Voltage control method based on dynamic cluster division of high-permeability photovoltaic power distribution network

Abstract

The invention provides a voltage control method based on dynamic cluster division of a high-permeability photovoltaic power distribution network, and the method comprises the steps: carrying out the Newton-Raphson method load flow calculation of a power distribution network power system based on the topological structure data of the power distribution network, line parameters, the access position of a photovoltaic inverter and energy storage equipment, and equipment parameter information, and constructing a comprehensive modular function; based on the comprehensive modularization function, according to the load capacity of each moment, taking the average load capacity as a standard, utilizing an improved artificial hummingbird algorithm to dynamically divide time segments of the clusters and selecting dominant nodes of each cluster; based on the dominant node, taking the minimum node voltage deviation and the minimum system loss as objective functions, and constructing a second-order cone programming model when a preset constraint condition is satisfied; and solving the second-order cone programming model by using a solver CPLEX to obtain the reactive power regulation quantity of the distributed photovoltaic inverter and the active power regulation quantity of the energy storage equipment, thereby improving the power supply reliability and the electric energy quality of the photovoltaic power distribution network.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network voltage control, and in particular, to a voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network. Background Art

[0002] Due to the advantages of being green, clean, safe and low-carbon, the proportion of renewable energy in the overall energy structure has been continuously increasing. Therefore, distributed photovoltaic (PV) has been massively connected to the distribution network, and the installation capacity of distributed PV has been growing rapidly. Its inherent intermittency will bring more uncertainties and dynamics to the distribution network aiming to meet energy demands, resulting in the operating state of the distribution network getting closer and closer to the limit. Reverse overload and reverse power flow phenomena have occurred in some regional power grids, causing problems such as local grid voltage over-limit and poor frequency stability, thus affecting the reliability of power grid power supply and power quality.

[0003] Therefore, how to improve the power supply reliability and power quality of the photovoltaic distribution network has become a technical problem that needs to be urgently solved by those skilled in the art. Summary of the Invention

[0004] The present invention provides a voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network to solve the defect of poor power supply reliability and power quality of the photovoltaic distribution network in the prior art.

[0005] In a first aspect, the present invention provides a voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network, including: Based on the distribution network topology structure data, line parameters, access positions of photovoltaic inverters and energy storage devices, and device parameter information, perform Newton-Raphson power flow calculation on the distribution network power system of distributed PV to obtain the voltage-power matrix, and the active power and reactive power provided by the photovoltaic inverters and energy storage devices, and construct a comprehensive modular function; Based on the comprehensive modular function, according to the load at each moment, taking the average load as the standard, use an improved artificial hummingbird algorithm to dynamically divide the clusters in time segments and select the dominant nodes of each cluster; Based on the dominant nodes, taking the minimization of node voltage deviation and the minimization of system loss as the objective function, construct a second-order cone programming model under the premise of meeting the preset constraint conditions; Use the solver CPLEX to solve the second-order cone programming model to obtain the reactive power regulation amount of the distributed PV inverter and the active power regulation amount of the energy storage device.

[0006] According to the voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network provided by the present invention, the Newton-Raphson power flow calculation on the distribution network power system of distributed PV includes: Using the power flow calculation equation of the Newton-Raphson method, matrix elements are obtained in the distribution network power system with distributed photovoltaic, and a Jacobian matrix is obtained using the matrix elements; The voltage-power sensitivity between each node of the distribution network power system and the electrical distance defined by the voltage-power sensitivity are obtained through the inverse transformation of the Jacobian matrix.

[0007] According to a voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network provided by the present invention, it further includes: Evaluating the regional voltage regulation ability using the reactive power of the photovoltaic inverter and the active power output of the energy storage device, as follows: ; In the formula, represents the voltage regulation ability of the cluster ; and respectively represent the active power and reactive power that can be provided by the photovoltaic inverter or energy storage device in ; ; In the formula, is the voltage overstep of the node with the maximum voltage in the cluster , if there is no voltage overlimit, then ; and respectively represent the maximum active power margin and the maximum reactive power margin that can be provided by the photovoltaic inverter and the energy storage device at the node ; and represent the active power sensitivity and reactive power sensitivity of the voltage amplitude of the node to the active power injection at the node .

[0008] According to a voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network provided by the present invention, the comprehensive modular function is: ; ; In the formula, is the comprehensive index of cluster performance; is the weighted coefficient of the edge between the nodes , ; is the sum of the edge weight coefficients of all nodes connected to the node ; is the node connected to The total sum of the edge weight coefficients of all connected nodes; is the sum of the weights of all nodes, represents the electrical distance, represents the cluster inner nodes the electrical distance between, represents the total number of clusters.

