Reactive power control method and device for household photovoltaic transformer area

Abstract

The invention provides a household photovoltaic area reactive power control method and device, and relates to the technical field of power system reactive power control. The method comprises the steps of establishing a network model of a household photovoltaic transformer area, constructing a target optimization model of the transformer area network model according to at least one optimization target of no out-of-limit node voltage, minimum total network loss, minimum three-phase unbalance degree and reactive self-balance of a low-voltage transformer area network, setting constraint conditions for the target optimization model, and obtaining a power adjustment value. The reactive power of the household photovoltaic transformer area is adjusted, and the stability of power grid operation under the condition of household photovoltaic high permeability can be remarkably improved. Moreover, when the adjustment is not carried out for the first time, the optimization result of the reactive power of the previous time section is used as the input quantity of the total target optimization model of the current time section, and the optimization result of the reactive power of the current time section is output, thereby achieving the coherent control of the reactive power of the household photovoltaic transformer area. And the stability of section adjustment in continuous time can be ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of reactive power control of electric power systems, and in particular to a method and device for reactive power control of household photovoltaic stations. Background Art

[0002] In modern power systems, distributed power sources represented by distributed photovoltaics are developing rapidly, especially household photovoltaics, that is, photovoltaic power generation equipment installed in residents' homes, which has been rapidly popularized. Distributed photovoltaics have brought many challenges to the stable operation of the power grid due to their randomness and volatility. Since the amount of electricity generated by these photovoltaic devices is unstable and sometimes more and sometimes less, after a large number of them are connected to the power grid, it may lead to a series of problems such as increased overvoltage risks in the substation area, significant increase in line network losses, and a sudden increase in three-phase imbalance. These problems seriously threaten the safety and stable operation of the power grid.

[0003] To solve the above problems, the traditional method is to control overvoltage by adjusting the position of the transformer tap in the substation. However, with high photovoltaic penetration, this method is difficult to ensure that the voltage is qualified at all times, especially at night, when photovoltaic equipment does not generate electricity, which easily leads to low voltage. In addition, the position of the transformer tap in the substation usually needs to be adjusted without load, that is, it needs to be adjusted during power outage, which seriously affects residents' electricity consumption at night.

[0004] Therefore, it is necessary to propose a method and device for controlling reactive power in household photovoltaic areas, which can improve the stability of power grid operation under conditions of high penetration of household photovoltaics. Summary of the invention

[0005] The purpose of the present invention is to provide a method and device for controlling reactive power in a household photovoltaic area, which can improve the stability of power grid operation under conditions of high household photovoltaic penetration.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a method for controlling reactive power of a household photovoltaic area, comprising the following steps: establishing a network model of a household photovoltaic area, constructing a target optimization model of the area network model according to an optimization target, and setting constraints for the target optimization model, wherein the optimization target includes at least one of the following: no voltage over-limit at the node of the low-voltage area network, minimum total network loss, minimum three-phase imbalance, and reactive self-balancing; obtaining an adjustment value of reactive power as an optimization result according to the constraints and the target optimization model, and adopting a coherent control strategy to adjust the reactive power of a continuous time section of the household photovoltaic area, comprising the following steps: determining whether the reactive power of the household photovoltaic area is adjusted for the first time in the current time section; if it is the first adjustment, then Obtain each target optimization model of each optimization target, and obtain the overall target optimization model of the current time section based on each target optimization model, and output the optimization result of the reactive power of the current time section; if it is not the first adjustment, obtain the optimization result of the reactive power output by the overall target optimization model of the previous time section, use the optimization result of the reactive power of the previous time section as the input of the overall target optimization model of the current time section, and output the optimization result of the reactive power of the current time section; use the optimization result of the reactive power of the current time section as the input of the overall target optimization model of the next time section, and roll-up the overall target optimization model to achieve continuous control and optimization of the reactive power of household photovoltaic stations in continuous time sections.

[0008] In order to further improve the optimization effect, more reasonably weigh the weights of various target optimizations, and obtain a more reliable overall target optimization model, the present invention provides a preferred scheme in the first aspect, wherein the adjustment value of reactive power is obtained according to the constraint conditions and the target optimization model, as the optimization result, to adjust the reactive power of the household photovoltaic area, wherein the variable weight principle is introduced, including: setting weight rules according to different weights corresponding to different preset operating conditions of the household photovoltaic area, determining the weights of the operating conditions corresponding to each target optimization model at the current time section according to the operating conditions of the household photovoltaic area at the current time section, wherein the operating conditions of the household photovoltaic area include load demand and photovoltaic output; obtaining the overall target optimization model according to each target optimization model and the weight of each target optimization model.

[0009] In a second aspect, the present invention provides a reactive power control device for a household photovoltaic area, which executes the above-mentioned control method, including: a model building module, which is used to establish a network model of the household photovoltaic area, build a target optimization model of the area network model according to the optimization target, and set constraints for the target optimization model, wherein the optimization target includes at least one of the node voltage of the low-voltage area network without exceeding the limit, the minimum total network loss, the minimum three-phase imbalance and reactive self-balancing; an adjustment module, which is used to obtain the adjustment value of reactive power as the optimization result according to the constraint conditions and the target optimization model, and adopt a coherent control strategy to adjust the reactive power of the continuous time section of the household photovoltaic area, including the following: a judgment submodule, which is used to judge whether the current time section is the first time to adjust the reactive power of the household photovoltaic area; an optimization result output submodule, which is used to obtain the reactive power of each household photovoltaic area if it is the first adjustment. The optimization targets are optimized by using the optimization models of the optimization targets, and according to the optimization models, the overall optimization model of the current time section is obtained, and the optimization result of the reactive power of the current time section is output; if it is not the first adjustment, the optimization result of the reactive power output by the overall optimization model of the previous time section is obtained, and the optimization result of the reactive power of the previous time section is used as the input of the overall optimization model of the current time section, and the optimization result of the reactive power of the current time section is output; a rolling update submodule is used to use the optimization result of the reactive power of the current time section as the input of the overall optimization model of the next time section, and to update the overall optimization model in a rolling manner, so as to realize the continuous control and optimization of the reactive power of household photovoltaic stations in continuous time sections.

