Distributed photovoltaic power distribution network voltage difference adjustment coefficient optimization method based on cluster partition

Abstract

The invention provides a distributed photovoltaic power distribution network voltage difference adjustment coefficient optimization method based on cluster partitioning. The method comprises the following steps: step 1, analyzing the voltage regulation and control capability of a photovoltaic power grid node; 2, constructing a load flow calculation model containing a voltage difference adjustment coefficient node according to an analysis result, and calculating to obtain photovoltaic output under different loads based on the load flow calculation model containing the voltage difference adjustment coefficient node; and 3, constructing a distributed photovoltaic voltage difference adjustment coefficient optimization setting model, and solving the distributed photovoltaic voltage difference adjustment coefficient optimization setting model by adopting an interior point method in combination with photovoltaic output under different loads to obtain a distributed photovoltaic power distribution network voltage difference adjustment coefficient optimization result. When adjustment control is carried out based on the partitions, reactive voltage adjustment can be carried out by taking the partitions as units, the control range is narrowed, mutual influence among the partitions is limited, photovoltaic output is more convenient to coordinate, and the requirements of reactive optimization accuracy and rapidity are better met.

Description

Technical Field

[0001] The invention relates to a method for optimizing voltage regulation coefficient of a distributed photovoltaic power distribution network based on cluster partitioning, and belongs to the technical field of optimized operation of power distribution networks. Background Art

[0002] Distributed photovoltaics occupy a major position among many renewable energy sources because of their advantages of improving the proportion of energy structure through large-scale access to distribution networks. The number of photovoltaic power sources connected to distribution networks has increased dramatically, and large-scale high-penetration distributed photovoltaics connected to distribution networks will bring many risks. Distributed photovoltaic output has the characteristics of strong dependence on the external environment, and the output has a strong randomness; the influence of photovoltaic access location and the mismatch between photovoltaic output and load capacity; in addition, the distribution network currently lacks continuously adjustable reactive backup resources. The above problems will affect the voltage of the distribution network, which may easily lead to the voltage of the distribution network exceeding the limit, becoming an important reason affecting the absorption of photovoltaics. Therefore, studying the voltage control problem in the distribution network with a high proportion of distributed photovoltaics has become an urgent problem to be solved.

[0003] When the active output of the photovoltaic power generation system does not reach the rated capacity, the photovoltaic inverter has a certain continuous reactive regulation resource. Therefore, the reactive compensation capability of photovoltaic can be used to adjust the system voltage, which has a positive effect on improving the voltage quality of the distribution network. At present, some people have proposed to participate in the reactive voltage control of the distribution network through the reactive support capability of distributed photovoltaics, but the method of photovoltaic participation in voltage regulation does not make more flexible and full use of the voltage regulation capability of photovoltaics itself to play the role of reactive voltage support for the power system in various scenarios. Therefore, it is possible to propose the optimization of the reactive voltage adjustment coefficient of the photovoltaic node, give full play to the reactive surplus of photovoltaics, so that photovoltaics can autonomously adjust reactive power according to external voltage fluctuations. In addition, there is reactive voltage control based on variable droop coefficient, but the voltage adjustment coefficient should not be changed frequently, so the adjustment coefficient suitable for various scenarios should be selected. In addition, with the increase in the number of distributed power sources connected to the grid, the topology of the distribution network has become more complex, the dimension of the control quantity has increased dramatically, and there are large differences in different regions. Zoning optimization is more conducive to meeting the requirements of reactive optimization accuracy and rapidity.

[0004] Defects and deficiencies of existing technologies: The above-mentioned methods for photovoltaic participation in voltage regulation do not make better use of the voltage regulation capability of photovoltaic itself to play a role in supporting the reactive voltage of the power system. Therefore, it is possible to propose optimization of the reactive voltage adjustment coefficient of photovoltaic nodes, give full play to the reactive surplus of photovoltaics, and enable photovoltaics to autonomously adjust reactive power according to external voltage fluctuations. In addition, the voltage adjustment coefficient should not be changed frequently, so the adjustment coefficient applicable to a variety of scenarios should be selected. In addition, with the increase in the number of distributed power sources connected to the grid, the topology of the distribution network has become more and more complex, the dimension of the control quantity has increased dramatically, and there are large differences between different regions. Zoning optimization is more conducive to meeting the requirements of accuracy and rapidity of reactive optimization. Summary of the invention

[0005] In order to solve the above technical problems, the present invention further proposes a method for optimizing the voltage regulation coefficient of a distributed photovoltaic distribution network based on cluster partitioning.

