A distribution network measurement optimization configuration method and terminal

Abstract

The present invention discloses a distribution network measurement optimization configuration method and terminal. The method uses the scalar function of the Fisher information matrix of multiple time sections as a parameter affecting the pheromone update in the first ant colony algorithm to compress the initial measurement set of the distribution network into a candidate measurement set; uses the state estimation accuracy value of multiple time sections as a parameter affecting the pheromone update in the second ant colony algorithm to solve the measurement configuration scheme with the best state estimation accuracy from the candidate measurement set. Therefore, during the second stage of calculation, the existence of the first stage greatly reduces the number of optional measurements when the ants walk, and saves calculation time while ensuring that the measurement configuration can obtain the optimal solution. At the same time, since the two-stage measurement optimization configuration of multiple time sections is taken into account, it can effectively solve the problem of insufficient observability of the previous active distribution network and improve the accuracy of solving the optimal measurement configuration scheme.

Description

Technical Field

[0001] The present invention relates to the technical field of power system distribution technology, and in particular to a distribution network measurement optimization configuration method and terminal. Background Art

[0002] The current active distribution network lacks observability, which greatly affects the distribution network's energy management capabilities to adapt to the access of large-scale distributed power sources; and most of the current active distribution networks are pseudo-measurements based on load forecasts, with poor accuracy. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a distribution network measurement optimization configuration method and terminal, which can improve the observability of the distribution network and improve the measurement configuration accuracy.

[0004] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0005] A method for optimizing the configuration of distribution network measurement includes the following steps:

[0006] The initial measurement set of the distribution network is updated with the scalar function of the Fisher information matrix of multiple time sections using the first ant colony algorithm, and the candidate measurement set is obtained by compression.

[0007] The candidate measurement set is subjected to a second ant colony algorithm pheromone update using state estimation accuracy values ​​of multiple time sections to solve a measurement configuration scheme with optimal state estimation accuracy.

[0008] In order to solve the above technical problems, another technical solution adopted by the present invention is:

[0009] A distribution network measurement optimization configuration terminal includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0010] The initial measurement set of the distribution network is updated with the scalar function of the Fisher information matrix of multiple time sections using the first ant colony algorithm, and the candidate measurement set is obtained by compression.

[0011] The candidate measurement set is subjected to a second ant colony algorithm pheromone update using state estimation accuracy values ​​of multiple time sections to solve a measurement configuration scheme with optimal state estimation accuracy.

[0012] The beneficial effects of the present invention are as follows: using the scalar function of the Fisher information matrix of multiple time sections as the parameter affecting the pheromone update in the first ant colony algorithm, the initial measurement set of the distribution network is compressed into a candidate measurement set. Therefore, by combining the ant colony algorithm with the optimal experimental design model based on FIM, the candidate measurement configuration scheme of the first stage is obtained, which greatly reduces the calculation time of the measurement optimization configuration. Using the state estimation accuracy value of multiple time sections as the parameter affecting the pheromone update in the second ant colony algorithm, the measurement configuration scheme with the best state estimation accuracy is solved from the candidate measurement set. Therefore, in the second stage calculation, due to the existence of the first stage, the number of optional measurements when the ants walk is greatly reduced, while ensuring that the measurement configuration can obtain the optimal solution, the calculation time is saved. At the same time, since the two-stage measurement optimization configuration of multiple time sections is taken into account, it can effectively solve the problem of insufficient observability of the previous active distribution network and improve the accuracy of solving the optimal measurement configuration scheme. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a flow chart of a distribution network measurement optimization configuration method according to an embodiment of the present invention;

[0014] Figure 2 A schematic diagram of a distribution network measurement optimization configuration terminal according to an embodiment of the present invention;

[0015] Figure 3 This is a schematic diagram of the two-stage measurement optimization configuration process according to the first embodiment of the present invention;

[0016] Figure 4 This is a flow chart of the first ant colony algorithm measurement optimization configuration according to the first embodiment of the present invention;

[0017] Figure 5 This is a flow chart of the second ant colony algorithm measurement optimization configuration according to the first embodiment of the present invention;

[0018] Figure 6 This is a calculation flow chart of the weighted least squares state estimation method according to the second embodiment of the present invention;

[0019] Description of labels:

[0020] 1. A distribution network measurement optimization configuration terminal; 2. A memory; 3. A processor. DETAILED DESCRIPTION

[0021] To illustrate the technical content, achieved objectives and effects of the present invention in detail, the following description is given in conjunction with the embodiments and accompanying drawings.