[0009] According to a voltage control method based on dynamic cluster division of a high - permeability photovoltaic distribution network provided by the present invention, the use of an improved artificial hummingbird algorithm to perform dynamic division of time segments for clusters and select the dominant nodes of each cluster includes: Select a preset number of individuals with the optimal fitness in the population for crossover operation and update the population fitness, compare the fitness of the new and old individuals, and select the better fitness; Combined with the better fitness, use a non - uniform mutation operator to perturb the population position, so that the hummingbird individuals perform a rough search in the global region at the initial stage of solving and perform a fine search in the local region at the later stage, perform dynamic division of time segments for the clusters and select the dominant nodes of each cluster.

[0010] According to a voltage control method based on dynamic cluster division of a high - permeability photovoltaic distribution network provided by the present invention, based on the dominant nodes, taking minimizing the node voltage deviation and minimizing the system loss as the objective function, includes: Determine the voltage amplitude and rated voltage value of the dominant nodes of each cluster at time ; Count the number of clusters and the total number of nodes after cluster division; Based on the voltage amplitude, the rated voltage value, the number of clusters and the total number of nodes, determine the objective function.

[0011] According to a voltage control method based on dynamic cluster division of a high - permeability photovoltaic distribution network provided by the present invention, the preset constraint conditions include at least one of the distribution network power flow constraint, the distributed photovoltaic power generation output constraint, the node voltage constraint, and the energy storage device operation constraint.

[0012] According to a voltage control method based on dynamic cluster division of a high - permeability photovoltaic distribution network provided by the present invention, constructing a second - order cone programming model under the premise of satisfying the preset constraint conditions includes: Use the second - order cone programming method to convexify the non - convex and non - linear problem of the objective function and transform it into a second - order cone programming problem; Based on the second - order cone programming problem, introduce auxiliary variables, and combine at least one of the distribution network power flow constraint, the distributed photovoltaic power generation output constraint, the node voltage constraint, and the energy storage device operation constraint to construct a second - order cone programming model.

[0013] In a second aspect, the present invention further provides a voltage control device based on dynamic cluster division of a high-penetration photovoltaic distribution network, including: An acquisition module, configured to perform Newton-Raphson power flow calculation on the power system of the distributed photovoltaic distribution network based on the distribution network topology structure data, line parameters, access positions of photovoltaic inverters and energy storage devices, and device parameter information, obtain a voltage-power matrix, and active power and reactive power provided by the photovoltaic inverters and energy storage devices, and construct a comprehensive modular function; A division module, configured to perform dynamic division of time segments on the clusters based on the comprehensive modular function, with reference to the average load amount according to the load amount at each moment, and select the leading nodes of each cluster by using an improved artificial hummingbird algorithm; A construction module, configured to construct a second-order cone programming model based on the leading nodes, with the minimization of node voltage deviation and the minimization of system loss as the objective functions, under the premise of meeting preset constraint conditions; A solution module, configured to solve the second-order cone programming model by using a solver CPLEX to obtain the reactive power regulation amount of the distributed photovoltaic inverter and the active power regulation amount of the energy storage device.

[0014] In a third aspect, the present invention further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where when the processor executes the program, the voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network as described in any one of the above is implemented.

[0015] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network as described in any one of the above is implemented.

[0016] In a fifth aspect, the present invention further provides a computer program product, including a computer program, and when the computer program is executed by a processor, the voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network as described in any one of the above is implemented.

[0017] The voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network provided by the present invention includes: performing Newton-Raphson power flow calculation on the distribution network power system based on the distribution network topology structure data, line parameters, access positions of photovoltaic inverters and energy storage devices, and device parameter information, and constructing a comprehensive modular function; based on the comprehensive modular function, according to the load at each moment, taking the average load as the standard, using an improved artificial hummingbird algorithm to dynamically divide the clusters in time periods and select the dominant nodes of each cluster; based on the dominant nodes, taking the minimization of node voltage deviation and the minimization of system loss as the objective function, constructing a second-order cone programming model under the satisfaction of preset constraint conditions; using the solver CPLEX to solve the second-order cone programming model to obtain the reactive power regulation amount of the distributed photovoltaic inverter and the active power regulation amount of the energy storage device, and through the coordinated cooperation of the reactive power regulation of the distributed photovoltaic inverter and the active power regulation of the energy storage device, combined with the improved artificial hummingbird algorithm to achieve dynamic cluster division, selecting the dominant nodes of each cluster on the basis of considering the adjustable voltage ability of the nodes, ensuring the flexibility of regulation of each cluster, and improving the power supply reliability and power quality of the photovoltaic distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0019] Figure 1 is a flowchart of the voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network provided in this embodiment; Figure 2 is a topological structure diagram of the IEEE33-node example provided in this embodiment; Figure 3 is a curve graph of the total photovoltaic output and load of the system provided in this embodiment; Figure 4 is a schematic diagram of voltage violation nodes at typical moments after the distributed photovoltaic is connected provided in this embodiment; Figure 5 is a flowchart of the voltage control strategy for dynamic cluster division of a high-penetration photovoltaic distribution network provided in this embodiment; Figure 6 is a graph of the dynamic cluster division results at 3 typical moments provided in this embodiment; Figure 7 is a voltage fluctuation graph under the control strategy regulation using the method of the present invention provided in this embodiment; Figure 8It is a comparison chart of the voltage control effects of two different control strategies provided in this embodiment at three typical moments; Figure 9 It is a schematic structural diagram of a voltage control device based on dynamic cluster division of a high-penetration photovoltaic distribution network provided in this embodiment; Figure 10 It is a schematic structural diagram of an electronic device provided in this embodiment. Detailed implementation manners

