A construction method, device and equipment of a voltage segment layered calculation architecture and a medium

Abstract

The application provides a construction method of a voltage segment layered calculation architecture, which comprises the following steps: collecting node voltage historical operation data and node active power historical operation data in a low-voltage distribution network and determining a daytime operation period and a nighttime operation period; obtaining a daytime training data set according to the daytime operation period and a nighttime training data set according to the nighttime operation period; constructing a basic layer and a regulation and control layer of the low-voltage distribution network in different operation periods according to the daytime training data set and the nighttime training data set; obtaining a basic operation voltage and a voltage change amount of the low-voltage distribution network in different periods, and calculating an actual voltage of the low-voltage distribution network in different periods according to the basic operation voltage and the voltage change amount. According to the technical scheme, the accuracy of voltage calculation without an electrical model of the low-voltage active distribution network can be improved.

Description

Technical Field

[0001] This invention relates to the field of low-voltage distribution network calculation technology, and in particular to a method, apparatus, electronic device, and storage medium for constructing a voltage segmented and hierarchical calculation architecture. Background Technology

[0002] Due to the complexity of network structures, insufficient coverage of measurement equipment, and the high cost of constructing accurate electrical models, low-voltage distribution networks often lack accurate topology and line parameter information, making traditional voltage calculation methods based on power flow calculations inapplicable. In recent years, with the widespread deployment of smart meters and the rapid development of data-driven technologies, voltage calculation methods independent of electrical models have been proposed. These methods use data-driven techniques such as least squares and machine learning to fit the mapping relationship between node power and node voltage in low-voltage distribution networks, thereby achieving voltage calculation. However, due to the mismatch between load demand and photovoltaic output in spatiotemporal distribution, bidirectional power flow often occurs in low-voltage distribution networks, leading to safety issues such as node voltage exceeding limits and three-phase imbalance, requiring voltage regulation. Ultimately, this results in poor accuracy of model-less voltage calculations in low-voltage active distribution networks under regulated conditions. Summary of the Invention

[0003] This invention aims to at least solve one of the technical problems existing in the prior art. To this end, this invention proposes a method, apparatus, device, and medium for constructing a voltage segmented and hierarchical calculation architecture, which can improve the accuracy of voltage calculation in low-voltage active distribution networks without electrical models.

[0004] In a first aspect, embodiments of the present invention provide a method for constructing a voltage segmented hierarchical computing architecture, including:

[0005] Summarize historical operating data of node voltage and node active power in the low-voltage distribution network;

[0006] Based on the historical operating data of the node voltage and the historical operating data of the node active power, the daytime operating period and the nighttime operating period are determined;

[0007] A daytime training dataset is obtained based on the daytime running period, and a nighttime training dataset is obtained based on the nighttime running period.

[0008] Based on the daytime training dataset, a daytime basic layer and a daytime control layer for the daytime operation period are constructed. The daytime basic layer is constructed based on the three-phase active net load matrix, the three-phase reactive net load matrix, and the input-output mapping relationship of the low-voltage distribution network.

[0009] Based on the nighttime training dataset, a nighttime base layer and a nighttime control layer are constructed for the nighttime operating period. The nighttime base layer is constructed based on the three-phase active net load matrix, the three-phase reactive net load matrix, and the input-output mapping relationship of the low-voltage distribution network.

[0010] The daytime base operating voltage and daytime voltage change are obtained, and the daytime base layer and the daytime control layer calculate the actual daytime voltage of the low-voltage distribution network based on the daytime base operating voltage and the daytime voltage change.

[0011] The nighttime base operating voltage and the nighttime voltage change are obtained, and the actual nighttime voltage of the low-voltage distribution network is calculated by the nighttime base layer and the nighttime control layer based on the nighttime base operating voltage and the nighttime voltage change.

[0012] In some embodiments of the present invention, determining the daytime and nighttime operating periods of the low-voltage distribution network includes:

[0013] Obtain the start and stop times of the preset photovoltaic nodes for the preset number of days in the low-voltage distribution network, as well as the total number of working days and the total number of photovoltaic nodes in the low-voltage distribution network;

[0014] The daytime start time of the low-voltage distribution network is calculated based on the start time, the total number of working days, and the total number of photovoltaic nodes;

[0015] The daytime shutdown time of the low-voltage distribution network is calculated based on the start time, the total number of working days, and the total number of photovoltaic nodes;

[0016] The nighttime operating hours of the low-voltage distribution network are determined based on the daytime start and end times.

[0017] In some embodiments of the present invention, the step of calculating the actual daytime voltage of the low-voltage distribution network based on the daytime base operating voltage and the daytime voltage change includes:

[0018] The three-phase voltage matrix of each node of the low-voltage distribution network during normal operation is obtained based on the three-phase net active load matrix, the three-phase net reactive load matrix and the input-output mapping relationship.

