Power distribution network voltage measurement-reactive power strategy relationship construction method and system based on dimensionality mapping

Abstract

The application relates to the technical field of power distribution network voltage control, and discloses a power distribution network voltage measurement-reactive power strategy relationship construction method and system based on dimension lifting mapping. The method trains a power distribution network voltage measurement-reactive power strategy relationship generation model based on dimension lifting mapping by using a training data set constructed by historical operation data of the power distribution network, performs high-dimensional mapping on the relationship between node power change and node voltage change, inputs real-time node voltage measurement data and node injected reactive power change into a power distribution network voltage measurement-reactive power strategy relationship matrix in a dimension lifting space, constructs a mapping relationship between measurement and control strategies, thereby uses a power distribution network voltage measurement-reactive power strategy mapping matrix to represent a power distribution network sensitivity relationship in a linear form, and realizes accurate depiction of the mapping relationship between power distribution network voltage measurement and each distributed power reactive power strategy in a weak parameter scene.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of distribution network voltage control, and particularly relates to a distribution network voltage measurement-reactive power strategy relationship construction method and system based on dimension lifting mapping. BACKGROUND

[0002] With the rapid development of renewable energy, the proportion of distributed power in modern power systems is increasing. However, the large-scale and high-proportion access of distributed power to distribution networks brings many challenges to the operation of distribution networks, such as voltage stability, power balance and protection problems. In order to effectively cope with these challenges, new methods and technologies need to be developed to improve the operation efficiency and stability of distribution networks.

[0003] Under this background, distribution network sensitivity analysis has become a key means to solve this problem. The sensitivity analysis of the distribution network is a method for quantitatively representing the relationship between the change of node power and the change of node voltage, which can be analyzed and extracted from historical data. The sensitivity relationship can support the rapid construction of distributed power control strategies. However, since the sensitivity relationship is highly related to the operation state of the distribution network, the sensitivity values are different under different distribution network states and power change amplitudes. This makes the sensitivity relationship of the distribution network present complex, high-dimensional and nonlinear characteristics, which is difficult to accurately describe with a single linear function. SUMMARY

[0004] The present application provides a distribution network voltage measurement-reactive power strategy relationship construction method and system based on dimension lifting mapping, which solves the technical problem that the sensitivity relationship of the distribution network presents complex, high-dimensional and nonlinear characteristics, which is difficult to accurately describe with a single linear function.

[0005] Therefore, the first aspect of the present application provides a distribution network voltage measurement-reactive power strategy relationship construction method based on dimension lifting mapping, comprising the following steps:

[0006] Constructing a training data set according to historical operation data of the distribution network, wherein the historical operation data of the distribution network includes node voltage of the distribution network and reactive power injected by the node of the distribution network;

[0007] Training a distribution network voltage measurement-reactive power strategy relationship generation model based on dimension lifting mapping using the training data set, and generating a distribution network voltage measurement-reactive power strategy relationship matrix;

[0008] Inputting real-time node voltage measurement data and the change amount of reactive power injected by the node into the distribution network voltage measurement-reactive power strategy relationship matrix to construct a voltage measurement-reactive power strategy mapping matrix of the distribution network.

[0009] Preferably, the method further comprises:

[0010] The training data set is constructed by taking the voltage of each node of the power distribution network and the reactive power injected by each node of the power distribution network as an input vector, and taking the voltage variation of each node of the power distribution network as an output vector.

[0011] Preferably, the step of training the power distribution network voltage measurement-reactive strategy relationship generation model based on the dimensionality increasing mapping to generate a power distribution network voltage measurement-reactive strategy relationship matrix comprises:

[0012] The input vector in the training data set is subjected to dimensionality increasing transformation to obtain an input vector extended by dimensionality increasing transformation;

[0013] Based on the input vector extended by dimensionality increasing transformation, the weight matrix of the power distribution network voltage measurement-reactive strategy relationship generation model is estimated by using the least square method, wherein the weight matrix is represented as:

[0014]

[0015] In the formula, M is the weight matrix, Y is the output vector set, X L represents the input vector set after dimensionality increasing, T is the transpose of the matrix, is the Moore-Penrose inverse matrix operator;

[0016] Based on the weight matrix, the linear mapping relationship between the input vector and the output vector in the training data set is determined to generate the power distribution network voltage measurement-reactive strategy relationship matrix, and the power distribution network voltage measurement-reactive strategy relationship matrix is:

[0017]

[0018] In the formula, y is the output vector, x l is the input vector after dimensionality increasing, and x is the input vector, is the dimensionality increasing representation of the input vector x.