[0020] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Obviously, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without making creative efforts shall fall within the protection scope of the present invention.

[0021] Figure 1 It is a schematic flowchart of a voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network provided in this embodiment.

[0022] As Figure 1 shown, the voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network provided in the embodiment of the present invention mainly includes the following steps: 101. Based on the distribution network topology structure data, line parameters, access positions of photovoltaic inverters and energy storage devices, and device parameter information, perform Newton-Raphson power flow calculation on the distribution network power system with distributed photovoltaics, obtain the voltage-power matrix, as well as the active power and reactive power provided by the photovoltaic inverters and energy storage devices, and construct an integrated modular function.

[0023] In a specific implementation process, obtain the distribution network power system data with a high proportion of distributed photovoltaics, including distribution network topology structure data, distribution network load, electrical quantity data such as new energy output, and access positions and device parameter data of photovoltaic inverters and energy storage devices. The device parameter data includes: the capacities of photovoltaic inverters and energy storage devices, the charge and discharge efficiency of energy storage devices, the maximum and minimum state of charge limits of energy storage devices, etc. Based on the obtained relevant data, perform Newton-Raphson power flow calculation on the distribution network power system with distributed photovoltaics to obtain the voltage-power matrix and the matrix as well as the active / reactive power that can be provided by the photovoltaic inverter / energy storage device and to construct an integrated modular function .

[0024] Specifically, it includes obtaining matrix elements in the distribution network power system with distributed photovoltaics by using the Newton-Raphson power flow calculation equation, and obtaining the Jacobian matrix from the matrix elements; obtaining the voltage-power sensitivity and the electrical distance defined by the voltage-power sensitivity between each node of the distribution network power system through the inverse transformation of the Jacobian matrix.

[0025] The voltage-power sensitivity matrix is transformed into the following form through the inverse operation of the Jacobian matrix, as shown in (1): (1) In the formula, and are the fluctuations of the voltage phase angle and amplitude respectively, V represents the current amplitude; and are the fluctuations of the active power and reactive power of the injection node respectively; is the full voltage-power sensitivity matrix, indicating the influence degree between the power fluctuation and the node voltage change, represents the sensitivity of the active power fluctuation to the voltage amplitude fluctuation ; represents the sensitivity of the reactive power fluctuation to the voltage amplitude fluctuation ; represents the sensitivity of the active power fluctuation to the voltage phase angle fluctuation ; represents the sensitivity of the reactive power fluctuation to the voltage phase angle fluctuation .

[0026] The calculation formula for the electrical distance is defined by the active and reactive-voltage amplitude sensitivity matrix as follows (2): (2) In the formula, and represent the sensitivity of the voltage amplitude of node to the injected active / reactive power of node ; and represent the sensitivity of the voltage amplitude of node to the injected active / reactive power of node ; and represent the sensitivity of the voltage amplitude of node to the injected active / reactive power of node ; and represent the sensitivity of the voltage amplitude of node to the injected active / reactive power of node Injected active / reactive power sensitivity. Represents the electrical distance defined by the active-voltage magnitude sensitivity matrix, Represents the electrical distance defined by the reactive-voltage magnitude sensitivity matrix.