[0019] The daytime base operating voltage of the low-voltage distribution network is obtained based on the three-phase voltage matrix. The daytime control layer obtains the daytime voltage change of the low-voltage distribution network based on the three-phase active net load matrix, the three-phase reactive net load matrix and the input-output mapping relationship.

[0020] In some embodiments of the present invention, calculating the actual nighttime voltage of the low-voltage distribution network based on the nighttime base operating voltage and the nighttime voltage change includes:

[0021] The three-phase voltage matrix of each node of the low-voltage distribution network during normal nighttime operation is obtained based on the three-phase active net load matrix, the three-phase reactive net load matrix, and the input-output mapping relationship.

[0022] The nighttime base operating voltage of the low-voltage distribution network is obtained based on the three-phase voltage matrix. The nighttime control layer obtains the nighttime voltage change of the low-voltage distribution network based on the three-phase active net load matrix and the three-phase reactive net load matrix of the low-voltage distribution network and the input-output mapping relationship.

[0023] In some embodiments of the present invention, constructing the basic layer and control layer of the low-voltage distribution network for different operating periods includes:

[0024] A voltage calculation model for the daytime base layer is constructed based on the total number of photovoltaic nodes, the input-output mapping relationship, the three-phase active net load matrix, and the three-phase reactive net load matrix.

[0025] Calculate the reference operating voltage of the daytime base layer at the current daytime voltage calculation time, wherein the reference operating voltage represents the system state when no control measures are introduced into the daytime base layer;

[0026] The basic operating state of the low-voltage distribution network under the control of the control equipment within the daytime control layer is calculated. The reference operating voltage is superimposed with the daytime voltage change to obtain the actual daytime voltage.

[0027] In some embodiments of the present invention, obtaining a daytime training dataset based on the daytime runtime segment and obtaining a nighttime training dataset based on the nighttime runtime segment includes:

[0028] Training data for the control layer is obtained by subtracting the base layer data from different time periods.

[0029] Determine the time delay required for the training data of the daytime control layer and the nighttime control layer;

[0030] The training data of the daytime base layer and the nighttime base layer are subjected to time difference processing. The data at each time point is subtracted from the data of time delay to obtain the difference data, wherein the difference data is the three-phase active static load change data, the reactive static load change data and the three-phase voltage change data.

[0031] The differential data is used as the daytime training dataset and the nighttime training dataset.

[0032] In some embodiments of the present invention, after constructing the voltage calculation model of the base layer, the method further includes:

[0033] Obtain the three-phase voltage change matrix, three-phase active power difference matrix, three-phase reactive power difference matrix, three-phase energy storage device active power output matrix, and three-phase photovoltaic inverter reactive power output matrix of each node in the low-voltage distribution network.

[0034] Acquire the data with and without voltage regulation in the low-voltage distribution network;

[0035] The three-phase voltage change matrix of each node in the control layer is constructed based on the no-voltage control data, the voltage control data, the three-phase voltage change matrix, the three-phase active power difference matrix, the three-phase reactive power difference matrix, the active power output matrix of the three-phase energy storage device, and the reactive power output matrix of the three-phase photovoltaic inverter.

[0036] A segmented and hierarchical calculation architecture is constructed based on the three-phase active net load matrix, the three-phase reactive net load matrix, the three-phase active difference matrix, and the three-phase reactive difference matrix.

[0037] In a second aspect, embodiments of the present invention provide a voltage calculation device without an electrical model, including at least one control processor and a memory for communicatively connecting to the at least one control processor; the memory stores instructions executable by the at least one control processor, which, when executed by the at least one control processor, enables the at least one control processor to perform the method for constructing a voltage segmented hierarchical calculation architecture as described in the first aspect above.

[0038] Thirdly, embodiments of the present invention provide an electronic device including a voltage calculation device without an electrical model as described in the second aspect above.

[0039] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing computer-executable instructions for executing the method for constructing a voltage segmentation and hierarchical computing architecture as described in the first aspect above.