[0019] Preferably, the step of subjecting the input vector in the training data set to dimensionality increasing transformation to obtain an input vector extended by dimensionality increasing transformation comprises:

[0020] The input vector in the training data set is subjected to dimensionality increasing transformation by a dimensionality increasing function to obtain an input vector extended by dimensionality increasing transformation, and the input vector extended by dimensionality increasing transformation is:

[0021]

[0022] In the formula, f l (·) represents a polyharmonic dimensionality increasing function, c irepresents the i-th dimension lifting basis vector, j represents the element index of the basis vector, w represents the total number of elements of the basis vector, c ij is the j-th element in the basis vector c i .

[0023] Preferably, the step of inputting the current acquired real-time node voltage measurement data and the change of the node injected reactive power into the power distribution network voltage measurement-reactive strategy relationship matrix to construct the power distribution network voltage measurement-reactive strategy mapping matrix comprises:

[0024] Based on the current acquired real-time node voltage measurement data and the change of the node injected reactive power, a new input vector is constructed, and the input vector is input into the power distribution network voltage measurement-reactive strategy relationship matrix to generate the elements of the power distribution network voltage measurement-reactive strategy mapping matrix, wherein the elements of the power distribution network voltage measurement-reactive strategy mapping matrix are:

[0025]

[0026] In the formula, represents the sensitivity of the change of the i-node reactive power ΔQ i,s to the change of the j-node voltage under the system voltage distribution at time t, ΔQ i,s represents the change of the reactive power of the s sampling points of node i in the voltage measurement-reactive strategy mapping matrix, N n represents the number of nodes, N s represents the number of sampling points of the change of the reactive power;

[0027] The elements of the power distribution network voltage measurement-reactive strategy mapping matrix are used to construct the power distribution network voltage measurement-reactive strategy mapping matrix, wherein the power distribution network voltage measurement-reactive strategy mapping matrix is:

[0028]

[0029] In the formula, represents the power distribution network voltage measurement-reactive strategy mapping matrix, represents the sensitivity of the change of the i-node reactive power ΔQ i,s to the change of the N n node voltage under the system voltage distribution at time t.

[0030] In a second aspect, the application further provides a power distribution network voltage measurement-reactive strategy relationship construction system based on dimension lifting mapping, comprising:

[0031] A training set construction module is configured to construct a training data set according to power distribution network historical operation data, wherein the power distribution network historical operation data comprises power distribution network node voltage and power distribution network node injected reactive power.

[0032] a training module configured to train a power grid voltage measurement-reactive power strategy relationship generation model based on a dimensionality increasing mapping using the training data set, and generate a power grid voltage measurement-reactive power strategy relationship matrix;

[0033] a matrix construction module configured to input real-time node voltage measurement data and a change in node injected reactive power into the power grid voltage measurement-reactive power strategy relationship matrix, and construct a power grid voltage measurement-reactive power strategy mapping matrix.

[0034] Preferably, the training set construction module is configured to construct the training data set by taking node voltages of the power grid and a change in node injected reactive power of the power grid as input vectors, and taking a change in node voltage of the power grid as an output vector.

[0035] Preferably, the training module specifically includes:

[0036] a dimensionality increasing transformation module configured to perform dimensionality increasing transformation on the input vectors in the training data set, and obtain input vectors extended by dimensionality increasing transformation;

[0037] a weight matrix estimation module configured to estimate a weight matrix of the power grid voltage measurement-reactive power strategy relationship generation model based on the input vectors extended by dimensionality increasing transformation using a least square method, wherein the weight matrix is represented as:

[0038]

[0039] wherein M is the weight matrix, Y is the output vector set, X L represents the input vector set after dimensionality increasing, T is a transpose of a matrix, is a Moore-Penrose inverse matrix operator;

[0040] a matrix generation module configured to determine a linear mapping relationship between the input vectors and the output vectors in the training data set based on the weight matrix, and generate a power grid voltage measurement-reactive power strategy relationship matrix, wherein the power grid voltage measurement-reactive power strategy relationship matrix is:

[0041]

[0042] wherein y is the output vector, x l is the input vector, and x is the input vector, is a dimensionality increased representation of the input vector x.

[0043] Preferably, the dimension lifting transformation module is specifically configured to perform dimension lifting transformation on the input vectors in the training data set by a dimension lifting function to obtain dimension lifting transformed input vectors, wherein the dimension lifting transformed input vectors are:

[0044]

[0045] wherein f l denotes a polyharmonic type dimension lifting function, c i denotes the i-th dimension lifting basis vector, j denotes the element index of the basis vector, w denotes the total number of elements of the basis vector, c ij is the j-th element in the basis vector c i .