[0010] Compared with the prior art, the above technical solution has the following advantages:

[0011] The reactive power control method and device for household photovoltaic areas of the present invention comprehensively consider multiple objectives such as voltage, total network loss, three-phase unbalance control and reactive self-balancing, construct a target optimization model, and adjust the reactive power of household photovoltaic areas according to the power adjustment value obtained based on the constraint conditions and the target optimization model, which can significantly improve the stability of power grid operation under the condition of high penetration rate of household photovoltaics. In addition, if it is not the first adjustment, the optimization result of the reactive power output by the total target optimization model of the previous time section is obtained, and the optimization result of the reactive power of the previous time section is used as the input of the total target optimization model of the current time section, and the optimization result of the reactive power of the current time section is output, so as to realize the coherent control of the reactive power of household photovoltaic areas, and ensure the stability of the adjustment in the continuous time section. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0013] Figure 1 A schematic diagram of the system structure of a household photovoltaic area in an implementation environment involved in various embodiments of the present invention;

[0014] Figure 2 A flowchart of a method for controlling reactive power in a household photovoltaic area provided in the first embodiment of the present invention;

[0015] Figure 3 This is a flow chart of S100 in the household photovoltaic area reactive power control method provided in the first embodiment of the present invention;

[0016] Figure 4 This is a flow chart of S200 in the household photovoltaic area reactive power control method provided in the first embodiment of the present invention;

[0017] Figure 5 A schematic diagram of a module of a household photovoltaic area reactive power control device provided in the first embodiment of the present invention;

[0018] Figure 6 This is a flow chart of S220 in the household photovoltaic area reactive power control method provided in the second embodiment of the present invention;

[0019] Figure 7 This is a flow chart of S220 in the household photovoltaic area reactive power control method provided in the third embodiment of the present invention:

[0020] Figure 8 A flowchart of a method for controlling reactive power of household photovoltaic stations at all times with high penetration rate based on the NSGA-III algorithm and coherent control strategy provided in the third embodiment of the present invention;

[0021] Fig. 9 This is a curve diagram of the intra-day voltage variation over time of the remote node voltage before and after control in Example 3 of the present invention;

[0022] Fig.10 This is a curve diagram of the total network loss of the substation area changing with time before and after the control of the third embodiment of the present invention;

[0023] Fig.11 This is a graph showing the change of the three-phase imbalance degree over time before and after the control of the third embodiment of the present invention;

[0024] Fig.12This is a graph showing the change in reactive power output over time in the substation area before and after control in the third embodiment of the present invention.

[0025] The reference numerals are as follows: model building module 100 , adjustment module 200 , judgment submodule 210 , first optimization result output submodule 220 , second optimization result output submodule 230 , rolling update submodule 240 . DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] Please refer to Figure 1 , Figure 1 The following is a schematic diagram of the system structure of a household photovoltaic power station in an implementation environment involved in various embodiments of the present invention. The household photovoltaic power station is a power station where household photovoltaic equipment is installed. The station is a basic unit in the power system, usually referring to the area powered by a transformer. In this area, there may be many households with photovoltaic power generation equipment installed. The electricity generated by these equipment can be connected to the power grid for use by other households or businesses. Figure 1 As shown, the household photovoltaic area in the example here includes 12 nodes, and a transformer is installed at the outlet of the area. Seven nodes in the system are load nodes, that is, nodes with loads mounted, and four nodes are installed with household photovoltaic units, and a total of 33 household photovoltaic units can be installed.

[0028] Embodiment 1: Please refer to Figure 2 First, this embodiment provides a method for controlling reactive power of household photovoltaic areas, including S100 establishing a network model of household photovoltaic areas, constructing a target optimization model of the area network model according to the optimization target, and setting constraints for the target optimization model, wherein the optimization target includes at least one of the node voltage of the low-voltage area network without exceeding the limit, the minimum total network loss, the minimum three-phase imbalance, and reactive self-balancing. S200 obtains the adjustment value of reactive power as the optimization result according to the constraint conditions and the target optimization model, and adopts a coherent control strategy to adjust the reactive power of the continuous time section of the household photovoltaic area. More detailed steps are as follows:

[0029] S100 establishes a network model of household photovoltaic substations, constructs a target optimization model of the substation network model according to the optimization target, and sets constraints for the target optimization model. As an example, this embodiment sets four optimization targets, namely: no over-limit of node voltage in the low-voltage substation network, minimum total network loss, minimum three-phase imbalance, and reactive self-balancing. The target optimization model constructed in this way is a multi-objective optimization model. Here, household photovoltaic substations specifically refer to household photovoltaic high penetration low-voltage substations.