[0040] The method for constructing a voltage segmented hierarchical calculation architecture according to embodiments of the present invention has at least the following beneficial effects:

[0041] This process involves summarizing historical operating data of node voltage and active power within the low-voltage distribution network; determining daytime and nighttime operating periods based on these data; obtaining daytime training datasets based on the daytime operating periods and nighttime training datasets based on the nighttime operating periods; constructing a daytime basic layer and a daytime control layer for the daytime operating periods using the daytime training datasets. The daytime basic layer is constructed based on the three-phase active power net load matrix, the three-phase reactive power net load matrix, and the input-output mapping relationship of the low-voltage distribution network. The process also includes constructing a system based on the nighttime training datasets. The nighttime operation phase includes a nighttime base layer and a nighttime control layer. The nighttime base layer is constructed based on the three-phase active net load matrix, the three-phase reactive net load matrix, and the input-output mapping relationship of the low-voltage distribution network. It acquires the daytime base operating voltage and daytime voltage change, and calculates the actual daytime voltage of the low-voltage distribution network based on these values ​​using the daytime base layer and the daytime control layer. Similarly, it acquires the nighttime base operating voltage and nighttime voltage change, and calculates the actual nighttime voltage of the low-voltage distribution network based on these values ​​using the nighttime base layer and the nighttime control layer. According to the technical solution of this embodiment, the accuracy of voltage calculation in the electrical-undefined model of a low-voltage active distribution network can be improved. Attached Figure Description

[0042] Figure 1 This is a flowchart of a method for constructing a voltage segmented and hierarchical calculation architecture according to an embodiment of the present invention;

[0043] Figure 2 This is a flowchart of determining the daytime and nighttime operating periods of a low-voltage distribution network according to an embodiment of the present invention;

[0044] Figure 3 This is a flowchart of calculating the actual daytime voltage of a low-voltage distribution network based on the daytime base operating voltage and the daytime voltage change, provided by an embodiment of the present invention;

[0045] Figure 4 This is a flowchart of calculating the actual nighttime voltage of a low-voltage distribution network based on the nighttime base operating voltage and the nighttime voltage change, provided by one embodiment of the present invention.

[0046] Figure 5 This is a flowchart of a method for constructing the basic layer and control layer of a low-voltage distribution network at different operating stages, provided by an embodiment of the present invention;

[0047] Figure 6 This is a flowchart illustrating the daytime voltage change of a calculation and control device under a low-voltage distribution network, provided in one embodiment of the present invention.

[0048] Figure 7 This is a flowchart illustrating the voltage calculation model of the base layer provided in one embodiment of the present invention;

[0049] Figure 8 This is a structural diagram of a voltage calculation device without an electrical model provided in another embodiment of the present invention. Detailed Implementation

[0050] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0051] In the description of this invention, it should be understood that the orientation descriptions, such as up, down, front, back, left, right, etc., are based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting this invention.

[0052] In the description of this invention, "several" means one or more, "more than" means two or more, "greater than," "less than," and "exceeding" are understood to exclude the stated number, while "above," "below," and "within" are understood to include the stated number. The use of "first" and "second" in the description is merely for distinguishing technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of indicated technical features, or implicitly indicating the order of the indicated technical features.

[0053] In the description of this invention, unless otherwise explicitly defined, terms such as "set up," "install," and "connect" should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this invention in conjunction with the specific content of the technical solution.

[0054] This invention provides a method for constructing a segmented and hierarchical voltage calculation architecture, extending the existing model-free single-phase voltage calculation method to three-phase systems. It fully considers the three-phase imbalance characteristics commonly found in low-voltage distribution networks, thus more accurately reflecting the actual operating status of low-voltage distribution systems. Furthermore, this embodiment uses different operating data from daytime and nighttime periods in the low-voltage distribution network to train the model, taking into account the time-dependent characteristics of photovoltaic power generation. This establishes voltage calculation models for different time periods, and through a data-driven approach, learns the power flow characteristics of different low-voltage distribution networks at different times, thereby improving the accuracy and adaptability of the all-day voltage calculation for the low-voltage distribution network.

[0055] Furthermore, the voltage calculation method based on this embodiment decomposes the voltage calculation process at different time periods into the superposition of changes in the base layer voltage and the control layer voltage. By independently modeling the voltage and power mapping characteristics between different nodes of the base layer and the control layer, it effectively solves the problem of decreased voltage calculation accuracy in scenarios with frequent power fluctuations in the prior art, thereby improving the robustness of the voltage calculation model.

[0056] The control method of the present invention will be further described below with reference to the accompanying drawings.

[0057] Reference Figure 1 , Figure 1 The flowchart illustrates a method for constructing a segmented and hierarchical voltage calculation architecture according to an embodiment of the present invention. This method includes, but is not limited to, the following steps:

[0058] Step S11: Summarize the historical operating data of node voltage and node active power in the low-voltage distribution network;

[0059] It should be noted that historical operating data of node voltage directly reflects the changes in the voltage of each node in the distribution network over time. By summarizing this data, we can understand the characteristics of node voltage fluctuation range, amplitude distribution, etc., under different operating conditions. Historical operating data of node active power reflects the distribution and flow of power in the distribution network. Changes in active power will affect node voltage. Summarizing this data helps to analyze the intrinsic relationship between power and voltage, thus providing accurate data support for calculation methods based on the power-voltage relationship.