[0046] Preferably, the matrix construction module specifically comprises:

[0047] an element generation module configured to construct a new input vector based on the currently acquired real-time node voltage measurement data and the change amount of the reactive power injected at the node, input the input vector into the power distribution network voltage measurement-reactive strategy relationship matrix, and generate an element of the power distribution network voltage measurement-reactive strategy mapping matrix, wherein the element of the power distribution network voltage measurement-reactive strategy mapping matrix is:

[0048]

[0049] wherein denotes the sensitivity of the change ΔQ i,s of the reactive power at the i-th node to the change of the voltage at the j-th node at the t-th time under the system voltage distribution, ΔQ i,s denotes the change amount of the reactive power at the s sampling points of the i-th node in the voltage measurement-reactive strategy mapping matrix, N n denotes the number of nodes, N s denotes the number of sampling points of the change amount of the reactive power.

[0050] a mapping matrix determination module configured to construct a power distribution network voltage measurement-reactive strategy mapping matrix by using the elements of the power distribution network voltage measurement-reactive strategy mapping matrix, wherein the power distribution network voltage measurement-reactive strategy mapping matrix is:

[0051]

[0052] wherein denotes the power distribution network voltage measurement-reactive strategy mapping matrix, denotes the sensitivity of the change ΔQ i,s of the reactive power at the i-th node to the change of the voltage at the N n -th node at the t-th time under the system voltage distribution.

[0053] From the above technical solution can be seen, the present application has the following advantages:

[0054] The present application trains the generation model of the voltage measurement-reactive power strategy relationship of the power distribution network based on the dimensional mapping by using the training data set constructed by the historical operation data of the power distribution network, maps the relationship between the node power change and the node voltage change in a high dimension, inputs the real-time node voltage measurement data and the reactive power change amount injected by the node into the voltage measurement-reactive power strategy relationship matrix of the power distribution network in the dimensional space, constructs the mapping relationship between the measurement and the control strategy, thereby using the voltage measurement-reactive power strategy mapping matrix of the power distribution network to represent the sensitivity relationship of the power distribution network in a linear form, and realizes the accurate description of the mapping relationship between the voltage measurement of the power distribution network and the reactive power strategy of each distributed power source in a weak parameter scene. BRIEF DESCRIPTION OF DRAWINGS

[0055] Figure 1 A flow chart of a power distribution network voltage measurement-reactive power strategy relationship construction method based on dimensional mapping provided for the embodiment of the present application is shown in the figure.

[0056] Figure 2 A structure schematic diagram of an IEEE 33 node power distribution system provided for the embodiment of the present application is shown in the figure.

[0057] Figure 3a A change curve schematic diagram of the per unit value and time of a load power generation profile provided for the embodiment of the present application is shown in the figure.

[0058] Figure 3b A change curve schematic diagram of the per unit value and time of a photovoltaic unit power generation profile provided for the embodiment of the present application is shown in the figure.

[0059] Figure 3c A change curve schematic diagram of the per unit value and time of a wind power unit power generation profile provided for the embodiment of the present application is shown in the figure.

[0060] Figure 4 A schematic diagram of the power change of different nodes provided for the embodiment of the present application is shown in the figure.

[0061] Figure 5 A prediction error distribution schematic diagram provided for the embodiment of the present application is shown in the figure.

[0062] Figure 6 A structure schematic diagram of a power distribution network voltage measurement-reactive power strategy relationship construction system based on dimensional mapping provided for the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0063] In the following, the technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all the other embodiments obtained by a person of ordinary skill in the art without creative effort belong to the scope of the present application.

[0064] It should be noted that the terms "comprising" and "having" and any variations thereof in the specification and claims of the present application and the above-described drawings are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to only those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to the process, method, product or device.

[0065] Sensitivity analysis of distribution network is a key means to solve this problem. Sensitivity analysis of distribution network is a method for quantitatively representing the relationship between node power change and node voltage change, which can be analyzed and extracted from historical data. The sensitivity relationship can support the rapid construction of distributed power control strategy. However, since the sensitivity relationship is highly related to the operation state of the distribution network, the sensitivity values are different under different distribution network states and power change amplitudes. This makes the sensitivity relationship of the distribution network complex, high-dimensional and nonlinear, which is difficult to accurately describe with a single linear function.