[0006] The technical solution adopted by the present invention to solve the above problems is: the present invention proposes a method for optimizing the voltage regulation coefficient of a distributed photovoltaic distribution network based on cluster partitioning, comprising:

[0007] Step 1: Analyze the voltage regulation capability of photovoltaics on grid nodes;

[0008] Step 2: Based on the analysis results, a power flow calculation model including voltage regulation coefficient nodes is constructed, and the photovoltaic output under different loads is calculated based on the power flow calculation model including voltage regulation coefficient nodes;

[0009] Step 3: Construct a distributed photovoltaic voltage regulation coefficient optimization setting model. Combined with the photovoltaic output under different loads, the distributed photovoltaic voltage regulation coefficient optimization setting model is solved using the interior point method to obtain the optimization result of the distributed photovoltaic distribution network voltage regulation coefficient.

[0010] Optionally, the analysis of the voltage regulation capability of photovoltaic power to the grid node in step 1 specifically includes:

[0011] The photovoltaic grid-connected point is defined as the PX node, and the PX node is used as the adjustment coefficient node to obtain the known quantities and unknown quantities of the PQ node, PV node, balancing node and PX node in the power flow calculation; among them, the known quantities of the PQ node are active power and reactive power, and the unknown quantities are voltage amplitude and phase angle, the known quantities of the PV node are active power and voltage amplitude, and the unknown quantities are reactive power and voltage phase angle, the known quantities of the balancing node are voltage amplitude and phase angle, and the unknown quantities are active power and reactive power, the known quantities of the PX node are active power and reactive voltage adjustment coefficient, and the unknown quantities are reactive power, voltage amplitude and phase angle.

[0012] Optionally, the step of calculating the photovoltaic output under different loads in step 2 includes:

[0013] Step 2.1: The first non-PV grid-connected node in the power grid is taken as a balancing node, where the power grid includes x PX nodes and nx-1 PQ nodes;

[0014] Step 2.2: Define the PX node according to the relationship between the reactive power Q and the voltage U of the photovoltaic grid-connected point, and obtain the polar coordinate form of the PX node;

[0015] Step 2.3: In the polar coordinate form of the X node, calculate the power flow correction equation containing the node with the regulation coefficient. On the basis of the selected regulation coefficient, obtain the photovoltaic output under different loads through power flow calculation, and perform voltage crossing of the limit node of the power grid.

[0016] Optionally, the step of obtaining the photovoltaic output under different loads by power flow calculation in step 2.3 includes:

[0017] Step 2.3.1: Calculate the power equation of the PX node in polar coordinates, derive the relationship between the power equation of the PX node in polar coordinates and the reactive power Q and voltage U, and obtain the nonlinear equation of the PX node;

[0018] Step 2.3.3: Obtain the active unbalance equation of the PX node, and construct the power flow correction equation containing the adjustment coefficient node by combining the nonlinear equation of the PX node;

[0019] Step 2.3.3: Solve the nonlinear equation of the PX node to obtain and

[0020] Step 2.3.4: and It is used as an element of the power flow correction equation containing the node with the adjustment coefficient to iteratively solve and obtain the photovoltaic output under different loads;

[0021] The power equation in polar coordinate form at the PX node is:

[0022]

[0023] The relationship between reactive power Q and voltage U is:

[0024]

[0025] The nonlinear equation of the PX node is:

[0026]

[0027] The active unbalance equation of the PX node is:

[0028]

[0029] The power flow correction equation containing the adjustment coefficient node is:

[0030]

[0031] The expression of the elements of the power flow correction equation containing the adjustment coefficient node is:

[0032]