[0060] Step S12: Determine the daytime and nighttime operating periods based on the historical operating data of node voltage and the historical operating data of node active power.

[0061] It should be noted that in low-voltage distribution networks, photovoltaic (PV) power output exhibits significant time-varying characteristics, generating substantial power during the day and ceasing operation at night. Simultaneously, due to the relatively high impedance of low-voltage distribution networks, fluctuations in active power at nodes significantly impact node voltage. Therefore, the correlation between voltage and node power dynamically changes with time and operating conditions: during the daytime, the correlation between voltage and PV node active power is high, while the correlation with load node active power is low; at night, the correlation between voltage and load node active power significantly increases. This dynamic characteristic indicates that the impact of PV output on the voltage-power correlation is particularly prominent.

[0062] However, in existing technologies, the time-of-day characteristics of photovoltaic output are usually ignored, and a single model is used to train and calculate all data from all low-voltage distribution networks. This causes the model to focus excessively on the characteristics of daytime photovoltaic nodes during the training process, while ignoring the impact of nighttime load nodes on voltage. This makes it difficult to fully capture the dynamic characteristics of grid power changes, thus limiting the accuracy and applicability of voltage calculation.

[0063] Therefore, in this embodiment, the voltage calculation model is segmented according to the time-period characteristics of photovoltaic power output, and a targeted model is used for calculation in different time periods.

[0064] Step S13: Obtain the daytime training dataset based on the daytime running time and the nighttime training dataset based on the nighttime running time.

[0065] It should be noted that there are significant differences in power load characteristics between daytime and nighttime. During the day, industrial, commercial, and residential electrical appliances are used frequently, resulting in diverse and highly volatile loads. In contrast, nighttime loads are relatively smaller and more stable, mainly consisting of residential lighting and small appliances. By obtaining separate training datasets for daytime and nighttime periods, the impact of load changes on low-voltage distribution network voltage can be more accurately reflected.

[0066] Step S14: Based on the daytime training dataset and the nighttime training dataset, construct the basic layer and the control layer for different operating periods of the low-voltage distribution network. The basic layer is constructed based on the three-phase active net load matrix, the three-phase reactive net load matrix and the input-output mapping relationship of the low-voltage distribution network.

[0067] It should be noted that the three-phase active power net load matrix and the three-phase reactive power net load matrix can comprehensively and meticulously describe the power distribution of each phase in a low-voltage distribution network. Active power determines the rate at which electrical energy is converted into other forms of energy, while reactive power is related to the energy exchange between electric and magnetic fields, directly affecting voltage quality and grid losses. Through the three-phase active power net load matrix and the three-phase reactive power net load matrix, the power flow status of each node and line in the low-voltage distribution network during daytime operation can be accurately grasped, providing accurate basic data for subsequent analysis and calculation. The input-output mapping relationship correlates the power matrix with other operating parameters of the low-voltage distribution network (such as voltage and current), thereby reflecting the dynamic changes in power input and output during daytime operation. Based on the three-phase active power net load matrix, the three-phase reactive power net load matrix, and the input-output mapping relationship, the daytime basic layer, daytime control layer, nighttime basic layer, and nighttime control layer of the low-voltage distribution network can be established.

[0068] Step S15: Based on the night training dataset, construct the night base layer and the night control layer for the night operation period. The night base layer is constructed based on the three-phase active net load matrix, the three-phase reactive net load matrix and the input-output mapping relationship of the low-voltage distribution network.

[0069] Step S16: Obtain the daytime base operating voltage and daytime voltage change, and calculate the actual daytime voltage of the low-voltage distribution network based on the daytime base layer and daytime control layer.

[0070] Step S17: Obtain the nighttime base operating voltage and the nighttime voltage change, and calculate the actual nighttime voltage of the low-voltage distribution network based on the nighttime base layer and the nighttime control layer.

[0071] It should be noted that, firstly, the reference normal operating voltage of the daytime base layer is calculated. This reference normal operating voltage reflects the system state of the low-voltage distribution network when no control measures are introduced. Subsequently, in the daytime and nighttime control layers, the voltage change of the low-voltage distribution network under the action of control equipment is calculated. Finally, the reference voltage of the daytime base layer is superimposed with the daytime voltage change of the daytime control layer, and the reference voltage of the nighttime base layer is superimposed with the nighttime voltage change of the nighttime control layer, thus obtaining the actual daytime voltage and the actual nighttime voltage of the low-voltage distribution network.