[0066] Therefore, the present application provides a distribution network voltage measurement-reactive strategy relationship construction method based on dimensionality mapping. The present application can be applied to the construction of the distribution network voltage measurement-reactive strategy relationship. The method can be executed by a distribution network voltage measurement-reactive strategy relationship construction device, which can be realized in the form of hardware and / or software and can be configured in a computer device.

[0067] Please refer to Figure 1 , Figure 1 The present application provides a distribution network voltage measurement-reactive strategy relationship construction method based on dimensionality mapping.

[0068] The present application provides a distribution network voltage measurement-reactive strategy relationship construction method based on dimensionality mapping, which includes the following steps S1-S3:

[0069] Step S1: Construct a training data set according to historical operation data of the distribution network, which includes node voltage of the distribution network and injected reactive power of the distribution network.

[0070] The training dataset is constructed using the voltage changes at each node of the distribution network and the reactive power changes injected at each node as input vectors, and the voltage changes at each node as output vectors.

[0071] The training dataset contains N f The group of training objects can be represented as:

[0072]

[0073]

[0074] In the formula, X represents the set of input vectors, Y represents the set of output vectors, and x (f) The input for the f-th group of training objects includes the voltages of each node in the system at time t. and the change in reactive power at each node at time t+1 y (f) The output of the f-th training group includes the voltage changes at each node.

[0075] The input vector and output vector can be represented as follows:

[0076]

[0077]

[0078] In the formula, x and y represent the input vector and the output vector, respectively, and T is the transpose of the matrix.

[0079] Step S2: Use the training dataset to train the distribution network voltage measurement-reactive power strategy relationship generation model based on the upgraded mapping, and generate the distribution network voltage measurement-reactive power strategy relationship matrix.

[0080] Specifically, step S2 includes:

[0081] Step S201: Perform a dimension-up transformation on the input vector in the training dataset to obtain the input vector expanded by the dimension-up transformation.

[0082] Specifically, the input vector in the training dataset is subjected to a dimension-up transformation using a dimension-up transformation function to obtain the dimension-up transformed and expanded input vector, which is:

[0083]

[0084] In the formula, f l (·) represents a polyharmonic-type function for increasing dimensionality, c i Let represent the i-th raised basis vector, j represent the element index of the basis vector, w represent the total number of elements in the basis vector, and c represent the total number of elements in the basis vector. ijthe jth element in the base vector c i .

[0085] where the basic mathematical structure of the dimension lifting function can be expressed as:

[0086] χ(x) = [χ1(x) χ2(x) … χ N (x)] T

[0087]

[0088] where χ i (x) represents the ith dimension lifting function, and χ(x) is the dimension lifting representation of the input vector x.

[0089] In step S202, the weight matrix of the power grid voltage measurement-reactive power strategy relationship generation model is estimated based on the input vector expanded by the dimension lifting transformation using the least square method, and the weight matrix is expressed as:

[0090]

[0091] where M is the weight matrix, Y is the output vector set, X L represents the dimension lifted input vector set, T is the transpose of the matrix, is the Moore-Penrose inverse matrix operator symbol;

[0092] In step S203, the linear mapping relationship between the input vector and the output vector in the training data set is determined based on the weight matrix, and the power grid voltage measurement-reactive power strategy relationship matrix is generated, and the power grid voltage measurement-reactive power strategy relationship matrix is:

[0093]

[0094] where y is the output vector, x l is the dimension lifted input vector, x is the input vector, and χ(x) is the dimension lifting representation of the input vector x.

[0095] In step S3, the real-time node voltage measurement data and the reactive power change amount injected by the node are input into the power grid voltage measurement-reactive power strategy relationship matrix to construct the voltage measurement-reactive power strategy mapping matrix of the power grid.

[0096] Specifically, step S3 specifically includes:

[0097] Step S301, based on the current acquired real-time node voltage measurement data and the change of the reactive power injected by the node, a new input vector is constructed, the input vector is input into the distribution network voltage measurement-reactive strategy relationship matrix, the elements of the distribution network voltage measurement-reactive strategy mapping matrix are generated, and the elements of the distribution network voltage measurement-reactive strategy mapping matrix are:

[0098]

[0099] In the formula, represents the change of the reactive power of the i node under the system voltage distribution at t time i,s sensitivity of j node voltage change at t time, ΔQ i,s represents the change of the reactive power of the s sampling points of the node i in the voltage measurement-reactive strategy mapping matrix, N n represents the number of nodes, N s represents the number of sampling points of the change of the reactive power.

