each stage.

2. The optimization configuration method for a monitoring device used for subsynchronous oscillation monitoring according to claim 1, characterized in that, The step S5 includes the following sub-steps: S51: Standardize the observable index and the damping influence index to obtain the standardized result of the observable index and the standardized processing result of the damping influence index; S52: Calculate the characteristic proportion of the observable index and the characteristic proportion of the damping influence index respectively according to the standardized result of the observable index and the standardized processing result of the damping influence index; S53: Calculate the information entropy of the observable index and the information entropy of the damping influence index respectively according to the characteristic proportion of the observable index and the characteristic proportion of the damping influence index; S54: Use the information entropy of the observable index and the information entropy of the damping influence index to calculate the weight of the observable index and the weight of the damping influence index of the nodes of the wind power system respectively.

3. The optimization configuration method for a monitoring device used for subsynchronous oscillation monitoring according to claim 2, characterized in that, In the step S51, the standardization of the observable index and the damping influence index is: Among them, j = 1 to 2. When j = 1, it represents the observable index; when j = 2, it represents the damping influence index, y nj represents the j-th index of the n-th node after standardization processing, x nj represents the j-th index of the n-th node, x jmax is the maximum value of the j-th index, x jmin is the minimum value of the j-th index; In the step S52, the characteristic proportion of the observable index and the damping influence index is: Among them, Y nj represents the characteristic proportion. When j = 1 to 2 and j = 1, it represents the observable index; when j = 2, it represents the damping influence index, and y nj represents the j-th index of the n-th node after standardization processing; N B is the number of system nodes; In the step S53, the information entropy is: Among them, e j is the information entropy, j = 1 to 2, when j = 1, it represents the observable index, and when j = 2, it represents the damping influence index, Y nj is the characteristic proportion, N B is the number of system nodes; In the step S54, the index weight is: where t is the number of indicators; W j is the weight of the j-th indicator, where j = 1 to 2, and when j = 1, it represents an observable indicator; when j = 2, it represents a damping influence indicator.

4. The method for optimizing the configuration of the monitoring device for subsynchronous oscillation monitoring according to claim 1, wherein, The criticality of the nodes of the wind power system is: Among them, C n represents the node criticality of node n in the wind power system; c n =(W1O n +W2D n )Pσ, and c n represents the node critical value of node n in the wind power system, that is, the node criticality C n is defined as the relative magnitude of the node critical value c n , O n is the observability index of oscillation mode node n, D n is the damping influence index of node n, W1 is the weight of the oscillation observability index; W2 is the weight of the damping influence degree index; P is the probability of the working condition occurrence; and P = P k P v ; P k is the disconnection probability of line k; is the reliability of line k; N L is the number of system lines, σ is the damping of this working condition; P v is the wind speed probability distribution and v is the wind speed, c, k are the scale parameter and shape parameter respectively; N B is the number of system nodes.

Images