[0022] Please refer to Figure 1 The embodiment of the present invention provides a method for optimizing the configuration of distribution network measurement, including the steps of:

[0023] The initial measurement set of the distribution network is updated with the scalar function of the Fisher information matrix of multiple time sections using the first ant colony algorithm, and the candidate measurement set is obtained by compression.

[0024] The candidate measurement set is subjected to a second ant colony algorithm pheromone update using state estimation accuracy values ​​of multiple time sections to solve a measurement configuration scheme with optimal state estimation accuracy.

[0025] As can be seen from the above description, the beneficial effects of the present invention are as follows: using the scalar function of the Fisher information matrix of multiple time sections as the parameter affecting pheromone updates in the first ant colony algorithm, the initial measurement set of the distribution network is compressed into a candidate measurement set. Therefore, by combining the ant colony algorithm with the FIM-based optimal experimental design model, a candidate measurement configuration scheme for the first stage is obtained, greatly reducing the calculation time of the measurement optimization configuration. Using the state estimation accuracy value of multiple time sections as the parameter affecting pheromone updates in the second ant colony algorithm, the measurement configuration scheme with the optimal state estimation accuracy is solved from the candidate measurement set. Therefore, in the second stage calculation, the existence of the first stage greatly reduces the number of optional measurements when the ants walk, ensuring that the measurement configuration can obtain the optimal solution while saving calculation time. At the same time, because the two-stage measurement optimization configuration taking into account multiple time sections can effectively solve the problem of insufficient observability of the previous active distribution network and improve the accuracy of solving the optimal measurement configuration scheme.

[0026] Furthermore, the initial measurement set of the distribution network is updated with the pheromone of the first ant colony algorithm using the scalar function of the Fisher information matrix of multiple time sections, and the candidate measurement set obtained by compression includes:

[0027] Setting relevant parameters of the first ant colony algorithm, initializing pheromones of each path of the first ant colony algorithm, and obtaining an initial measurement set of the distribution network;

[0028] The scalar function of the Fisher information matrix is ​​cyclically used to calculate multiple time sections for each measurement point in the initial measurement set, and the pheromone of each path is updated after the optimal path is obtained. After the cycle reaches a preset number of times, a shrunken candidate measurement set is obtained.

[0029] As can be seen from the above description, the first-stage measurement optimization configuration solution is obtained by combining the ant colony algorithm with the FIM-based optimal experimental design model. Based on the calculation results of the first stage, the candidate measurement set is compressed. This step greatly reduces the calculation time of the measurement optimization configuration and improves computational efficiency. In addition, the measurement optimization configuration method considering multiple time sections can improve the energy management capabilities of the distribution network for large-scale distributed generation.

[0030] Furthermore, the scalar function of the Fisher information matrix is:

[0031]

[0032] Where z=(z1,…,z m ) T represents the m×1-dimensional measurement vector, x represents the n×1-dimensional state vector, i and j represent nodes i and j in the state vector, and l m represents the joint log-likelihood function, and E(·) represents the mean value.

[0033] From the above description, it can be seen that the scalar function based on the Fisher information matrix facilitates the compression of the initial measurement set.

[0034] Furthermore, the pheromone update of the second ant colony algorithm is performed on the candidate measurement set using the state estimation accuracy values ​​of multiple time sections to solve the measurement configuration scheme with the optimal state estimation accuracy, which includes:

[0035] Setting relevant parameters of the second ant colony algorithm, and initializing pheromones of each path of the second ant colony algorithm according to the candidate measurement set;

[0036] The state estimation accuracy value is cyclically used to calculate each time section of each measurement point in the candidate measurement set, and the pheromone of each path is updated after the optimal path is obtained. After the cycle reaches a preset number of times, a measurement configuration scheme with the optimal state estimation accuracy is obtained.

[0037] As can be seen from the above description, the first phase of calculation narrows the set of candidate measurements, significantly reducing the algorithm's complexity. In the second phase, the first phase significantly reduces the number of optional measurements during the ant's walk. This ensures that the measurement configuration achieves the optimal solution while saving computation time and improving efficiency.