[0072] It should be noted that this embodiment extends the model-free single-phase voltage calculation method in the prior art to three-phase systems, fully considering the three-phase imbalance characteristics commonly found in low-voltage distribution networks, thereby more accurately reflecting the actual operating status of low-voltage distribution systems. Furthermore, this embodiment is used to train different time-period operating data in the low-voltage distribution network during the day and night periods, respectively, to establish voltage calculation models for different time periods, taking into account the time-period characteristics of photovoltaic power generation. By using a data-driven method to learn the power flow characteristics of different low-voltage distribution networks at different times, the accuracy and adaptability of the all-day voltage calculation for the low-voltage distribution network are improved.

[0073] Additionally, in one embodiment, reference is made to Figure 2 ,exist Figure 1 Step S11 in the illustrated embodiment also includes, but is not limited to, the following steps:

[0074] Step S21: Obtain the start and stop times of photovoltaic nodes of the low-voltage distribution network for a preset number of days, as well as the total number of working days and the total number of photovoltaic nodes of the low-voltage distribution network;

[0075] Step S22: Calculate the daytime start time of the low-voltage distribution network based on the start time of power output, the total number of working days, and the total number of photovoltaic nodes;

[0076] Step S23: Calculate the daytime shutdown time of the low-voltage distribution network based on the start time of power output, the total number of working days, and the total number of photovoltaic nodes;

[0077] Step S24: Determine the nighttime working hours of the low-voltage distribution network based on the daytime start and end times of the working hours.

[0078] It should be noted that the voltage segmentation rules set in this embodiment are as follows:

[0079] Historical node voltage operation data and historical node active power operation data are obtained from the node voltage historical operation data and the node active power historical operation data, respectively. The times when all photovoltaic power generation begins and stops each day are recorded in both the node voltage historical operation data and the node active power historical operation data. The average of the two sets is taken and rounded to the nearest integer as the start and end times of the daytime period. Time outside the daytime period is defined as the nighttime period. This is expressed by the following first formula:

[0080]

[0081] in, This is the time to begin exerting effort during the day. D represents the time when power output stops during the day; D represents the total number of working days; M represents the total number of photovoltaic nodes in the low-voltage active distribution network. This represents the start time of power output for the m-th photovoltaic node at the m-th node on day d. The moment when the m-th photovoltaic node stops outputting power on day d; This is the floor symbol.

[0082] Additionally, in one embodiment, reference is made to Figure 2 ,exist Figure 1 Step S11 in the illustrated embodiment also includes, but is not limited to, the following steps:

[0083] Step S31: Obtain the three-phase voltage matrix of each node in the low-voltage distribution network during normal operation based on the three-phase active net load matrix, the three-phase reactive net load matrix, and the input-output mapping relationship;

[0084] Step S32: Obtain the daytime basic operating voltage of the low-voltage distribution network based on the three-phase voltage matrix. The daytime control layer obtains the daytime voltage change of the low-voltage distribution network based on the three-phase active net load matrix, the three-phase reactive net load matrix, and the input-output mapping relationship.

[0085] It should be noted that, due to the mutual influence of self-impedance and mutual impedance among the three phases in low-voltage distribution lines, this embodiment establishes a three-phase Distflow power flow model, expressed by the following second formula:

[0086] ;

[0087] in, Three-phase net active load matrix on branch mn Let mn be the reactive power matrix on branch mn; S n Let n be the set of child nodes; Let n be the square matrix of the effective values ​​of the three-phase voltages at node n; z mn and i mn These are the three-phase impedance matrix and the square matrix of the effective values ​​of the three-phase currents for branch mn, respectively. P PV,n Let n be the net active power load matrix of the photovoltaic three-phase system at node n. P L,n and Q L,n These are the three-phase active power net load matrix and the three-phase reactive power net load matrix for node n, respectively; * indicates the conjugate of the matrix. The three-phase matrix has the following form: .

[0088] Furthermore, z mn , r mn and x mn This can be expressed by the following third formula:

[0089] ;

[0090] in, The self-impedance of phase A is given. ; α represents the mutual impedance between phase A and phase B; α is the symmetric component transformation matrix.

[0091] Additionally, in one embodiment, reference is made to Figure 2 ,exist Figure 1 Step S11 in the illustrated embodiment also includes, but is not limited to, the following steps:

[0092] Step S41: Obtain the three-phase voltage matrix of each node of the low-voltage distribution network during normal operation at night based on the three-phase active net load matrix, the three-phase reactive net load matrix and the input-output mapping relationship;

[0093] Step S42: Obtain the nighttime basic operating voltage of the low-voltage distribution network based on the three-phase voltage matrix. The nighttime control layer obtains the nighttime voltage change of the low-voltage distribution network based on the three-phase voltage active net load matrix, the three-phase reactive net load matrix, and the input-output mapping relationship of the low-voltage distribution network.