[0100] Wherein, the elements of the distribution network voltage measurement-reactive strategy mapping matrix can describe the influence of the power change of each node of the distribution network on the voltage distribution under the current time running state, the matrix elements are represented as a three-dimensional matrix That is, the voltage measurement-reactive strategy mapping matrix of the distribution network is generated. Wherein,

[0101]

[0102]

[0103] Step S302, the elements of the voltage measurement-reactive strategy mapping matrix of the distribution network are used to constitute the voltage measurement-reactive strategy mapping matrix of the distribution network, and the voltage measurement-reactive strategy mapping matrix of the distribution network is:

[0104]

[0105] In the formula, represents the voltage measurement-reactive strategy mapping matrix of the distribution network, represents the change of the reactive power of the i node under the system voltage distribution at t time i,s sensitivity of N n node voltage change.

[0106] It should be noted that the training data set constructed by the historical operation data of the power distribution network is used to train the generation model of the power distribution network voltage measurement-reactive power strategy relationship based on dimensionality mapping, the relationship between the node power change and the node voltage change is high-dimensional mapping, the real-time node voltage measurement data and the reactive power change of the node injection are input into the power distribution network voltage measurement-reactive power strategy relationship matrix in the high-dimensional space, and the mapping relationship between the measurement and the control strategy is constructed, so that the voltage measurement-reactive power strategy mapping matrix of the power distribution network is used to represent the sensitivity relationship of the power distribution network in a linear form, and the mapping relationship between the voltage measurement of the power distribution network and the reactive power strategy of each distributed power supply in a weak parameter scene is accurately described.

[0107] The following is an example of a power distribution network voltage measurement-reactive power strategy relationship construction method based on dimensionality mapping provided by the embodiments of the present application.

[0108] In this example, the effectiveness of the power distribution network voltage measurement-reactive power strategy relationship construction method based on dimensionality mapping provided by the embodiments of the present application is verified by using an improved IEEE 33-node power distribution system.

[0109] The improved IEEE 33-node power distribution system includes one substation and 32 branches, and the rated voltage level is 12.66 kV. The structure of the IEEE 33-node power distribution system is as shown in Figure 2 .

[0110] The total active demand is 3715.0 kW, and the total reactive demand is 2300.0 kvar. In order to consider the influence of the distributed power supply, 7 photovoltaic units with a capacity of 200.0 kVA and 5 WT units with a capacity of 300.0 kVA are integrated in the test system. The real-time fluctuation curve of the power generation and the load is as shown in Figures 3a to 3c . Figure 3a , wherein, Figure 3b represents the change curve of the load power generation profile (Load profile) in per unit and time (time), Figure 3c represents the change curve of the photovoltaic unit power generation profile (PV profile) in per unit and time (time), represents the change curve of the wind turbine power generation profile (WT profile) in per unit and time (time). The upper limit and the lower limit of the legal voltage range are set to 1.10 p.u. and 0.90 p.u.

[0111] To support the accurate description of the nonlinear mapping relationship, a voltage measurement-reactive strategy mapping matrix is established based on the Koopman operator. From the perspective of a node, the voltage measurement-reactive strategy mapping matrix can accurately depict the relationship between the voltage measurement and the distributed power reactive strategy. Compared with the linear relationship derived at the operating point, the estimated mapping matrix has higher accuracy in describing this complex relationship. At the same time, this method can also avoid the interference caused by the perturbation method to the active distribution network.

[0112] The impact of different scenarios is shown in Figure 4 Figure 4 The power change of different nodes is shown, where Voltage-power relationship represents the voltage power relationship, Nodal reactive power variation represents the node reactive power variation, Estimated value represents the estimated value, True value represents the true value, Linear value represents the linear value, Error between linear / true value represents the error curve between the estimated value and the true value, Error between estimated / true value represents the error between the estimated value and the true value, and Prediction error represents the prediction error. In terms of adapting to complex distribution network environment, the Koopman-based model has strong generalization ability. The estimation error between the estimated value and the true value is mostly within 1%, which can adapt to the change of active distribution network state. The prediction error distribution is shown in Figure 5 Figure 5 where Number of sample represents the number of samples.

[0113] The numerical experiment is carried out on an Intel(R) Core(TM) i7-9750H CPU processor with a running frequency of 2.60 GHz and a memory of 16 GB.