[0030] Please refer to Figure 3 Specifically, S100 has the following steps: S110 establishes a network model for household photovoltaic stations, S120 constructs a multi-objective optimization model with the optimization objectives of no-limit node voltage, minimum total network loss, minimum three-phase imbalance and reactive power self-balancing in the low-voltage station network, and S130 sets the constraints of the multi-objective optimization model. More detailed steps are as follows:

[0031] S110 establishes a network model for household photovoltaic areas, and establishes a complete network model for household photovoltaic areas based on the network topology, line parameters, household photovoltaic installation location and quantity, and household photovoltaic access method of the household photovoltaic low-voltage area. The network topology can include different topologies such as tree, star, and mesh. Each structure has different effects on the performance of the network, such as reliability, transmission rate, and scalability. Line parameters can include wire diameter, wire material, resistance, reactance, etc. These parameters directly affect the power loss and voltage distribution of the network. The installation location, quantity, and access method of household photovoltaics will affect the power flow distribution and power balance of the network.

[0032] S120 builds a multi-objective optimization model with the optimization objectives of no-limit node voltage, minimum total network loss, minimum three-phase imbalance and reactive power self-balancing in the low-voltage area network. Among them:

[0033] The node voltage is within the limit, that is, to ensure that the voltage of all nodes in the low-voltage substation network is within the allowable range, and the voltage is not too high or too low. Specifically, the node voltage is within the limit target optimization model to ensure that the voltage of all nodes is within the allowable range. The node voltage is within the limit target optimization model can be implemented by the following expression (1): (1) In formula (1) represents the node voltage at the p-phase node i, represents the minimum voltage allowed at the p-phase node i, represents the maximum voltage allowed at the p-phase node i, Represents the penalty function constant parameter to be set.

[0034] The total network loss is minimized, that is, by adjusting the network structure and operation mode, the power loss of the entire network is reduced. Specifically, the network loss can be minimized through the total network loss minimum target optimization model. The total network loss minimum target optimization model can be implemented through the following expression (2): (2) In formula (2), N represents the total number of network nodes. , They represent the active and reactive loads at the node i of the corresponding phase p, respectively. represents the node voltage at the node i of the corresponding phase p, , Represents the resistance and reactance of the line between node i and node j.

[0035] The three-phase unbalance is minimized, that is, the three-phase load distribution is optimized to reduce the unbalance of the three-phase current and voltage. The three-phase load distribution can be optimized through the three-phase unbalance minimum target optimization model. The three-phase unbalance minimum target optimization model can be implemented in turn through the following expressions (3) and (4): (3) (4) In formulas (3) and (4) , , , Respectively represent the average current, A phase current, B phase current, and C phase current. Take 1, 2, and 3 to represent phases A, B, and C respectively. Indicates the corresponding phase current.

[0036] Reactive power self-balancing, that is, achieving reactive power balance in the network through reactive power compensation devices and other means. The reactive power balance in the network is ensured through the reactive power self-balancing target optimization model. The reactive power self-balancing target optimization model can be implemented through the following expression (5): (5) In formula (5) express The reactive power injected into the upper power grid at the exit of the corresponding substation.

[0037] In this embodiment, after obtaining the above-mentioned target optimization models, the final overall target optimization model is the sum of the above-mentioned models. The overall target optimization model can be realized by the following expression (6): (6) The above target optimization model is a function used to mathematically describe the relationship between the optimization target and the variables. By solving the function, the optimal solution to achieve the optimization target can be found.

[0038] S130 sets the constraints of the multi-objective optimization model. The constraints of this embodiment include the upper and lower limits of the voltage of each phase of each node in the photovoltaic area, the upper limit constraint of the current allowed to flow through each branch, the network flow constraint, and the household photovoltaic inverter adjustment constraint.

[0039] The node phase voltage constraint ensures that the voltage of each node in the network is within a safe and stable range to avoid damage to equipment and users caused by excessively high or low voltage. This constraint is particularly important in the case of high penetration of household photovoltaic access, because the volatility and intermittency of photovoltaic power sources may affect voltage stability. The node phase voltage constraint can be achieved by the following expression (7): (7) In formula (7) represents the node voltage at node i at time t, , , They represent the lowest allowed node voltage values ​​of phases A, B, and C at node i, respectively. , , They respectively represent the maximum allowed node voltage values ​​of phases A, B, and C at node i.

[0040] The branch current constraint is that the current flowing through each branch should not exceed its maximum value allowed to prevent overheating, damage to equipment or safety accidents. The branch current constraint can be achieved by the following expression (8): (8) In formula (8) , , They represent the current values ​​of the three phases A, B, and C flowing through the line between node i and node j at time t, It represents the maximum current value allowed to flow through the line between nodes i and j.

[0041] The power flow constraint is to ensure that the input power and output power of each node in the network are balanced and satisfy the power flow equation of the network. This is the basic requirement for the stable operation of the power system. The power flow constraint can be achieved by the following expression (9): (9) In formula (9) , represents the active part of the load power and photovoltaic power at the p-phase node i, , represents the reactive part of the load power and photovoltaic power at the p-phase node i, , represents the conductance and susceptance of the line between node i and node j, Represents the phase angle difference between the node voltages at node i and node j.

[0042] The regulation constraints of household photovoltaic inverters are as follows: As the interface device between the photovoltaic system and the power grid, the regulation capability of the photovoltaic inverter has an important impact on the stability and efficiency of the system. Therefore, the regulation constraints of the inverter, such as maximum power point tracking, voltage and frequency regulation range, etc., need to be considered in the optimization process. The regulation constraints of household photovoltaic inverters can be realized by the following expressions (10), (11) and (12): (10) (11) (12) In formulas (10) to (12), , They represent the active and reactive power sent by the distributed photovoltaic at the p-phase node i, represents the maximum apparent power of the distributed photovoltaic installed at node i connected to phase p, It represents the maximum active power of distributed photovoltaic installed at node i connected to phase p, , They respectively represent the maximum and minimum reactive power of the distributed photovoltaic installed at node i connected to phase p.