[0038] Furthermore, the state estimation accuracy value is:

[0039]

[0040] Where V i ρ represents the measured value of the voltage amplitude of the ρ phase node at node i, represents the true solution of the voltage amplitude flow of the ρ-phase node at node i, represents the measured value of the ρ phase node voltage phase angle at node i, represents the true solution of the voltage phase angle power flow of the ρ phase node at node i, and ρ∈(A, B, C) represents the phase sequence;

[0041] The updating of the pheromones of each path after obtaining the optimal path includes:

[0042] Update the ant colony algorithm pheromone based on the state estimation accuracy value:

[0043]

[0044] Where Q2 represents the pheromone increase coefficient of the second ant colony algorithm, f2 represents the state estimation accuracy value obtained by the k-th ant selection measurement, and f2 = min(SE accuracy ).

[0045] From the above description, it can be seen that updating the ant colony algorithm pheromone by calculating the minimum state estimation accuracy value can effectively solve the shortcomings of poor accuracy of the previous pseudo-measurement based on load forecasting.

[0046] Please refer to Figure 2 Another embodiment of the present invention provides a distribution network measurement optimization configuration terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0047] The initial measurement set of the distribution network is updated with the scalar function of the Fisher information matrix of multiple time sections using the first ant colony algorithm, and the candidate measurement set is obtained by compression.

[0048] The candidate measurement set is subjected to a second ant colony algorithm pheromone update using state estimation accuracy values ​​of multiple time sections to solve a measurement configuration scheme with optimal state estimation accuracy.

[0049] As can be seen from the above description, the scalar function of the Fisher information matrix at multiple time intervals is used as the parameter affecting pheromone updates in the first ant colony algorithm, compressing the initial measurement set of the distribution network into a candidate measurement set. Therefore, by combining the ant colony algorithm with the FIM-based optimal experimental design model, the candidate measurement configuration scheme for the first stage is obtained, significantly reducing the computational time for the measurement optimization configuration. The state estimation accuracy values ​​at multiple time intervals are used as the parameter affecting pheromone updates in the second ant colony algorithm, and the measurement configuration scheme with the optimal state estimation accuracy is solved from the candidate measurement set. Therefore, in the second stage calculation, the existence of the first stage significantly reduces the number of optional measurements during the ant walk, ensuring that the measurement configuration can obtain the optimal solution while saving computational time. Furthermore, by considering the two-stage measurement optimization configuration at multiple time intervals, it can effectively address the previous problem of insufficient observability of active distribution networks and improve the accuracy of solving the optimal measurement configuration scheme.

[0050] Furthermore, the initial measurement set of the distribution network is updated with the pheromone of the first ant colony algorithm using the scalar function of the Fisher information matrix of multiple time sections, and the candidate measurement set obtained by compression includes:

[0051] Setting relevant parameters of the first ant colony algorithm, initializing pheromones of each path of the first ant colony algorithm, and obtaining an initial measurement set of the distribution network;

[0052] The scalar function of the Fisher information matrix is ​​cyclically used to calculate multiple time sections for each measurement point in the initial measurement set, and the pheromone of each path is updated after the optimal path is obtained. After the cycle reaches a preset number of times, a shrunken candidate measurement set is obtained.

[0053] As can be seen from the above description, the first-stage measurement optimization configuration solution is obtained by combining the ant colony algorithm with the FIM-based optimal experimental design model. Based on the calculation results of the first stage, the candidate measurement set is compressed. This step greatly reduces the calculation time of the measurement optimization configuration and improves computational efficiency. In addition, the measurement optimization configuration method considering multiple time sections can improve the energy management capabilities of the distribution network for large-scale distributed generation.

[0054] Furthermore, the scalar function of the Fisher information matrix is:

[0055]

[0056] Where z=(z1,...,z m ) T represents the m×1-dimensional measurement vector, x represents the n×1-dimensional state vector, i and j represent nodes i and j in the state vector, lm represents the joint log-likelihood function, and E(·) represents the average value.

[0057] From the above description, it can be seen that the scalar function based on the Fisher information matrix facilitates the compression of the initial measurement set.