[0094] It should be noted that the specific calculation process of the nighttime base layer and the nighttime control layer is the same as that of the daytime base layer and the daytime control layer. Therefore, this embodiment will not elaborate on the specific calculation process.

[0095] Additionally, in one embodiment, reference is made to Figure 2 ,exist Figure 1 Step S11 in the illustrated embodiment also includes, but is not limited to, the following steps:

[0096] Step S51: Construct a voltage calculation model for the base layer based on the total number of photovoltaic nodes, input-output mapping relationship, three-phase active net load matrix, and three-phase reactive net load matrix;

[0097] Step S52: Calculate the reference operating voltage of the daytime base layer at the current daytime voltage calculation time, wherein the reference operating voltage characterizes the system state when no control measures are introduced at the daytime base layer;

[0098] Step S53: Calculate the basic operating state of the low-voltage distribution network under the control of the control equipment within the control layer, and superimpose the reference operating voltage and the voltage change to obtain the actual voltage.

[0099] It should be noted that β can be any one of the three phases A, B, and C. An implicit functional relationship exists between the β-phase voltage at node n and the β-phase voltage at node m, the three-phase active power injected at node n, the three-phase reactive power, and the three-phase line loss of the mn line. Furthermore, the β-phase voltage at node m and the three-phase line loss of the mn line are also affected by the active and reactive power of other nodes. Therefore, this embodiment uses historical LVADN operating data as the training set, including three-phase active load, reactive load, PV output, and corresponding node voltages. A data-driven approach is used to mine the potential power flow information in this data and establish mapping relationships, thereby replacing the traditional electrical model and enabling voltage calculation without network parameters.

[0100] In this embodiment, the voltage calculation model of the base layer is expressed by the following fourth formula:

[0101] ;

[0102] In the formula, This represents the three-phase voltage matrix of each node in a low-voltage active distribution network during normal operation. This refers to the input-output mapping relationship of the model-free voltage calculation method at the foundation layer; P and These are the active power and net reactive power matrices for each of the three phases at each node.

[0103] The specific meaning of each variable is expressed by the following fifth formula:

[0104] ;

[0105] In the formula, N is the total number of nodes; Un is the three-phase voltage matrix of n nodes during normal operation. ; Let be the voltage of phase A at node n.

[0106] Furthermore, this embodiment utilizes the reactive and active power coordinated control capabilities of photovoltaic inverters and energy storage systems to achieve precise voltage regulation when node voltage exceedances occur in the low-voltage active distribution network, thereby ensuring the safe and stable operation of the low-voltage distribution network. This embodiment uses the regulation of photovoltaic inverters and energy storage systems as an example; the proposed model is universal and can be extended to other types of regulation equipment. The power balance model under low-voltage distribution network regulation conditions is expressed by the following sixth formula:

[0107] ;

[0108] In the formula, the superscript reg represents the variable during voltage regulation; This is the active power matrix of the energy stored at node n during voltage regulation, which is either emitted or absorbed. ; For voltage regulation, the matrix of reactive power generated or absorbed by the n-node photovoltaic inverter is expanded in the same way. .

[0109] Additionally, in one embodiment, reference is made to Figure 2 ,exist Figure 1 Step S11 in the illustrated embodiment also includes, but is not limited to, the following steps:

[0110] Step S61: Obtain training data for the control layer by subtracting the base layer data from different time periods;

[0111] Step S62: Determine the time delay required for training data of the daytime control layer and the nighttime control layer;

[0112] Step S63: Perform time difference processing on the training data of the daytime base layer and the nighttime base layer. Subtract the data of each time point from the data of the time delay to obtain the difference data. The difference data is the three-phase active static load change data, reactive static load change data and three-phase voltage change data.

[0113] Step S64: Use the differential data as the daytime training dataset and the nighttime training dataset.

[0114] It should be noted that the differential data obtained through subtraction can highlight the changes in three-phase active static load, reactive static load, and three-phase voltage over different time periods. This helps the model focus more on load variations rather than the absolute values ​​in the original data, thereby better capturing the patterns and trends of load changes and improving the ability to predict and regulate load changes. The data dimensionality may be reduced after differential processing, resulting in a relatively smaller data volume, which speeds up model training and reduces training time and computational resource consumption. Simultaneously, because the data is more representative and targeted, the model can converge to a better solution more quickly during training, improving training efficiency and effectiveness.