[0043] By comprehensively considering the above constraints and combining with the target optimization model, the corresponding power adjustment value can be calculated. Incorporating the above constraints into the model can ensure that the solved power adjustment value can meet all constraints.

[0044] S200 obtains the adjustment value of reactive power according to the constraint conditions and the target optimization model as the optimization result, and adopts a coherent control strategy to adjust the reactive power of the continuous time section of the household photovoltaic area.

[0045] Please refer to Figure 4 , S200 is implemented through the following steps:

[0046] S210 determines whether the reactive power of the household photovoltaic area is adjusted for the first time in the current time section. The time section may be a specific time or time period in the power system. The first adjustment means that the reactive power of the household photovoltaic area has not been adjusted before the current time section.

[0047] If it is the first adjustment, then obtain each target optimization model of each optimization target, and obtain the total target optimization model of the current time section according to each target optimization model, and output the optimization result of reactive power of the current time section. Taking the optimization target of no over-limit node voltage, minimum total network loss, minimum three-phase imbalance and reactive self-balance of the low-voltage substation network as an example, for each target to be optimized, respectively establish a target optimization model, and the specific expression refers to formulas (1) to (5) of this embodiment.

[0048] If S230 is not the first adjustment, the optimization result of the reactive power output by the overall target optimization model of the previous time section is obtained, and the optimization result of the reactive power of the previous time section is used as the input of the overall target optimization model of the current time section, and the optimization result of the reactive power of the current time section is output.

[0049] S240 uses the optimization result of the reactive power of the current time section as the input of the overall target optimization model of the next time section, and updates the overall target optimization model in a rolling manner, so as to realize the continuous control and optimization of the reactive power of household photovoltaic stations in continuous time sections, and ensure the stability of regulation in continuous time periods.

[0050] Specifically, the above steps S230 and S240 are as follows: if the current time section is not the first time to adjust the reactive power of the household photovoltaic area, then the overall target optimization model is updated according to the area status of the current time section to achieve coherent control of the reactive power of the household photovoltaic area. In other words, if the current time section is not the first adjustment, the overall target optimization model of the previous time section can be used as a basis, and the overall target optimization model can be updated according to the optimization results of the previous time section and the actual situation of the current time section to achieve coherent control of the reactive power of the household photovoltaic area, so as to ensure the stability of the adjustment in the continuous period.

[0051] The reactive power regulation of this implementation can be achieved by adjusting the input or removal of reactive compensation equipment, adjusting the tap position of the transformer, etc. Specifically, it can be implemented according to the size of the power adjustment value and the actual situation of the household photovoltaic area. By adjusting the reactive power, the voltage quality can be improved and the stability of the power system can be improved.

[0052] The reactive power control method for household photovoltaic areas in this embodiment comprehensively considers multiple objectives such as voltage, total network loss, three-phase imbalance control and reactive self-balancing, constructs a target optimization model, and adjusts the reactive power of household photovoltaic areas according to the power adjustment value obtained based on constraint conditions and the target optimization model, which can significantly improve the stability of grid operation under conditions of high household photovoltaic penetration.

[0053] Please refer to Figure 5Secondly, this embodiment provides a household photovoltaic area reactive power control device, which executes the above control method and mainly consists of the following modules:

[0054] The model building module 100 is used to establish a network model of a household photovoltaic area, build a target optimization model of the area network model according to the optimization target, and set constraints for the target optimization model. The optimization target includes at least one of the node voltage of the low-voltage area network is not exceeded, the total network loss is minimized, the three-phase imbalance is minimized, and reactive power self-balancing;

[0055] The adjustment module 200 is used to obtain the adjustment value of reactive power according to the constraint conditions and the target optimization model as the optimization result, and adopt a coherent control strategy to adjust the reactive power of the continuous time section of the household photovoltaic area, including the following:

[0056] The judgment submodule 210 is used to judge whether the reactive power of the household photovoltaic area is adjusted for the first time in the current time section;

[0057] The first optimization result output submodule 220 is used to obtain each target optimization model of each optimization target if it is the first adjustment, and obtain the overall target optimization model of the current time section according to each target optimization model, and output the optimization result of reactive power of the current time section;

[0058] The second optimization result output submodule 230, if it is not the first adjustment, obtains the optimization result of reactive power output by the total target optimization model of the previous time section, uses the optimization result of reactive power of the previous time section as the input of the total target optimization model of the current time section, and outputs the optimization result of reactive power of the current time section;

[0059] The rolling update submodule 240 is used to use the optimization result of the reactive power of the current time section as the input of the overall target optimization model of the next time section, and to rolling update the overall target optimization model to achieve continuous control and optimization of the reactive power of household photovoltaic stations in continuous time sections.

[0060] It should be noted that the examples and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the contents disclosed in the method of the above embodiment. It should be noted that the above modules as part of the device can be run in Figure 1 In the environment shown, it can be implemented by software or by hardware.

[0061] This embodiment reconstructs the multi-objective optimization model through a coherent control strategy to achieve coherent control of the reactive power of household photovoltaic stations, thereby ensuring the stability of regulation in continuous time sections.