[0058] Furthermore, the pheromone update of the second ant colony algorithm is performed on the candidate measurement set using the state estimation accuracy values ​​of multiple time sections to solve the measurement configuration scheme with the optimal state estimation accuracy, which includes:

[0059] Setting relevant parameters of the second ant colony algorithm, and initializing pheromones of each path of the second ant colony algorithm according to the candidate measurement set;

[0060] The state estimation accuracy value is cyclically used to calculate each time section of each measurement point in the candidate measurement set, and the pheromone of each path is updated after the optimal path is obtained. After the cycle reaches a preset number of times, a measurement configuration scheme with the optimal state estimation accuracy is obtained.

[0061] As can be seen from the above description, the first phase of calculation narrows the set of candidate measurements, significantly reducing the algorithm's complexity. In the second phase, the first phase significantly reduces the number of optional measurements during the ant's walk. This ensures that the measurement configuration achieves the optimal solution while saving computation time and improving efficiency.

[0062] Furthermore, the state estimation accuracy value is:

[0063]

[0064] Where V i ρ represents the measured value of the voltage amplitude of the ρ phase node at node i, represents the true solution of the voltage amplitude flow of the ρ-phase node at node i, represents the measured value of the ρ phase node voltage phase angle at node i, represents the true solution of the voltage phase angle power flow of the ρ phase node at node i, and ρ∈(A, B, C) represents the phase sequence;

[0065] The updating of the pheromones of each path after obtaining the optimal path includes:

[0066] Update the ant colony algorithm pheromone based on the state estimation accuracy value:

[0067]

[0068] Where Q2 represents the pheromone increase coefficient of the second ant colony algorithm, f2 represents the state estimation accuracy value obtained by the k-th ant selection measurement, and f2 = min(SE accuracy ).

[0069] From the above description, it can be seen that updating the ant colony algorithm pheromone by calculating the minimum state estimation accuracy value can effectively solve the shortcomings of poor accuracy of the previous pseudo-measurement based on load forecasting.

[0070] The distribution network measurement optimization configuration method and terminal described above are suitable for optimizing the measurement configuration of a distribution network. They can effectively address the shortcomings of previous active distribution networks, such as insufficient observability and poor pseudo-measurement accuracy based on load forecasting, and achieve optimal configuration of branch power measurement devices. The following is an explanation of the specific implementation methods:

[0071] Example 1

[0072] Please refer to Figure 1 , a distribution network measurement optimization configuration method, comprising the steps of:

[0073] S1. The initial measurement set of the distribution network is updated with the scalar function of the Fisher information matrix of multiple time sections using the first ant colony algorithm to compress and obtain a candidate measurement set.

[0074] In this embodiment, the first ant colony algorithm is the first stage ant colony algorithm; the initial measurement set is of order N×1, and after the first stage ant colony algorithm processing, the candidate measurement set is compressed to order n×1, where n <N。

[0075] Please refer to Figure 3 During measurement configuration, the measurements are divided into an initial measurement set containing M measurements and a candidate measurement set containing N measurements. To achieve observability of the active distribution network system, some measurements in the candidate measurement set are added to the initial measurement set. When adding measurements, investment cost constraints and state estimation accuracy requirements must be met, so measurement optimization configuration is required.

[0076] S11. Setting relevant parameters of a first ant colony algorithm, initializing pheromones of each path of the first ant colony algorithm, and obtaining an initial measurement set of the distribution network.

[0077] For details, please refer to Figure 4 , set the relevant parameters of the first stage ant colony algorithm, initialize the pheromone of each path, and after one traversal, evaluate the paths walked by all ants.

[0078] S12. cyclically use the scalar function of the Fisher information matrix to perform multi-time section calculations on each measurement point in the initial measurement set, and update the pheromone of each path after obtaining the optimal path. After the cycle reaches a preset number of times, a shrunken candidate measurement set is obtained.

[0079] In this embodiment, the number of measurement installations is limited by investment costs during the operation of the active power distribution network system. When state estimation is performed after configuring the same measurement model under different power flow distributions, the state estimation results will be affected by the power flow distribution at different times, which in turn affects the state estimation accuracy. Therefore, when optimizing measurement configuration, in order to meet the requirements of state estimation accuracy and comprehensive measurement benefits, the impact of power flow distribution changes on state estimation accuracy should be fully considered, and multiple time sections should be considered in the actual modeling.