[0115] Finally, the calculation model for the voltage change of the incremental layer proposed in this embodiment can be expressed by the following seventh formula:

[0116] ;

[0117] In the formula, This represents the matrix representing the changes in three-phase voltage at each node of the voltage regulation layer; and These represent the three-phase active power difference matrix and reactive power difference matrix at each node, respectively. This represents the active power output matrix of the three-phase energy storage devices at each node; This represents the reactive power output matrix of the three-phase photovoltaic inverter at each node.

[0118] Additionally, in one embodiment, reference is made to Figure 2 ,exist Figure 1 Step S11 in the illustrated embodiment also includes, but is not limited to, the following steps:

[0119] Step S71: Obtain the three-phase voltage change matrix, three-phase active power difference matrix, three-phase reactive power difference matrix, three-phase energy storage device active power output matrix, and three-phase photovoltaic inverter reactive power output matrix of each node in the low-voltage distribution network.

[0120] Step S72: Obtain voltage regulation data and voltage regulation data in the low-voltage distribution network;

[0121] Step S73: Construct the three-phase voltage change matrix for each node in the control layer based on the no-voltage control data, voltage control data, three-phase voltage change matrix, three-phase active power difference matrix, three-phase reactive power difference matrix, three-phase energy storage device active power output matrix, and three-phase photovoltaic inverter reactive power output matrix.

[0122] Step S74: Construct a segmented and hierarchical calculation architecture based on the three-phase active net load matrix, the three-phase reactive net load matrix, the three-phase active difference matrix, and the three-phase reactive difference matrix.

[0123] It should be noted that the relationship between the actual voltage of node n at a certain moment after regulation and the base operating voltage before regulation during normal operation is expressed by the following eighth formula:

[0124] ;

[0125] In the formula, This is the matrix representing the change in three-phase voltage at node n during regulation. This is the three-phase voltage matrix of the n nodes after regulation.

[0126] By comparing equations (3) and (7), it can be found that the mapping relationship between power and voltage in a low-voltage active distribution network differs significantly between normal operation and voltage regulation. Under voltage regulation, the model needs to incorporate input parameters from regulation devices such as photovoltaic inverters and energy storage systems to accurately characterize the impact of regulation measures on system operating characteristics. Furthermore, as shown in equation (8), the regulated node voltage can be calculated by linearly superimposing the voltage under normal operation and the voltage change.

[0127] Based on the above analysis, this invention proposes a hierarchical voltage calculation architecture, which clearly divides the voltage calculation process of a low-voltage active distribution network into two layers: the base layer and the control layer. First, the reference normal operating voltage of the base layer is calculated. This reference voltage reflects the system state without any control measures. Subsequently, in the control layer, the voltage change caused by the control equipment is calculated. Finally, the reference voltage of the base layer will be... Voltage change of the control layer Superimpose the values ​​to complete the voltage calculation without an electrical model.

[0128] For the control layer The effect of changes in the output of the control equipment on the node voltage can usually be described using a voltage sensitivity matrix. The mapping relationship between the output changes of other nodes is expressed by the following ninth formula:

[0129] ;

[0130] According to the ninth formula, this embodiment uses the changes in the three-phase active power, reactive power, and corresponding voltage changes of the control equipment as training data, and uses a data-driven voltage segmented hierarchical calculation architecture to calculate the voltage changes during the voltage control process.

[0131] The specific meaning of each variable is expressed by the following tenth formula:

[0132] ;

[0133] When there is no voltage regulation data, , and These are the difference matrices of three-phase voltage, three-phase injected active power, and three-phase injected reactive power at node n at time t and time t-t_lag, respectively. ; ; .

[0134] In summary, this embodiment decouples the calculation of normal operating voltage from the impact of control measures on voltage, enabling more targeted selection and processing of training data at different levels.

[0135] Taking an uncontrolled data scenario as an example, the integrated voltage segmentation and hierarchical calculation architecture can be expressed by the following eleventh formula:

[0136] ;

[0137] in, For daytime training dataset, For daytime training dataset, ( ) represents the input correspondence for the construction method of the voltage segmented and hierarchical calculation architecture. ( ) represents the output correspondence of the construction method of the voltage segmented hierarchical calculation architecture.

[0138] like Figure 8 As shown, Figure 8 This is a structural diagram of a voltage calculation device without an electrical model according to an embodiment of the present invention. The present invention also provides a voltage calculation device without an electrical model, comprising:

[0139] The processor 801 can be implemented using a general-purpose central processing unit (CPU), microprocessor, application specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0140] The memory 802 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 802 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 802 and is called and executed by the processor 801 to execute the voltage segmentation and hierarchical calculation architecture construction method of the embodiments of this application.

[0141] The 803 input / output interface is used to implement information input and output.