[0062] Embodiment 2: Based on Example 1, this embodiment provides a more preferred method for controlling reactive power in household photovoltaic areas, wherein S200 obtains an adjustment value of reactive power according to constraints and a target optimization model as an optimization result to adjust the reactive power of household photovoltaic areas, wherein a variable weight principle is introduced to reconstruct the multi-objective optimization model.

[0063] Please refer to Figure 6 , where S220 is mainly implemented through the following steps:

[0064] S221 sets weight rules according to different weights corresponding to different working conditions of household photovoltaic areas. The working condition of the area is the operating status of the low-voltage area at a specific time section. Load demand reflects the user's demand for electricity in the time section, which may change with factors such as time of day, season, weather, etc. Photovoltaic output depends on the installed capacity, lighting conditions, temperature, etc. of the photovoltaic system, indicating the power that the photovoltaic system can provide at the current time section. According to the preset working conditions, such as the typical daily load curve of the low-voltage area, photovoltaic output forecast, etc., weights are assigned to each optimization target (such as node voltage without over-limit, minimum total network loss, minimum three-phase imbalance, reactive self-balancing, etc.). These weights reflect the relative importance of each optimization target under the preset working conditions. The weight can be a fixed weight or a variable weight, that is, the weight can be determined according to the preset working conditions and the variable weight principle. It should be noted that the variable weight principle is relative to the fixed weight principle. The variable weight principle takes into account the dynamic changes in the state of the area and adaptively adjusts the weights of each optimization target. For example, during peak load periods, more attention is paid to voltage stability and minimization of network losses; while during periods of high photovoltaic output, more attention is paid to reactive self-balancing and reduction of three-phase imbalance. As an example, this embodiment provides a variable weight rule, which is established through the following steps:

[0065] First, the generated weights are as shown in expression (13): (13) The number of nodes in the household photovoltaic area where the photovoltaic output is greater than the load demand is counted using the following expression (14): (14) The judgment variable is generated according to the number of nodes obtained by statistics, which is implemented by the following expression (15): (15) The weight rule is generated according to the judgment variable, which is implemented by the following expression (16): (16) In formulas (13) to (16), Indicates the weight coefficient of the voltage over-limit degree target of the substation area, Indicates the weight coefficient of the total network loss target of the substation area, Indicates the weight coefficient of the three-phase unbalance target of the substation area, Indicates the weight coefficient of the reactive power self-balancing target of the substation. Indicates the number of nodes in the area where the photovoltaic output is greater than the load demand. Indicates the total number of photovoltaic nodes installed in the area. , They represent the PV active output and active load demand at node i respectively. represents the weight combination of each optimization objective at time t, Represents the weight combination of each optimization objective at time t-1.

[0066] S222 determines the weights of the operating conditions corresponding to the operating conditions of each target optimization model at the current time section according to the operating conditions of the household photovoltaic area at the current time section, wherein the operating conditions of the household photovoltaic area include load demand and photovoltaic output.

[0067] S223 obtains the overall target optimization model according to each target optimization model and the weight of each target optimization model. Thus, the overall target optimization model reconstructed by the variable weight principle is obtained, which is realized by the following expression (6): (6)', Exemplarily, first collect the substation status information of the current time section, including but not limited to key parameters such as substation load demand, photovoltaic output, node voltage, branch current, three-phase imbalance, etc. Compare the substation status information of the current time section with the substation status information of the previous time section to analyze the trend and amplitude of the state change. According to the change of the substation state, dynamically adjust the weight coefficients of each target to be optimized (such as voltage without over-limit, minimum total network loss, minimum three-phase imbalance, reactive self-balancing, etc.). For example, if the current voltage fluctuation is large, the weight of the "voltage without over-limit" target can be appropriately increased. According to the new weight coefficient, adjust the form of the overall target optimization function model. The overall target optimization model is the weighted sum of each target optimization model. Therefore, the change of the weight coefficient will directly affect the shape of the function and the position of the optimal solution. Substitute the new weight coefficient and other relevant parameters into the overall target optimization model to obtain the updated function expression.

[0068] This embodiment reconstructs the multi-objective optimization model by introducing the variable weight principle, and combines it with a coherent control strategy to take into account the dynamic changes in the substation status. It can adaptively adjust the weights of each optimization objective within a continuous time period according to the operating conditions, thereby improving the reliability and accuracy of the multi-objective optimization model and further improving the reactive power regulation capability.

[0069] Embodiment three: In this embodiment, S200 obtains the reactive power adjustment value according to the constraint conditions and the target optimization model, and as the optimization result, introduces the objective function of the relative change amount of reactive power adjustment penalty of the household photovoltaic area in the continuous time section to update the overall target optimization model. Figure 7 , wherein S220, based on S221 and S222 of the second embodiment, is mainly implemented by the following steps:

[0070] S224 When the relative change in reactive power of a household photovoltaic area in a continuous time section is greater than a preset value, an adjustment penalty function is obtained to encourage a smooth change in reactive power, that is, an objective function for adjusting the penalty for the relative change in reactive power of a household photovoltaic area in a continuous time section. Specifically, by monitoring the reactive power change of the photovoltaic system in a continuous time section and calculating its relative change. When this relative change exceeds a preset threshold, the adjustment penalty function will be triggered. The specific form of the penalty function may be a mathematical expression related to the size of the reactive power change, which is used to "penalize" this sharp change in the subsequent optimization process, thereby encouraging the system to find an operating strategy that makes the reactive power change more stable. The adjustment penalty function of this embodiment is implemented by the following expression (17): (17) In formula (17), M represents the number of actions of the optimization algorithm. , Respectively time, The i-th reactive action value at time instant.