[0080] Among them, the scalar function of the Fisher information matrix is:

[0081]

[0082] Where z=(z1,…,z m ) Trepresents the m×1-dimensional measurement vector, x represents the n×1-dimensional state vector, i and j represent nodes i and j in the state vector, and l m represents the joint log-likelihood function, and E(·) represents the mean value.

[0083] For details, please refer to Figure 4 , select the next measurement configuration point for each ant, and calculate the scalar function value f of each ant in n sections respectively 1,n , get pheromone feedback information min(f 1,1 +……+f 1,n ), if the current optimal solution is better than the previous one, then replace the current optimal configuration and update the path pheromone of each ant; otherwise, directly update the path pheromone of each ant. Repeat this process until the maximum number of traversals is reached, and then obtain the shrunken candidate measurement set based on the latest pheromone table.

[0084] S2. Perform a second ant colony algorithm pheromone update on the candidate measurement set using the state estimation accuracy values ​​of multiple time sections to solve a measurement configuration scheme with the best state estimation accuracy.

[0085] In this embodiment, the second ant colony algorithm is the second stage ant colony algorithm, please refer to Figure 3 ,In the second stage, in the n×1 order candidate measurement set, the ant colony algorithm is used to find the optimal measurement configuration scheme with the highest state estimation accuracy under the investment cost constraint.

[0086] S21 . Set relevant parameters of the second ant colony algorithm, and initialize pheromones of each path of the second ant colony algorithm according to the candidate measurement set.

[0087] For details, please refer to Figure 5 , set the relevant parameters of the second-stage ant colony algorithm, initialize the pheromones of each path in the second-stage ant colony algorithm; after one traversal, evaluate the paths walked by all ants.

[0088] S22, cyclically use the state estimation accuracy value to calculate each time section of each measurement point in the candidate measurement set, and update the pheromone of each path after obtaining the optimal path, until the measurement configuration scheme with the optimal state estimation accuracy is obtained after the preset number of cycles.

[0089] For details, please refer to Figure 5 , select the next measurement configuration point for each ant, and calculate the state estimation accuracy value f of each ant in n sections respectively 2,n , get pheromone feedback information min((f 2,1 +……+f 2，n) / n), if the current optimal solution is better than the previous generation, the current optimal configuration scheme is replaced and the path pheromone of each ant is updated. Otherwise, the path pheromone of each ant is directly updated. This process is repeated until the maximum number of traversals is reached, and the optimal measurement configuration scheme is obtained according to the latest pheromone table.

[0090] Specifically, in the calculation of the measurement optimization configuration scheme in the second stage, the weighted least squares state estimation method is used to model the system state estimation, and then the state estimation accuracy value is calculated:

[0091]

[0092] Where V i ρ represents the measured value of the voltage amplitude of the ρ phase node at node i, represents the true solution of the voltage amplitude flow of the ρ-phase node at node i, represents the measured value of the ρ phase node voltage phase angle at node i, The true solution of the voltage phase angle flow of the ρ phase node at node i is represented, and ρ∈(A, B, C) represents the phase sequence. In the subsequent measurement optimization configuration, the weighted least squares state estimation method is used as the basis to calculate the accuracy of the state estimation under different measurement configurations. The indicator measuring the accuracy of the state estimation is used as the standard to judge whether the measurement configuration scheme has reached the optimal standard, and the optimal SE is found under the investment cost constraint. accuracy Minimal measurement configuration.

[0093] The calculated state estimation accuracy value is used as the influencing parameter for pheromone update in the second stage ant colony algorithm, and the ant colony algorithm pheromone is updated based on the state estimation accuracy value:

[0094]

[0095] Where Q2 represents the pheromone increase coefficient of the second ant colony algorithm, f2 represents the state estimation accuracy value obtained by the k-th ant selection measurement, and f2 = min(SE accuracy ).

[0096] Configure the corresponding constraints for the two-stage measurement optimization described above:

[0097] 1. Relationship between state variables and measurement equations: z = h(x) + e;

[0098] Where z∈R m represents the measurement vector, h(x) represents the nonlinear measurement function that relates the measurement value to the state variable, x∈R 2n-1 Represents the state vector consisting of the node voltage amplitude and node voltage phase angle, e∈R mrepresents the measurement error vector, m represents the number of measurements, and n represents the number of nodes.