[0027] When the electrical connection between nodes and is closer, the active / reactive-voltage sensitivities between them are closer. At the same time, the electrical distance between them is smaller. The difference in the impact of active and reactive power regulation of node injection on voltage is reflected by the magnitude of the sensitivity coefficient, and there is no need to additionally increase the weight coefficient. Therefore, considering the same weight, the distance is (3): (3) Then, use whether the reactive power of the PV inverter and the active power output of the energy storage device can solve the voltage over-limit problem in the cluster to evaluate the regional voltage regulation ability. The regional voltage regulation ability is defined as (4): (4) In the formula, represents the voltage regulation ability of the cluster ; and respectively represent the active power and reactive power that can be provided by the PV inverter or the energy storage device in (5) (6) In the formula, is the voltage overshoot of the node with the maximum voltage in the cluster . If there is no voltage over-limit, then ; and respectively represent the maximum active margin and maximum reactive margin that can be provided by the PV inverter and the energy storage device at node ; and represent the active power sensitivity and reactive power sensitivity of the voltage magnitude of node to the active power injection of node .

[0028] Then, combine the voltage-power sensitivity and the electrical distance described by the regional regulation ability to calculate the edge weight between two nodes in the distribution network. It is defined that at any time and any node , The comprehensive modular function between them is (7): (7) (8) Wherein, is the comprehensive index of cluster performance; is the node , The weighted coefficient of the edge between; is the sum of the edge weight coefficients of all nodes connected to the node ; represents the electrical distance, represents the cluster The electrical distance between the nodes inside; represents the total number of clusters, is the sum of the edge weight coefficients of all nodes connected to the node ; is the sum of the weights of all nodes. Assuming that the node and the node are divided into the same cluster, , otherwise .

[0029] 102. Based on the comprehensive modular function, according to the load at each moment, taking the average load as the standard, use the improved artificial hummingbird algorithm to dynamically divide the clusters by time period and select the leading nodes of each cluster.

[0030] Set the maximum number of iterations of the artificial hummingbird algorithm , the population size of hummingbirds, the current iteration value of the algorithm , the food source location and access table of the hummingbird population. Combining the relevant data obtained above, according to the load at each moment, taking the average load as the standard, using the artificial hummingbird algorithm based on the crossover operator and non-uniform mutation operator can solve the problems of high computing power consumption, long solution time and easy to fall into local optimum in the later stage when the traditional artificial hummingbird algorithm solves the optimal solution of high-dimensional complex problems.

[0031] Specifically, select a preset number (for example, the top 1 / 3 of the high-quality individuals) of individuals with the best fitness in the population for crossover operation and update the population fitness. Compare the fitness of the new and old individuals, and select the better fitness to increase the probability of the algorithm jumping out of the trap of local optimum.

[0032] Combined with the better fitness, use the non-uniform mutation operator to moderately perturb the population position, so that the hummingbird individuals can search in a relatively global area in the initial stage of the solution and perform fine search in the local area in the later stage.

[0033] Among them, the non-uniform operator can be expressed as (9): (9) In the formula: represents the position of the food source at dimension,[[]] at the th iteration; and respectively represent the upper and lower limits of the dimensional space; (10) In the formula: is the maximum number of iterations set for the algorithm, is the parameter determining the non-uniformity, r represents a constant between 0 and 1, used to control the change rate of the function as the number of iterations increases.

[0034] Use the improved artificial hummingbird algorithm to perform dynamic cluster division for time segments, and select nodes with stronger adjustable ability and higher sensitivity to over-limit nodes as the leading nodes of each cluster, so that the leading nodes can reflect and regulate the voltage level within the cluster.

[0035] 103. Based on the leading nodes, with minimizing the node voltage deviation and minimizing the system loss as the objective function, construct a second-order cone programming model under the premise of meeting the preset constraint conditions.

[0036] Specifically, determine the voltage amplitude and rated voltage value of the leading nodes of each cluster at time ; count the number of clusters and the total number of nodes after cluster division; based on the voltage amplitude, rated voltage value, number of clusters and total number of nodes, determine the objective function, such as (11): (11) In the formula: and represent the weight coefficients in the objective function, and their values are set to 0.5 and 0.5 in this embodiment; represents the voltage amplitude of the leading nodes within each cluster at time ; represents the rated voltage value; and respectively represent the number of clusters and the total number of nodes after cluster division; represents the line impedance value between nodes ; represents the sum of the squares of the currents between nodes .

[0037] Among them, the preset constraint conditions include at least one of the following: power flow constraints of the distribution network, output constraints of distributed photovoltaic power generation, node voltage constraints, and operation constraints of energy storage devices.

[0038] The power flow constraint conditions of the distribution network are as shown in (12) and (13): (12) (13) In the formula, respectively represent the active power and reactive power of the photovoltaic at node ; respectively represent the active power and reactive power of the load at node ; respectively represent the active power and reactive power at the head end of node , represents the voltage amplitude of node ; represents the voltage amplitude of node ; represents the conductance between nodes ; represents the susceptance between nodes ; represents the voltage phase angle difference between nodes .