[0071] S225 determines the weight of the penalty function for the current time section according to the preset household photovoltaic area working conditions corresponding to different weights and the household photovoltaic area working conditions at the current time section;

[0072] S226 obtains the overall target optimization model according to each target optimization model, the adjustment penalty function, and each weight. That is, after the target function is introduced, it is added to the overall target optimization model as an additional optimization target. According to the importance of each optimization target and penalty function, the corresponding weight is assigned, and the final overall target optimization model can be obtained by weighted summation. The final overall target optimization model is updated to the following expression (18): (18) In formula (18), The weight coefficient representing the voltage over-limit degree target of the substation area; Indicates the weight coefficient of the total network loss target of the substation area; Indicates the weight coefficient of the three-phase unbalance target of the substation; Indicates the weight coefficient of the reactive power self-balancing target of the substation; Represents the weight coefficient of the continuous control target; represents the overall objective function of the multi-objective optimization problem, , , , , Represents each sub-objective function.

[0073] This embodiment introduces an adjustment penalty function and considers the weight of the adjustment penalty function as one of the objective functions to reconstruct the multi-objective optimization model, thereby further improving the reliability and accuracy of the multi-objective optimization model and further improving the reactive power regulation capability.

[0074] Embodiment 4: This embodiment provides an algorithm for solving the overall target optimization model of the above-mentioned embodiment 1 and embodiment 2. That is, for S200, the adjustment value of reactive power is obtained according to the constraint conditions and the target optimization model, as the optimization result, based on the constraint conditions, the overall target optimization model is solved by the NSGA-III algorithm to obtain the power adjustment value, which is the optimization result.

[0075] The model usually includes one or more optimization objectives, such as minimizing network losses, maximizing voltage stability, minimizing the investment in reactive power compensation equipment, etc. Then, the above constraints are incorporated into the model to ensure that the solved power adjustment value can meet all constraints.

[0076] An optimization algorithm called NSGA-Ⅲ (Non-dominated Sorting Genetic Algorithm Ⅲ) can be used to solve this model. The NSGA-Ⅲ algorithm is a multi-objective optimization algorithm that can find a balance between multiple optimization objectives so that the solved power adjustment value can not only meet various constraints but also be as close to the optimization goal as possible. Specifically, the NSGA-Ⅲ algorithm can search and evaluate possible power adjustment values ​​based on constraints and target optimization models. During the search process, the algorithm will continuously generate new candidate solutions, and retain excellent solutions and eliminate inferior solutions by comparing their advantages and disadvantages. Finally, when the algorithm reaches the stopping condition, such as reaching the maximum number of iterations or finding a solution that meets the requirements, the optimal power adjustment value can be selected from the retained valid solutions.

[0077] Please refer to Figure 8This embodiment provides a method for controlling reactive power of household photovoltaic stations with high penetration rate at all times based on NSGA-Ⅲ algorithm and coherent control strategy, which is mainly divided into two steps: initialization and multi-objective optimization under coherent control. It is implemented by the following steps:

[0078] S11 inputs the distribution network and photovoltaic equipment parameters. The parameters in this step can be obtained by referring to the above-mentioned embodiment 1 step S110 to establish a network model of the household photovoltaic area, and establishing a complete network model of the household photovoltaic area based on the network topology structure, line parameters, household photovoltaic installation location and quantity, and household photovoltaic access method of the household photovoltaic low-voltage area.

[0079] S12 Input population size and iteration parameters. The population size determines the number of individuals included in each generation of the genetic algorithm. A larger population size helps to better explore the search space, increase the algorithm's search ability and diversity, but it will also increase the computational cost and time accordingly. A smaller population size may lead to insufficient coverage of the search space and easily fall into a local optimal solution. Usually, the population size is set according to the complexity of the problem and the computing resources. For complex multi-objective optimization problems, a larger population size may be required to ensure that a high-quality solution is searched. The number of iterations determines how many generations the algorithm will run, and each generation will perform crossover, mutation, and selection operations to gradually optimize the individuals in the population. A larger number of iterations helps the algorithm to search for the optimal solution more fully, but it will also increase the computational cost and time. A smaller number of iterations may cause the algorithm to fail to fully converge and fail to find a high-quality solution. The appropriate number of iterations needs to be set according to the complexity of the problem and the convergence speed. For complex problems, more iterations may be required to ensure that the algorithm can fully search for the optimal solution. At the same time, the limitations of computing resources need to be considered to avoid unnecessary waste of computing. Here, you can set it as needed based on the actual situation, such as the number of optimization targets, chip computing power, speed requirements, cost, etc.

[0080] S13 determines whether it is the first optimization control. This step can be understood as the step S210 of the first embodiment above, which determines whether the reactive power of the household photovoltaic area is adjusted for the first time in the current time section.

[0081] When S14 is the first optimization control, the control strategy is initialized to 0, and then S16 is executed.

[0082] S15, when it is not the first optimization control, inputs the control strategy, that is, the result output by the total target optimization model of the previous time section, which can refer to step S230 of the first embodiment above. If it is not the first adjustment, the optimization result of the reactive power output by the total target optimization model of the previous time section is obtained, and the optimization result of the reactive power of the previous time section is used as the input of the total target optimization model of the current time section, that is, the control strategy described in step S15. In this way, coherent control in continuous time sections is achieved.

[0083] S16 adaptively selects weights according to the working conditions. This step indicates the start of multi-objective optimization. This step can be understood as step S222 of the above-mentioned embodiment 2.