[0099] 2. Power constraints of measurement equation branches:

[0100]

[0101]

[0102]

[0103]

[0104] Where ρ, β, κ∈(A, B, C) represents the phase sequence, and are the measured active power and reactive power values ​​of the ρ-phase branch between node i and node j, respectively. and are the actual values ​​of the active power and reactive power of the ρ phase branch between node i and node j, and are the measurement errors of the active power and reactive power of the ρ-phase branch between node i and node j, respectively. i ρ is the ρ phase voltage amplitude at node i, is the voltage amplitude of phase κ at node j, where represents the phase angle difference between the ρ phase voltage at node i and the κ phase voltage at node i, The ρ phase voltage phase angle of node i is is the κ-phase voltage phase angle difference at node j, is the element in the node admittance matrix.

[0105] 3. Power constraints of measurement equation nodes:

[0106]

[0107]

[0108]

[0109]

[0110] Where, and They represent the measured value of active power and reactive power injected into the ρ phase of node i, respectively. i ρ and They represent the actual value of the injected active power and reactive power of the ρ phase at node i, respectively. and They represent the measurement errors of the ρ-phase node-injected active power and node-injected reactive power of node i, respectively.

[0111] 4. Branch current amplitude measurement:

[0112]

[0113]

[0114] Where, is the real part of the ρ-phase branch current between nodes i and j, is the imaginary part of the ρ-phase branch current between nodes i and j, is the measured value of the ρ-phase branch current amplitude between nodes i and j, is the actual value of the ρ-phase branch current amplitude between nodes i and j, is the measurement error corresponding to the square of the ρ-phase branch current amplitude measurement between node i and node j.

[0115] 5. Node voltage amplitude measurement value:

[0116]

[0117]

[0118] Where, is the measured value of the voltage amplitude of the ρ phase node at node i; is the measurement error corresponding to the voltage amplitude measurement of the ρ phase node at node i.

[0119] 6. Weighted Least Squares State Estimation Model:

[0120]

[0121] Where, is the covariance matrix of the measurement error, which is generally considered to be an unbiased, uncorrelated random variable that obeys a Gaussian distribution. is the measurement error e i The covariance of , m is the number of measurement equations.

[0122] 7. State estimation solution of the system under the measurement configuration scheme:

[0123]

[0124] Where x (k) represents the state variable value obtained at the kth iteration; H T (x (k) ) is the measurement Jacobian matrix that is updated as the state variables are updated during calculation; G(x(k) )=H T (x (k) )R -1 H(x (k) ) is the (2n-1)×(2n-1)-order symmetric positive definite gain matrix calculated during iteration.

[0125] 8. Conditions for ending the solution:

[0126] max|Δx (k) |≤∈;

[0127] Where ∈ represents the given convergence criterion.

[0128] 9. Gain matrix G(x):

[0129]

[0130] Where h(x) represents the nonlinear function that relates the quantity measurement to the state variable.

[0131] in, This formula represents the partial derivative of the point voltage amplitude measurement with respect to the node voltage amplitude and node voltage phase angle.

[0132] Therefore, in this embodiment, the ant colony algorithm is applied to the measurement optimization configuration problem. Each node that an ant walks while foraging corresponds to a measurement in the candidate measurement set. The ant's walking path represents a feasible measurement configuration solution for the measurement optimization configuration problem, and all paths of the entire ant colony constitute the solution space for the measurement optimization configuration problem. Taking the solution of optimal state estimation accuracy as an example, when state estimation accuracy is the optimization objective of the measurement optimization configuration model based on the ant colony algorithm, ants with lower state estimation accuracy release more pheromones. Over time, the accumulated pheromone concentration on the path with lower state estimation accuracy gradually increases, and the number of ants choosing this path will also increase. Ultimately, due to the positive feedback effect, all the ants will converge on the optimal path, and the optimal measurement configuration solution with the lowest state estimation accuracy is obtained.