[0114] As can be seen from the system test results, without accurate network parameters, the distribution network voltage measurement-reactive strategy relationship mapping matrix constructed by the dimensionality mapping can effectively extract the mapping rule between the distribution network voltage measurement and the reactive strategy of each distributed power from the historical data, thereby accurately depicting the mapping relationship between the voltage measurement and the reactive strategy of the distributed power, and accurately representing the sensitivity relationship of the distribution network.

[0115] The above is a detailed description of an embodiment of the distribution network voltage measurement-reactive strategy relationship construction method based on dimensionality mapping provided by the present application, and the following is a detailed description of an embodiment of a distribution network voltage measurement-reactive strategy relationship construction system based on dimensionality mapping provided by the present application.​​

[0116] As shown in Figure 6 , Figure 6 The application provides a structure of a power distribution network voltage measurement-reactive power strategy relationship construction system based on dimensionality mapping.

[0117] The application further provides a power distribution network voltage measurement-reactive power strategy relationship construction system based on dimensionality mapping, comprising:

[0118] A training set construction module 100 is configured to construct a training data set according to historical operation data of the power distribution network, wherein the historical operation data of the power distribution network comprises node voltages of the power distribution network and reactive power injected by nodes of the power distribution network.

[0119] The training set construction module 100 is configured to construct the training data set by taking the node voltages of the power distribution network and the reactive power injected by the nodes of the power distribution network as input vectors and taking the node voltage changes of the power distribution network as output vectors.

[0120] A training module 200 is configured to train a power distribution network voltage measurement-reactive power strategy relationship generation model based on dimensionality mapping by using the training data set, so as to generate a power distribution network voltage measurement-reactive power strategy relationship matrix.

[0121] A matrix construction module 300 is configured to input real-time node voltage measurement data and reactive power injected by nodes into the power distribution network voltage measurement-reactive power strategy relationship matrix, so as to construct a power distribution network voltage measurement-reactive power strategy mapping matrix.

[0122] In one specific embodiment, the training module specifically comprises:

[0123] A dimensionality transformation module is configured to perform dimensionality transformation on the input vectors in the training data set, so as to obtain input vectors extended by dimensionality transformation.

[0124] A weight matrix estimation module is configured to estimate a weight matrix of the power distribution network voltage measurement-reactive power strategy relationship generation model by using a least square method based on the input vectors extended by dimensionality transformation, wherein the weight matrix is represented as:

[0125]

[0126] In the formula, M represents the weight matrix, Y represents the output vector set, X represents the input vector set after dimensionality transformation, T represents the transpose of the matrix, and represents the Moore-Penrose inverse matrix operator. L

[0127] ​​The matrix generation module is configured to determine a linear mapping relationship between an input vector and an output vector in the training data set based on the weight matrix, and generate a power distribution network voltage measurement-reactive power strategy relationship matrix, wherein the power distribution network voltage measurement-reactive power strategy relationship matrix is:

[0128]

[0129] wherein y is the output vector, x is the input vector, and x is the dimensionally upgraded input vector. l is a dimensionally upgraded representation of the input vector x.

[0130] In one specific embodiment, the dimensionally upgrading module is specifically configured to perform dimensionally upgrading transformation on the input vector in the training data set through a dimensionally upgrading function to obtain an input vector upgraded by the dimensionally upgrading transformation, wherein the input vector upgraded by the dimensionally upgrading transformation is:

[0131]

[0132] wherein f l (·) represents a polyharmonic dimensionally upgrading function, c i represents the i-th dimensionally upgrading basis vector, j represents an element index of the basis vector, w represents a total number of elements of the basis vector, c ij is the j-th element in the basis vector c i .

[0133] In one specific embodiment, the matrix construction module specifically comprises:

[0134] The element generation module is configured to construct a new input vector based on the currently obtained real-time node voltage measurement data and the reactive power variation injected by the node, input the input vector into the power distribution network voltage measurement-reactive power strategy relationship matrix, and generate an element of the power distribution network voltage measurement-reactive power strategy mapping matrix, wherein the element of the power distribution network voltage measurement-reactive power strategy mapping matrix is:

[0135]

[0136] wherein represents a sensitivity of the j-th node voltage variation when the reactive power variation AQ i,s is injected by the i-th node under the system voltage distribution at the t-th moment, AQ i,s represents reactive power variation of s sampling points of the i-th node in the voltage measurement-reactive power strategy mapping matrix, N n represents a number of nodes, and N s represents a number of sampling points of the reactive power variation.