[0039] The output constraint conditions of distributed photovoltaic power generation are (14) and (15): (14) (15) In the formula: and respectively represent the active and reactive power that the -th photovoltaic inverter can output at time ; represents the capacity of the photovoltaic inverter; and respectively represent the minimum and maximum active power outputs of the -th photovoltaic inverter.

[0040] The node voltage constraint condition is (16): (16) In the formula: and respectively represent the upper and lower limits of the voltage of node .

[0041] The operation constraint conditions of the energy storage device are (17) and (18): (17) (18) Wherein: and respectively represent the active power of energy storage charging and discharging at the lower node at time ; represents the energy stored at node at time ; represents the energy capacity of the energy storage device at node ; and respectively represent the minimum and maximum state - of - charge limits of the energy storage device at node , represents the maximum power limit of the energy storage device during charging or discharging.

[0042] Then, the non - convex and non - linear problem of the objective function is convexified by using the second - order cone programming method, and transformed into a second - order cone programming problem that can be effectively solved; based on the second - order cone programming problem, auxiliary variables are introduced, and combined with at least one of the distribution network power flow constraint, distributed photovoltaic power generation output constraint, node voltage constraint, and energy storage device operation constraint, a second - order cone programming model is constructed.

[0043] Specifically, variable substitution is performed on the objective function. Let , and two sets of auxiliary variables and are introduced, and the following additional constraint (19) is added: (19) The objective function formula is transformed into the following form, such as (20): (20) 104. Use the solver CPLEX to solve the second - order cone programming model to obtain the reactive power regulation amount of the distributed photovoltaic inverter and the active power regulation amount of the energy storage device.

[0044] Use the commercial solver CPLEX to solve the voltage optimization control model for the reactive power regulation amount of the distributed photovoltaic inverter and the active power regulation amount of the energy storage device, obtain the optimal solution, solve the problem of voltage over - limit under the randomness of photovoltaic output, reduce the voltage fluctuation degree of the distribution network, and make the voltage - over - limit nodes in the cluster return to the normal range.

[0045] In order to verify the effect of the present invention, the following simulation tests are carried out.

[0046] The present invention builds a simulation example of a distribution network IEEE33 node with distributed photovoltaics by means of MATLAB, and the system topology is as Figure 2 shown. The system base capacity = 10 MVA, PV is connected to nodes 10, 13, 18, 20, 24, 28, 32. The capacity of each PV inverter is set to 600 kVA, and its maximum reactive power adjustable range is kVar. Energy storage devices are installed at nodes 12 and 27, with a maximum capacity of 600 kVA, a charge-discharge efficiency of 0.95, and maximum and minimum state of charge limits are 0.8 and 0.2 respectively. PV output and load data come from the actual monitoring data of a certain place within 24 hours. The total system PV output and load curves are as Figure 3 shown. The parameters of the improved artificial hummingbird algorithm are selected as the individual scale D = 30, and the maximum number of algorithm iterations = 100.

[0047] PV will generate a large amount of power at noon due to high light intensity. Then, as time goes by, the power output gradually weakens, and the power output is 0 at night. Due to the influence of load demand fluctuations throughout the day, power reverse flow and voltage over-limit phenomena will occur at some nodes of the distribution network at different times.

[0048] According to the load at each moment, this invention takes the average load as the standard, divides 24 hours into three time periods: from 20:00 of the previous day to 09:00 of the current day, 09:00 - 14:00, and 14:00 - 20:00, and conducts specific analysis on the output conditions at three typical times 09:00, 14:00, and 20:00. As Figure 4 can be seen, at 9 o'clock, as the light intensity increases, the absolute value of the voltage deviation of each node decreases slightly, and the system voltage regulation demand is low; at 14 o'clock, the PV output increases significantly. Due to the mismatch between PV output and load demand, the probability of voltage over-limit phenomena in the system increases, and the system voltage regulation demand is large; at 20 o'clock, the active power output of PV is 0, the load demand of the system ushers in the second peak period, the overall voltage level of the system decreases, the voltage deviation values of each node are large, and the voltage regulation demand is slightly high.