[0133] Example 2

[0134] The difference between this embodiment and the first embodiment is that the specific steps of using the weighted least squares state estimation method to perform state estimation modeling on the system and thus calculate the state estimation accuracy value are further limited in the calculation of the second-stage measurement optimization configuration solution:

[0135] Please refer to Figure 6 , the weighted least squares state estimation needs to be solved iteratively to obtain the state estimation solution of the system under the measurement configuration scheme:

[0136]

[0137] Among them, x (k) Indicates the state variable value obtained at the kth iteration, H T (x (k) ) represents the measurement Jacobian matrix that is updated as the state variables are updated during calculation, G(x (k) )=H T (x (k) )R -1 H(x (k) ) represents the (2n-1)×(2n-1)-order symmetric positive definite gain matrix calculated during iteration.

[0138] When the model solution satisfies max|Δx (k) When |≤∈, the solution ends.

[0139] Where ∈ represents the given convergence criterion.

[0140] Through iterative calculation, the model gradually converges, and the gain matrix G(x) can be written as:

[0141]

[0142] Where h(x) represents the nonlinear function that relates the quantity measurement to the state variable.

[0143] The calculation process of the Jacobian matrix corresponding to the measurement equation is:

[0144]

[0145] The calculation expressions for each element in the Jacobian matrix are as follows:

[0146] (1) Partial derivatives of the node injected active power with respect to the node voltage amplitude and node voltage phase angle:

[0147]

[0148]

[0149] (2) Partial derivatives of node injected reactive power with respect to node voltage amplitude and node voltage phase angle:

[0150]

[0151]

[0152] (3) Partial derivatives of branch active power with respect to node voltage amplitude and node voltage phase angle:

[0153]

[0154]

[0155] (4) Partial derivatives of branch reactive power with respect to node voltage amplitude and node voltage phase angle:

[0156]

[0157]

[0158] (5) Partial derivatives of the square of the branch current amplitude measurement with respect to the node voltage amplitude and node voltage phase angle:

[0159]

[0160]

[0161] (6) Partial derivatives of node voltage amplitude measurement with respect to node voltage amplitude and node voltage phase angle:

[0162]

[0163]

[0164] The definition of the state estimation accuracy value is as follows:

[0165]

[0166] in, represents the true solution of the voltage amplitude power flow of phase ρ at node i; represents the true solution of the voltage phase angle flow of the ρ phase node at node i; ρ∈(A,B,C) represents the phase sequence.

[0167] When optimizing the measurement configuration in the future, the weighted least squares state estimation method is used as the basis to calculate the accuracy of the state estimation under different measurement configurations. The indicator of state estimation accuracy is used as the standard to judge whether the measurement configuration scheme has reached the optimal level, and the optimal value of SE is found under the investment cost limit. accuracy Minimal measurement configuration.

[0168] Example 3

[0169] Please refer to Figure 2 A distribution network measurement optimization configuration terminal 1 includes a memory 2, a processor 3, and a computer program stored in the memory 2 and executable on the processor 3. When the processor 3 executes the computer program, each step of a distribution network measurement optimization configuration method of embodiment one or two is implemented.

[0170] In summary, the present invention provides a distribution network measurement optimization configuration method and terminal, which is mainly aimed at the fact that the existing distribution network measurement optimization configuration method cannot perform measurement configuration based on a single-time flow section, and thus proposes a measurement optimization configuration method that considers multiple time sections. The distribution network is optimized using the ant colony algorithm. In the first stage, the scalar function of the FIM is used as the objective function value that affects the pheromone update in the ant colony algorithm to achieve compression of the candidate measurement set; in the second stage, the state estimation accuracy is used as the objective function value that affects the pheromone update in the second stage ant colony algorithm, and finally the optimal measurement configuration scheme that achieves the highest state estimation accuracy is solved. While ensuring the accuracy of state estimation, the measurement optimization configuration efficiency is effectively improved, and it has a higher comprehensive measurement configuration benefit. It can effectively solve the problems of insufficient observability of the previous active distribution network and poor pseudo-measurement accuracy based on load forecasting, improve the accuracy of solving the optimal measurement configuration scheme, and achieve the optimal configuration of the branch power measurement device.