[0137] ​The mapping matrix determining module is configured to construct a voltage measurement-reactive power strategy mapping matrix of the power distribution network by using elements of the voltage measurement-reactive power strategy mapping matrix of the power distribution network, wherein the voltage measurement-reactive power strategy mapping matrix of the power distribution network is:

[0138]

[0139] In the formula, wherein, represents the voltage measurement-reactive power strategy mapping matrix of the power distribution network, represents the change of the reactive power of the i-th node under the voltage distribution of the system at the t-th moment i,s The sensitivity of the voltage of the N n th node to the change of the voltage of the i

[0140] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system described above can refer to the corresponding process in the foregoing method embodiments, and will not be described here.

[0141] In several embodiments provided in the present application, it can be understood that each block in the flowchart or block diagram can represent a module, a program segment or a part of code, and the module, the program segment or the part of code include one or more executable instructions for implementing the specified logic function. It should also be noted that, in some alternative implementation manners, the functions marked in the blocks can also occur in an order different from that marked in the drawings. For example, two consecutive blocks can actually be executed substantially in parallel, and they can also be executed in reverse order, depending on the functions involved.

[0142] In several embodiments provided in the present application, it should be understood that the disclosed system and method can be implemented in other ways. For example, the above-described device embodiments are only schematic, for example, the division of the units is only a logical function division, and actual implementation can be in another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interface, device or unit, and can be electrical, mechanical or in other forms.

[0143] The above embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for constructing a relationship between voltage measurement and reactive power strategy of a power distribution network based on dimensionality lifting mapping, characterized in that, The method comprises the following steps: constructing a training data set according to historical operation data of a power distribution network, wherein the historical operation data of the power distribution network comprises node voltage of the power distribution network and reactive power injected by a node of the power distribution network; training a power distribution network voltage measurement-reactive strategy relationship generation model based on dimensionality increasing mapping by using the training data set, to generate a power distribution network voltage measurement-reactive strategy relationship matrix; inputting real-time node voltage measurement data and a change amount of reactive power injected by a node into the power distribution network voltage measurement-reactive strategy relationship matrix, to construct a power distribution network voltage measurement-reactive strategy mapping matrix, comprising: constructing a new input vector based on the real-time node voltage measurement data and the change amount of reactive power injected by the node, inputting the input vector into the power distribution network voltage measurement-reactive strategy relationship matrix, to generate an element of the power distribution network voltage measurement-reactive strategy mapping matrix, wherein the element of the power distribution network voltage measurement-reactive strategy mapping matrix is: ； In the formula, represents the voltage measurement-reactive power strategy mapping matrix, at the moment of time under the system voltage distribution the reactive power change of the node, the sensitivity of the node voltage change, represents the reactive power change of the node of the sampling points in the voltage measurement-reactive power strategy mapping matrix, represents the number of nodes, represents the number of sampling points of the reactive power change. constructing a power distribution network voltage measurement-reactive strategy mapping matrix by using the element of the power distribution network voltage measurement-reactive strategy mapping matrix, wherein the power distribution network voltage measurement-reactive strategy mapping matrix is: ； wherein represents a voltage measurement-reactive power strategy mapping matrix of the power distribution network, represents at a time instant a change in reactive power at the node at a time instant a sensitivity of the node voltage change.

2. The method of claim 1, wherein the method is characterized by: further comprising: constructing the training data set by taking node voltage of the power distribution network and a change amount of reactive power injected by a node of the power distribution network as an input vector, and taking a change amount of node voltage of the power distribution network as an output vector.

3. The method of claim 1, wherein the method further comprises: The step of training the power distribution network voltage measurement-reactive strategy relationship generation model based on dimensionality increasing mapping by using the training data set, to generate a power distribution network voltage measurement-reactive strategy relationship matrix, specifically comprises: performing dimensionality increasing transformation on the input vector in the training data set, to obtain an input vector extended by dimensionality increasing transformation; estimating a weight matrix of the power distribution network voltage measurement-reactive strategy relationship generation model by using a least square method based on the input vector extended by dimensionality increasing transformation, wherein the weight matrix is represented as: ； wherein is a weight matrix, is a set of output vectors, denotes a set of input vectors of increased dimensionality, T is the transpose of a matrix, is a Moore-Penrose inverse matrix operator notation; determining a linear mapping relationship between the input vector and the output vector in the training data set based on the weight matrix, to generate a power distribution network voltage measurement-reactive strategy relationship matrix, wherein the power distribution network voltage measurement-reactive strategy relationship matrix is: ； where y is an output vector, is the input vector after dimensionality increase, is the input vector, is the input vector is the input vector after dimensionality increase.