[0049] As Figure 5 shown. The method of this invention inputs the system network structure and related parameters, the access positions of PV inverters and energy storage devices, and the source-load data curve; then calculates the comprehensive electrical distance between nodes and the comprehensive modular function of any node Set the parameters of the Artificial Hummingbird Algorithm (AHA), the maximum number of iterations, and the population size; initialize the food source positions, access tables, and iteration counts of the hummingbird population; guide foraging, area foraging, and migratory foraging; determine whether G is less than the maximum number of iterations. If so, output the optimal hummingbird individual and fitness value to obtain the cluster division result. If not, then G = G + 1 and perform another iteration. After the division is completed, build a simulation example of the IEEE 33-node distribution network with distributed photovoltaics, input the distribution network topology structure and parameter information, and perform system power flow calculation through the Newton-Raphson method; based on the topology structure after cluster division, construct a voltage control model with the objective function of minimizing voltage deviation and minimizing power loss function, use the second-order cone method to linearly process the non-convex and non-linear parts of the objective function, and finally use the commercial solver CPLEX to solve the voltage control model to obtain the optimal solutions for the processing and regulation of photovoltaic inverters and energy storage devices that meet various constraint conditions.

[0050] On the basis of considering the voltage regulation ability of nodes, select the nodes with higher sensitivity to over-limit nodes as the leading nodes of each cluster, and regulate the output of photovoltaic inverters and energy storage devices in the distribution network to achieve voltage and power stability.

[0051] As Figure 6 shown. After the dynamic cluster division of the distribution network by the present invention, the results at three typical times are as shown in the figure. As can be seen from Table 1, although the sensitivity of node 32 to node 33 is higher, considering that the power regulation ability of node 28 is better than that of node 32, the improved Artificial Hummingbird Algorithm selects node 28 as the leading node.

[0052] Table 1 Sensitivity ranking of over-limit nodes

[0053] As Figure 7 shown. The present invention takes the balanced node voltage (1.0 p.u.) as the reference, and uses the leading nodes after dynamic cluster division to cooperate with the reactive power regulation of distributed photovoltaic inverters and the active power regulation of energy storage devices to adjust the voltage of the distribution network. As can be seen from Figure 7 , the 24-hour operating voltage of the distribution network is controlled within the permitted range.

[0054] As Figure 8 shown. To verify the efficiency of the strategy proposed by the present invention, under the condition of high-proportion photovoltaic access in the distribution network, the per-unit value of voltage amplitude and the voltage deviation values under different strategies will be compared by using the power factor control method of photovoltaic inverters and the strategy of this paper. Under two different voltage regulation strategies, the system voltage results at three typical times 09:00, 14:00, and 20:00 are as Figure 8 shown.

[0055] In Figures (a) and (b), the lighting conditions are good at 09:00 and 14:00. The photovoltaic inverter is greatly affected by the light intensity based on the power factor. At 14:00, the photovoltaic system outputs a large amount of active power. The inverter is in a high-power operation state and can affect the voltage by regulating the reactive power. , At these two moments, the photovoltaic inverter controlled based on the power factor controls the voltage level within the normal range, and the total system voltage deviations are 0.493 p.u. and 0.554 p.u. respectively. In Figure (c), at 20:00 in the non-lighting scenario, since the photovoltaic system does not generate electricity, the photovoltaic inverter based on the power factor cannot generate power for voltage regulation, and the total system voltage deviation is 1.232 p.u. However, under the control method of the present invention, the leading nodes of each cluster regulate the output of the photovoltaic inverter and the energy storage device to adjust the voltage. The influence of the light intensity on the voltage regulation control of this strategy is small, and the total system voltage deviations at each moment are 0.224 p.u., 0.090 p.u., and 0.313 p.u. respectively, which is significantly better than the voltage regulation effect of the former.

[0056] Based on the same general inventive concept, the present invention also protects a voltage control device based on dynamic cluster division of a high-penetration photovoltaic distribution network. The voltage control device based on dynamic cluster division of a high-penetration photovoltaic distribution network described below can be correspondingly referred to the voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network described above.

[0057] Figure 9 is a schematic structural diagram of the voltage control device based on dynamic cluster division of a high-penetration photovoltaic distribution network provided in this embodiment. As Figure 9 shown, a voltage control device based on dynamic cluster division of a high-penetration photovoltaic distribution network provided in this embodiment includes: An acquisition module 901, configured to perform Newton-Raphson power flow calculation on the distribution network power system of distributed photovoltaics based on the distribution network topology structure data, line parameters, access positions of photovoltaic inverters and energy storage devices, and device parameter information, obtain a voltage-power matrix, and the active power and reactive power provided by the photovoltaic inverters and energy storage devices, and construct a comprehensive modular function; A division module 902, configured to perform dynamic division of time segments on the clusters based on the comprehensive modular function according to the load at each moment, with the average load as the standard, and select the leading nodes of each cluster by using an improved artificial hummingbird algorithm; A construction module 903, configured to construct a second-order cone programming model based on a dominant node, with minimizing node voltage deviation and minimizing system loss as objective functions, under the premise of satisfying preset constraint conditions; A solving module 904, configured to solve the second-order cone programming model by using a solver CPLEX to obtain the reactive power regulation amount of the distributed photovoltaic inverter and the active power regulation amount of the energy storage device.