[0142] The communication interface 804 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0143] Bus 805 transmits information between various components of the device (e.g., processor 801, memory 802, input / output interface 803, and communication interface 804);

[0144] The processor 801, memory 802, input / output interface 803, and communication interface 804 are connected to each other within the device via bus 805.

[0145] This application also provides an electronic device, including a voltage calculation device without an electrical model as described above.

[0146] This application embodiment also provides a storage medium, which is a computer-readable storage medium, storing a computer program that, when executed by a processor, implements the above-described method for constructing a voltage segmentation and hierarchical calculation architecture.

[0147] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof. The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separate, and may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0148] It will be understood by those skilled in the art that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically include computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0149] The above provides a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the above embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of the present invention.

Claims

1. A method for constructing a voltage segmented and hierarchical calculation architecture, characterized in that, include: Summarize historical operating data of node voltage and node active power in the low-voltage distribution network; Based on the historical operating data of the node voltage and the historical operating data of the node active power, the daytime operating period and the nighttime operating period are determined; A daytime training dataset is obtained based on the daytime running period, and a nighttime training dataset is obtained based on the nighttime running period. Using the three-phase active net load matrix and the three-phase reactive net load matrix corresponding to the low-voltage distribution network as inputs and the three-phase voltage matrix as outputs, an input-output mapping relationship is constructed. The input-output mapping relationship is trained using the daytime training dataset to obtain the daytime basic layer. At the same time, the input-output mapping relationship is trained using the nighttime training dataset to obtain the nighttime basic layer. After constructing a voltage change calculation model based on the three-phase active power difference matrix, the three-phase reactive power difference matrix, the active power output matrix of the three-phase energy storage device, and the reactive power output matrix of the three-phase photovoltaic inverter at each node, the model is trained using the daytime training dataset to obtain the daytime control layer. At the same time, the voltage change calculation model is trained using the nighttime training dataset to obtain the nighttime control layer. The daytime base operating voltage is obtained through the daytime base layer, the daytime voltage change is obtained through the daytime control layer, and the daytime actual voltage of the low-voltage distribution network is calculated based on the daytime base operating voltage and the daytime voltage change. The nighttime base operating voltage is obtained through the nighttime base layer, and the nighttime voltage change is obtained through the nighttime regulation layer. The actual nighttime voltage of the low-voltage distribution network is calculated based on the nighttime base operating voltage and the nighttime voltage change.

2. The method for constructing the voltage segmented hierarchical calculation architecture according to claim 1, characterized in that, The determination of daytime and nighttime operating periods includes: Obtain the start and stop times of photovoltaic nodes for a preset number of days in the low-voltage distribution network, as well as the total number of working days and the total number of photovoltaic nodes in the low-voltage distribution network; The daytime start time of the low-voltage distribution network is calculated based on the start time of power output, the total number of working days, and the total number of photovoltaic nodes; The daytime shutdown time of the low-voltage distribution network is calculated based on the shutdown time, the total number of working days, and the total number of photovoltaic nodes; The nighttime operating hours of the low-voltage distribution network are determined based on the daytime start and end times.

3. The method for constructing the voltage segmented hierarchical calculation architecture according to claim 1, characterized in that, The daytime voltage change is determined through the following steps: In the absence of voltage regulation data, the daytime voltage change is determined based on the three-phase active power difference matrix and the three-phase reactive power difference matrix. In the presence of the voltage regulation data, the daytime voltage change is determined based on the active power output matrix of the three-phase energy storage device and the reactive power output matrix of the three-phase photovoltaic inverter.

4. The method for constructing the voltage segmented hierarchical calculation architecture according to claim 1, characterized in that, The process of obtaining the daytime training dataset based on the daytime runtime segment and the nighttime training dataset based on the nighttime runtime segment includes: Training data for the control layer is obtained by subtracting the base layer data from different time periods, including: Determine the time delay required for the training data of the daytime control layer and the nighttime control layer; The training data of the daytime base layer and the nighttime base layer are subjected to time difference processing. The data at each time point is subtracted from the data of the time delay to obtain the difference data, wherein the difference data is the three-phase active net load change data, the three-phase reactive net load change data and the three-phase voltage change data. The differential data is used as the daytime training dataset and the nighttime training dataset.

5. A voltage calculation device without an electrical model, characterized in that, It includes at least one control processor and a memory for communicatively connecting to the at least one control processor; The memory stores instructions that can be executed by the at least one control processor to enable the at least one control processor to perform the method for constructing the voltage segmentation and hierarchical computing architecture as described in any one of claims 1 to 4.

6. An electronic device, characterized in that, Includes the voltage calculation device without an electrical model as described in claim 5.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions for causing a computer to perform the method for constructing a voltage segmentation and hierarchical computing architecture as described in any one of claims 1 to 4.

Images