[0171] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's description and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A distribution network measurement optimization configuration method, characterized in that: Including steps: The scalar function of the Fisher information matrix of multiple time sections is used to perform pheromone update of the first ant colony algorithm on the initial measurement set of the distribution network, and the candidate measurement set is compressed to obtain the following: relevant parameters of the first ant colony algorithm are set, the pheromone of each path of the first ant colony algorithm is initialized, and the initial measurement set of the distribution network is obtained; the scalar function of the Fisher information matrix is ​​cyclically used to perform multi-time section calculations on each measurement point in the initial measurement set, and the pheromone of each path is updated after the optimal path is obtained, until a preset number of cycles are completed, and a shrunken candidate measurement set is obtained; The candidate measurement set is updated with the pheromone of the second ant colony algorithm using the state estimation accuracy values ​​of multiple time sections to solve the measurement configuration scheme with the optimal state estimation accuracy: relevant parameters of the second ant colony algorithm are set, and the pheromone of each path of the second ant colony algorithm is initialized according to the candidate measurement set; the state estimation accuracy value is cyclically used to calculate each measurement point in the candidate measurement set at each time section, and the pheromone of each path is updated after the optimal path is obtained, until a preset number of cycles are completed, and the measurement configuration scheme with the optimal state estimation accuracy is obtained; The state estimation accuracy value is: ； Where, Represents the node i Phase node voltage amplitude measurement value, Represents the node i True solution of phase node voltage amplitude flow, Represents the node i Phase node voltage phase angle measurement value, Represents the node i True solution of phase node voltage phase angle power flow, Indicates phase sequence; The updating of the pheromones of each path after obtaining the optimal path includes: Update the ant colony algorithm pheromone based on the state estimation accuracy value: ； Where Q2 represents the pheromone increase coefficient of the second ant colony algorithm, f2 represents the state estimation accuracy value obtained by the k-th ant selection measurement, and f2=min(SE accuracy ).

2. A distribution network measurement optimization configuration method according to claim 1, characterized in that: The scalar function of the Fisher information matrix is: ； Where, express dimensional measurement vector, x represents dimensional state vector, i, j represents nodes i, j in the state vector, l m represents the joint log-likelihood function, Indicates the average value.

3. A distribution network measurement optimization configuration terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the following steps are implemented: The scalar function of the Fisher information matrix of multiple time sections is used to perform pheromone update of the first ant colony algorithm on the initial measurement set of the distribution network, and the candidate measurement set is compressed to obtain the following: relevant parameters of the first ant colony algorithm are set, the pheromone of each path of the first ant colony algorithm is initialized, and the initial measurement set of the distribution network is obtained; the scalar function of the Fisher information matrix is ​​cyclically used to perform multi-time section calculations on each measurement point in the initial measurement set, and the pheromone of each path is updated after the optimal path is obtained, until a preset number of cycles are completed, and a shrunken candidate measurement set is obtained; The candidate measurement set is updated with the pheromone of the second ant colony algorithm using the state estimation accuracy values ​​of multiple time sections to solve the measurement configuration scheme with the optimal state estimation accuracy: relevant parameters of the second ant colony algorithm are set, and the pheromone of each path of the second ant colony algorithm is initialized according to the candidate measurement set; the state estimation accuracy value is cyclically used to calculate each measurement point in the candidate measurement set at each time section, and the pheromone of each path is updated after the optimal path is obtained, until a preset number of cycles are completed, and the measurement configuration scheme with the optimal state estimation accuracy is obtained; The state estimation accuracy value is: ； Where, Represents the node i Phase node voltage amplitude measurement value, Represents the node i True solution of phase node voltage amplitude flow, Represents the node i Phase node voltage phase angle measurement value, Represents the node i True solution of phase node voltage phase angle power flow, Indicates phase sequence; The updating of the pheromones of each path after obtaining the optimal path includes: Update the ant colony algorithm pheromone based on the state estimation accuracy value: ； Where Q2 represents the pheromone increase coefficient of the second ant colony algorithm, f2 represents the state estimation accuracy value obtained by the k-th ant selection measurement, and f2=min(SE accuracy ).

4. A distribution network measurement optimization configuration terminal according to claim 3, characterized in that: The scalar function of the Fisher information matrix is: ； Where, express dimensional measurement vector, x represents dimensional state vector, i, j represents nodes i, j in the state vector, l m represents the joint log-likelihood function, Indicates the average value.

Images