[0058] Figure 10 It is a schematic structural diagram of the electronic device provided in this embodiment.

[0059] As Figure 10 shown, the electronic device may include: a processor 1010, a communications interface 1020, a memory 1030, and a communication bus 1040. Among them, the processor 1010, the communications interface 1020, and the memory 1030 complete mutual communication through the communication bus 1040. The processor 1010 may call logic instructions in the memory 1030 to execute a voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network. The method includes: performing Newton-Raphson power flow calculation on the distribution power system of distributed photovoltaics based on distribution network topology structure data, line parameters, access positions of photovoltaic inverters and energy storage devices, and device parameter information to obtain a voltage-power matrix and the active power and reactive power provided by the photovoltaic inverters and the energy storage devices, and constructing a comprehensive modular function; based on the comprehensive modular function, according to the load amount at each moment, with the average load amount as the standard, using an improved artificial hummingbird algorithm to perform dynamic division of time segments for the clusters and select the dominant nodes of each cluster; based on the dominant node, with minimizing node voltage deviation and minimizing system loss as objective functions, constructing a second-order cone programming model under the premise of satisfying preset constraint conditions; using a solver CPLEX to solve the second-order cone programming model to obtain the reactive power regulation amount of the distributed photovoltaic inverter and the active power regulation amount of the energy storage device.

[0060] In addition, when the logical instructions in the above-mentioned memory 1030 are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0061] On the other hand, the present invention also provides a computer program product. The computer program product includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network provided by the above-mentioned various methods. The method includes: based on the distribution network topology structure data, line parameters, access positions of photovoltaic inverters and energy storage devices, and device parameter information, performing Newton-Raphson power flow calculation on the distribution network power system of distributed photovoltaics to obtain a voltage-power matrix and the active power and reactive power provided by photovoltaic inverters and energy storage devices, and constructing a comprehensive modular function; based on the comprehensive modular function, according to the load amount at each moment, using the average load amount as a standard, using an improved artificial hummingbird algorithm to dynamically divide the clusters in time segments and select the leading nodes of each cluster; based on the leading nodes, using the minimization of node voltage deviation and the minimization of system loss as the objective function, constructing a second-order cone programming model under the condition of satisfying preset constraints; using the solver CPLEX to solve the second-order cone programming model to obtain the reactive power regulation amount of the distributed photovoltaic inverter and the active power regulation amount of the energy storage device.

[0062] In another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network provided by the above-mentioned various methods. The method includes: based on the distribution network topology structure data, line parameters, access positions and device parameter information of photovoltaic inverters and energy storage devices, performing Newton-Raphson power flow calculation on the distribution network power system of distributed photovoltaics to obtain a voltage-power matrix and the active power and reactive power provided by photovoltaic inverters and energy storage devices, and constructing a comprehensive modular function; based on the comprehensive modular function, according to the load at each moment, taking the average load as the standard, using an improved artificial hummingbird algorithm to dynamically divide the clusters in time segments and select the dominant nodes of each cluster; based on the dominant nodes, taking minimizing the node voltage deviation and minimizing the system loss as the objective function, constructing a second-order cone programming model under the satisfaction of preset constraint conditions; using the solver CPLEX to solve the second-order cone programming model to obtain the reactive power regulation amount of the distributed photovoltaic inverter and the active power regulation amount of the energy storage device.

[0063] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0064] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course also by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disc, etc., including several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0065] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the 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 invention.

Claims

1. A voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network, characterized in that Including: Based on the distribution network topology structure data, line parameters, access positions of photovoltaic inverters and energy storage devices, and device parameter information, perform Newton-Raphson power flow calculation on the distribution network power system with distributed photovoltaics, obtain the voltage-power matrix, and the active power and reactive power provided by the photovoltaic inverters and energy storage devices, and construct an integrated modular function; Based on the integrated modular function, according to the load at each moment, with the average load as the standard, use the improved artificial hummingbird algorithm to dynamically divide the time segments of the cluster and select the leading nodes of each cluster; Based on the leading nodes, with minimizing the node voltage deviation and minimizing the system loss as the objective function, construct a second-order cone programming model under the premise of meeting the preset constraint conditions; Use the solver CPLEX to solve the second-order cone programming model to obtain the reactive power regulation amount of the distributed photovoltaic inverter and the active power regulation amount of the energy storage device.