[0084] S17 calculates the fitness value of each objective function. The fitness value refers to the degree of excellence of an individual in a specific environment or problem, and it is usually closely related to the value of the objective function. In the NSGA-Ⅲ algorithm, the fitness value is used to guide the evolutionary direction of the population, that is, to select which individuals to perform crossover, mutation and other operations to produce new individuals. The calculation of the fitness value is usually based on the value of the objective function. In the NSGA-Ⅲ algorithm, the fitness value is usually determined by non-dominated sorting and reference point selection mechanisms.

[0085] S18 ENS non-dominated sorting. In the NSGA-Ⅲ algorithm, non-dominated sorting is used to determine the rank of each individual so that individuals with higher ranks can be selected first in subsequent selection operations. Prioritize selection based on the rank of the individual. Generally, individuals with higher ranks have a higher probability of being selected to ensure that excellent individuals can be retained in the next generation.

[0086] S19 determines whether the convergence condition is met. If the convergence condition is not met, S20 is executed. If the convergence condition is met, S23 is directly jumped. According to the preset convergence condition, such as the number of iterations input in step S12, it is judged whether the algorithm converges. If the total objective function converges, the optimal solution in the current population is output, and this optimal solution is used as the household photovoltaic output control strategy, that is, the reactive power adjustment value. Otherwise, iterative evolution continues.

[0087] S20 selects individuals based on the reference point mapping distance.

[0088] S21 simulates the binary crossover algorithm.

[0089] S22 polynomial mutation method, and then returns to S17.

[0090] The NSGA-Ⅲ algorithm continuously iterates the evolutionary population through three key steps: S20 selecting individuals based on reference point mapping distance, S21 simulating binary crossover algorithm, and S22 polynomial mutation method to find a uniformly distributed Pareto frontier solution. These steps together constitute the core framework of the NSGA-Ⅲ algorithm, which enables it to have excellent performance in dealing with multi-objective optimization problems.

[0091] S23 Household photovoltaic output control strategy. When the total objective function converges and the iteration is completed, the optimal solution in the current population is output as the household photovoltaic output control strategy for the current time section. At this point, the reactive power adjustment of the current time section is completed.

[0092] S24 Control strategy replacement: Replace the control strategy of the previous time section, and the control strategy of the current time section can be used for reactive power adjustment of the next time section.

[0093] This embodiment uses the NSGA-III algorithm to solve the overall target optimization model to obtain the power adjustment value, which can achieve multiple optimization goals such as node voltage without exceeding the limit, minimum total network loss, minimum three-phase imbalance and reactive self-balancing, thereby more accurately realizing the regulation of reactive power in household photovoltaic areas.

[0094] In order to further compare the advantages of NSGA-Ⅲ coherent control in the full-time reactive power control method of high-penetration household photovoltaic stations in this embodiment, the control results of different control strategies (NSGA-Ⅲ traditional control, PSO traditional control and NSGA-Ⅲ coherent control) are given, mainly in terms of average voltage over-limit rate, average station output reactive power (kW), average station total network loss (kW), and average three-phase imbalance. The comparison is shown in Table 1 below.

[0095] Table 1: Comparison of control results of different control strategies

[0096] In addition, the daily voltage variation curve of the remote node voltage before and after the control of the high penetration household photovoltaic area full-time reactive power control method based on the NSGA-Ⅲ algorithm and the coherent control strategy of this embodiment is also given (reference Fig. 9 )、The curve of total network loss in the substation area changing with time before and after control (reference Fig.10 ), the curve of the three-phase imbalance degree changing with time before and after control (reference Fig.11 , the output reactive power of the station area changes with time before and after control (reference Fig.12 ).

[0097] It can be seen that after the NSGA-Ⅲ coherent control strategy of this embodiment optimizes the reactive power control of the high-penetration household photovoltaic area, the average voltage over-limit rate is low, the average area output reactive power (kW), the average area total network loss (kW), and the average three-phase imbalance result parameter performance has great advantages. Therefore, the NSGA-Ⅲ coherent control strategy of this embodiment has a very good self-reactive power regulation capability, which reduces the total network loss of the area and improves the stability and economy of the low-voltage distribution network operation.

[0098] In summary of the above embodiments, the present invention discloses a method for optimizing reactive power control of household photovoltaic power stations with continuous control at all times under high penetration rates, aiming to solve the problems of strong lag, high degree of manual intervention, and high equipment investment cost in traditional regulation methods of low-voltage power stations under high penetration rates of household photovoltaic power stations. The method takes household photovoltaic inverters with single-phase and three-phase access as reactive power regulation objects, takes voltage control of the power station, reduction of total network loss, three-phase imbalance control, and reactive power self-balancing as goals, constructs a variable-weight multi-objective reactive power optimization model for each time section, and further comprehensively considers the economy, stability, and computational efficiency of regulation within a continuous period. The proposed method can fully tap the reactive power regulation potential of household photovoltaic power stations with high penetration rates, realize comprehensive power quality management, reduce total network loss in the power station, and improve the stability and economy of low-voltage distribution network operation.

[0099] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the above-mentioned embodiments of the methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0100] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0101] The above-described embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the present invention. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the attached claims.