4. The method of claim 3, wherein the method further comprises: The step of performing dimensionality increasing transformation on the input vector in the training data set, to obtain an input vector extended by dimensionality increasing transformation, specifically comprises: performing dimensionality increasing transformation on the input vector in the training data set by using a dimensionality increasing function, to obtain an input vector extended by dimensionality increasing transformation, wherein the input vector extended by dimensionality increasing transformation is: ； wherein, represents a polyharmonic type higher dimensional function, represents the th higher dimensional basis vector, represents the element index of the basis vector, represents the total number of elements of the basis vector, is the th element of the basis vector .

5. A power distribution network voltage measurement-reactive power strategy relationship construction system based on dimensionality mapping, characterized in that, comprising: a training set construction module, configured to construct a training data set according to historical operation data of a power distribution network, wherein the historical operation data of the power distribution network comprises node voltage of the power distribution network and reactive power injected by a node of the power distribution network; a training module, configured to train a power distribution network voltage measurement-reactive strategy relationship generation model based on dimensionality increasing mapping by using the training data set, to generate a power distribution network voltage measurement-reactive strategy relationship matrix; The matrix construction module is configured to input the current real-time node voltage measurement data and the change in the reactive power injected by the node into the power distribution network voltage measurement-reactive strategy relationship matrix to construct a voltage measurement-reactive strategy mapping matrix of the power distribution network. The matrix construction module specifically includes: The element generation module is configured to construct a new input vector based on the current real-time node voltage measurement data and the change in the reactive power injected by the node, input the input vector into the power distribution network voltage measurement-reactive strategy relationship matrix, and generate an element of the voltage measurement-reactive strategy mapping matrix of the power distribution network, where the element of the voltage measurement-reactive strategy mapping matrix of the power distribution network is: ； In the formula, represents the voltage of the node at the time The system voltage distribution at the time The reactive power change of the node The sensitivity of the node voltage change at the time represents the reactive power change of the node in the voltage measurement-reactive power strategy mapping matrix The reactive power change of the node The number of sampling points of the reactive power change represents the number of nodes represents the number of sampling points of the reactive power change The mapping matrix determination module is configured to construct the voltage measurement-reactive strategy mapping matrix of the power distribution network by using the element of the voltage measurement-reactive strategy mapping matrix of the power distribution network, where the voltage measurement-reactive strategy mapping matrix of the power distribution network is: ； wherein represents a voltage measurement-reactive power strategy mapping matrix of the power distribution network, represents the sensitivity of the node voltage change to the node reactive power change at the system voltage distribution at the time instant.

6. The power distribution network voltage measurement-reactive strategy relationship construction system based on dimension lifting mapping according to claim 5, characterized in that, The training set construction module is configured to construct the training data set by taking the voltage of each node of the power distribution network and the change in the reactive power injected by each node of the power distribution network as an input vector and taking the change in the voltage of each node of the power distribution network as an output vector.

7. The power distribution network voltage measurement-reactive strategy relationship construction system based on dimension lifting mapping according to claim 5, characterized in that, The training module specifically includes: The dimensionality increasing transformation module is configured to perform dimensionality increasing transformation on the input vector in the training data set to obtain an input vector extended by dimensionality increasing transformation. The weight matrix estimation module is configured to estimate a weight matrix of a power distribution network voltage measurement-reactive strategy relationship generation model based on the input vector extended by dimensionality increasing transformation by using a least square method, where the weight matrix is represented as: ； wherein is a weight matrix, is a set of output vectors, denotes a set of up-sampled input vectors, T is the transpose of a matrix, is a Moore-Penrose inverse matrix operator notation; The matrix generation module is configured to determine a linear mapping relationship between the input vector and the output vector in the training data set based on the weight matrix to generate a power distribution network voltage measurement-reactive strategy relationship matrix, where the power distribution network voltage measurement-reactive strategy relationship matrix is: ； where y is an output vector, is the input vector after dimensionality increase, is the input vector, is the input vector is the input vector after dimensionality increase.

8. The power distribution network voltage measurement-reactive strategy relationship construction system based on dimension lifting mapping according to claim 7, characterized in that, The dimensionality increasing transformation module is specifically configured to perform dimensionality increasing transformation on the input vector in the training data set by using a dimensionality increasing function to obtain an input vector extended by dimensionality increasing transformation, where the input vector extended by dimensionality increasing transformation is ； wherein, represents a polyharmonic type higher dimensional function, represents the th higher dimensional basis vector, represents the element index of the basis vector, represents the total number of elements of the basis vector, is the th element in the basis vector .

Images