2. The voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network according to claim 1, wherein The Newton-Raphson power flow calculation for the distribution network power system with distributed photovoltaics includes: Using the Newton-Raphson power flow calculation equation, obtain the matrix elements in the distribution network power system with distributed photovoltaics, and use the matrix elements to obtain the Jacobian matrix; Obtain the voltage-power sensitivity between each node of the distribution network power system and the electrical distance defined by the voltage-power sensitivity through the inverse transformation of the Jacobian matrix.

3. The voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network according to claim 2, wherein Also including: Use the reactive power of the photovoltaic inverter and the active power output of the energy storage device to evaluate the regional voltage regulation ability as follows: ； In the formula, represents the voltage regulation ability of the cluster ; and respectively represent the active power and reactive power that can be provided by the photovoltaic inverter or energy storage device in , that is, the voltage regulation ability in the area, as follows: ； ； wherein, is the cluster the node with the maximum voltage voltage increment. If there is no voltage over-limit, then ; and respectively represent the maximum active power margin and the maximum reactive power margin that the photovoltaic inverter and the energy storage device can provide at the node ; and represent the sensitivities of the voltage amplitude of the node to the active power injection and the reactive power injection at the node .

4. The voltage control method based on dynamic clustering division of a high-penetration photovoltaic distribution network according to claim 3, wherein The integrated modular function is: ； ； Wherein, is the comprehensive index of cluster performance; is the node , is the weighted coefficient of the edge between; is the sum of all node edge weight coefficients connected to the node ; is the sum of all node edge weight coefficients connected to the node ; is the sum of the weights of all nodes, represents the electrical distance, represents the cluster the electrical distance between the internal nodes ; represents the total number of clusters.

5. The voltage control method based on dynamic cluster division of a high-permeability photovoltaic distribution network according to claim 1, wherein, The use of the improved artificial hummingbird algorithm to dynamically divide the time segments of the cluster and select the leading nodes of each cluster includes: Select a preset number of individuals with the optimal fitness in the population for crossover operation and update the population fitness, compare the fitness of the new and old individuals, and select the better fitness; Combined with the better fitness, use the non-uniform mutation operator to perturb the population position, so that the hummingbird individuals perform a rough search in the global area at the initial stage of the solution and perform a fine search in the local area at the later stage, dynamically divide the time segments of the cluster and select the leading nodes of each cluster.

6. The voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network according to claim 1, wherein The based on the leading nodes, with minimizing the node voltage deviation and minimizing the system loss as the objective function includes: Determine the voltage amplitude and rated voltage value of the leading node of each of the clusters at the moment ; Count the number of clusters and the total number of nodes after cluster division; Based on the voltage amplitude, the rated voltage value, the number of clusters and the total number of nodes, determine the objective function.

7. The voltage control method based on dynamic clustering division of a high-penetration photovoltaic distribution network according to claim 1, wherein The preset constraint conditions include at least one of the distribution network power flow constraint, the distributed photovoltaic power generation output constraint, the node voltage constraint, and the energy storage device operation constraint.

8. The voltage control method based on dynamic cluster division of a high-penetration photovoltaic distribution network according to claim 7, characterized in that, The constructing a second-order cone programming model under the premise of meeting the preset constraint conditions includes: Use the second-order cone programming method to convexify the non-convex and non-linear problem of the objective function and transform it into a second-order cone programming problem; Based on the second-order cone programming problem, introduce auxiliary variables, and combine at least one of the distribution network power flow constraint, the distributed photovoltaic power generation output constraint, the node voltage constraint, and the energy storage device operation constraint to construct a second-order cone programming model.

9. A voltage control device based on dynamic cluster division of a photovoltaic distribution network with high penetration rate, characterized in that, Including: The acquisition module is used to perform Newton-Raphson power flow calculation on the distribution power system of distributed photovoltaics based on the distribution network topology structure data, line parameters, access locations of photovoltaic inverters and energy storage devices, and device parameter information, obtain the voltage-power matrix, as well as the active power and reactive power provided by the photovoltaic inverters and energy storage devices, and construct an integrated modular function; The division module is used to dynamically divide the time segments of the cluster based on the integrated modular function, according to the load at each moment, with the average load as the standard, and select the leading nodes of each cluster by using the improved artificial hummingbird algorithm; The construction module is used to construct a second-order cone programming model with minimizing the node voltage deviation and minimizing the system loss as the objective function based on the leading nodes under the condition of meeting the preset constraints; The solution module is used to solve the second-order cone programming model by using the solver CPLEX to obtain the reactive power regulation amount of the distributed photovoltaic inverter and the active power regulation amount of the energy storage device.

10. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the voltage control method based on the dynamic cluster division of the high-penetration photovoltaic distribution network as described in any one of claims 1 to 8.

Images