Claims

1. A method for controlling reactive power in a household photovoltaic area, characterized in that: The steps include: Establish a network model of household photovoltaic substations, construct a target optimization model of the substation network model according to the optimization target, and set constraints for the target optimization model, wherein the optimization target includes at least one of no-limit node voltage of the low-voltage substation network, minimum total network loss, minimum three-phase imbalance, and reactive power self-balancing; According to the constraints and the target optimization model, the reactive power adjustment value is obtained as the optimization result, and a coherent control strategy is adopted to adjust the reactive power of the continuous time section of the household photovoltaic area, including the following: Determine whether the reactive power of the household photovoltaic area is adjusted for the first time in the current time section; If it is the first adjustment, then the target optimization models of each optimization target are obtained, and according to each target optimization model, the overall target optimization model of the current time section is obtained, and the optimization result of the reactive power of the current time section is output; If it is not the first adjustment, the optimization result of the reactive power output by the total target optimization model of the previous time section is obtained, the optimization result of the reactive power of the previous time section is used as the input of the total target optimization model of the current time section, and the optimization result of the reactive power of the current time section is output; The optimization result of the reactive power of the current time section is used as the input of the overall target optimization model of the next time section, and the overall target optimization model is updated in a rolling manner to achieve coherent control and optimization of the reactive power of household photovoltaic stations in continuous time sections.

2. The household photovoltaic area reactive power control method according to claim 1, characterized in that: The reactive power adjustment value is obtained according to the constraint conditions and the target optimization model as the optimization result to adjust the reactive power of the household photovoltaic area, wherein the variable weight principle is introduced, including: setting weight rules according to different weights corresponding to different working conditions of the preset household photovoltaic area, determining the weights of the working conditions corresponding to each target optimization model at the current time section according to the working conditions of the household photovoltaic area at the current time section, wherein the working conditions of the household photovoltaic area include load demand and photovoltaic output; and obtaining the overall target optimization model according to each target optimization model and the weight of each target optimization model.

3. The household photovoltaic area reactive power control method according to claim 1, characterized in that: The constraint conditions include at least one of the upper and lower limits of the voltage of each phase of each node in the photovoltaic area, the upper limit constraint of the current allowed to flow through each branch, the network flow constraint, and the household photovoltaic inverter regulation constraint.

4. The household photovoltaic area reactive power control method according to claim 3, characterized in that: The optimization objectives include no over-limit node voltage in the low-voltage substation network, minimum total network loss, minimum three-phase imbalance and reactive power self-balancing, and the target optimization model constructed by this is a multi-objective optimization model.

5. The household photovoltaic area reactive power control method according to claim 4, characterized in that: The constraints include upper and lower limits of the voltage of each phase at each node in the photovoltaic area, upper limit constraints of the current allowed to flow through each branch, network flow constraints, and household photovoltaic inverter regulation constraints.

6. The household photovoltaic area reactive power control method according to claim 1, characterized in that: The network model of the household photovoltaic area is established based on the network topology structure, line parameters, household photovoltaic installation location and quantity, and household photovoltaic access method of the household photovoltaic low-voltage area.

7. The method for controlling reactive power of household photovoltaic stations according to claim 1, characterized in that: The reactive power adjustment value is obtained according to the constraint conditions and the target optimization model as the optimization result. Based on the constraint conditions, the overall target optimization model is solved by the NSGA-III algorithm to obtain the power adjustment value as the optimization result.

8. The method for controlling reactive power of household photovoltaic stations according to claim 2, characterized in that: The optimization result of reactive power is obtained according to the constraint conditions and the target optimization model, and further includes: introducing the objective function of adjusting the penalty of the relative change amount of reactive power of household photovoltaic stations in continuous time sections, and updating the overall target optimization model, including: When the relative change of reactive power in the continuous time section of the household photovoltaic area is greater than the preset value, a regulation penalty function is obtained to encourage a smooth change of reactive power; According to the preset different weights corresponding to different working conditions of household photovoltaic areas, the weight of the current time section adjustment penalty function is determined according to the working conditions of the household photovoltaic area at the current time section; The overall objective optimization model is obtained according to each objective optimization model, adjusting the penalty function, and each weight.

9. The household photovoltaic area reactive power control method according to claim 2, characterized in that: The weight rule is established by the following steps: Count the number of nodes in the household photovoltaic area where the photovoltaic output is greater than the load demand; Generate judgment variables based on the number of nodes obtained by statistics; The weight rule is generated according to the judgment variable.

10. A household photovoltaic area reactive power control device, executing the control method according to any one of claims 1 to 9, characterized in that: include: A model building module is used to establish a network model of a household photovoltaic area, build a target optimization model of the area network model according to the optimization target, and set constraints for the target optimization model, wherein the optimization target includes at least one of the following: no voltage exceeding the limit of the node of the low-voltage area network, minimum total network loss, minimum three-phase imbalance, and reactive self-balancing; The regulation module is used to obtain the adjustment value of reactive power according to the constraint conditions and the target optimization model as the optimization result, and adopts a coherent control strategy to adjust the reactive power of the continuous time section of the household photovoltaic area, including the following: A judgment submodule is used to judge whether the reactive power of the household photovoltaic area is adjusted for the first time in the current time section; The first optimization result output submodule is used to obtain each target optimization model of each optimization target if it is the first adjustment, and obtain the overall target optimization model of the current time section according to each target optimization model, and output the optimization result of reactive power of the current time section; The second optimization result output submodule, if it is not the first adjustment, obtains the optimization result of reactive power output by the total target optimization model of the previous time section, uses the optimization result of reactive power of the previous time section as the input of the total target optimization model of the current time section, and outputs the optimization result of reactive power of the current time section; The rolling update submodule is used to use the optimization result of the reactive power of the current time section as the input of the overall target optimization model of the next time section, and to update the overall target optimization model in a rolling manner to achieve coherent control and optimization of the reactive power of household photovoltaic stations in continuous time sections.

Images