[0033] Optionally, the step of obtaining the optimization result of the voltage regulation coefficient of the distributed photovoltaic distribution network in step 3 includes:

[0034] Step 3.1: Based on the power flow calculation model with voltage regulation coefficient nodes, the photovoltaic output under different loads is obtained, the objective function is set, and the voltage of the internal nodes of the power grid under different operation scenarios is optimized so that the voltage of the internal nodes of the power grid under all operation scenarios is close to the rated value of the power grid voltage;

[0035] Step 3.2: Set the constraints for different partitions in different scenarios of the power grid;

[0036] Step 3.3: Set coupling constraints under different scenarios based on the constraints of different partitions in different grid scenarios to build large-scale distributed photovoltaic grid-connected scenarios;

[0037] Step 3.4: Use the interior point method to solve the nonlinear optimization of the large-scale distributed photovoltaic grid-connected scenario and obtain the optimization result of the voltage regulation coefficient of the distributed photovoltaic distribution network;

[0038] The expression of the objective function is:

[0039]

[0040] In formula (7), i is the node number, n is the number of nodes in the interval, t is the operation scenario, and m is the total number of different operation scenarios. is the probability of the t-th operation mode occurring, is the node voltage value of node i in interval k under the tth operation scenario, U iref is the grid voltage reference value;

[0041] The expressions of coupling constraints in different scenarios are:

[0042] X 1 =X 2 =...=X t =...=X m (8);

[0043] In formula (8), t is the tth operation scenario, m is the total number of scenarios with different operation conditions; X t is the adjustment coefficient under the tth operation scenario.

[0044] Optionally, the constraint conditions for different partitions in different power grid scenarios set in step 3.2 include equality constraints and inequality constraints;

[0045] The equality constraints specifically include: Taking the grid-side voltage U0 as the voltage reference value V ref , X is used as the adjustment coefficient to calculate the power balance equation constraint;

[0046] The inequality constraints specifically include: photovoltaic power supply adjustment coefficient constraint, photovoltaic reactive power output subject to inverter capacity constraint, node voltage constraint and line transmission power constraint on the section interconnection line;

[0047] The expression of the power balance equality constraint is:

[0048]

[0049] The expression of photovoltaic power source regulation coefficient constraint is:

[0050] X i.min ≤X i ≤X i.max (10);

[0051] The expression of photovoltaic reactive output constrained by inverter capacity is:

[0052]

[0053] In formula (11), S PV is the rated capacity of the photovoltaic inverter, P PV is the active power output by the inverter at this moment;

[0054] The expression of node voltage constraint is:

[0055] U i.min ≤U i ≤U i.max (12);

[0056] The expression of line transmission power constraint on the section interconnection line is:

[0057] S ij.min ≤S ij ≤S ij.max (13);

[0058] In formula (13), S ij , S ij.min , S ij.max The two zone connection lines L ij The transmission power and amplitude of the line between the two end nodes i and j, i and j are nodes in different intervals, and the line L ij It is the connecting line between the two areas.

[0059] Optionally, obtaining the node voltage constraint specifically includes:

[0060] When the reactive power of the power source reaches the limit, the node type in the power flow calculation is converted from PX node to PQ node, the reactive injection power of the node is specified as the reactive upper limit value, and the node voltage constraint is calculated based on the reactive upper limit value.

[0061] The beneficial effects of the present invention are:

[0062] (1) The present invention controls the relationship between photovoltaic output and node voltage through the reactive-voltage adjustment coefficient, which can make full use of the voltage adjustment capability of photovoltaic itself, give full play to the reactive power surplus of photovoltaic, and take into account the uncertainty of its operating power and load, so that photovoltaic can autonomously adjust reactive power according to external voltage fluctuations; optimize and select voltage adjustment coefficients suitable for various scenarios, avoid frequent adjustment of voltage adjustment coefficients in different scenarios, and only need to adjust the reactive power output of the node based on the selected adjustment coefficient to control the node voltage, which can solve the voltage limit problem in various different scenarios;

[0063] (2) When adjusting and controlling based on partitions, the present invention can adjust reactive voltage based on partitions, narrow the control range, limit the mutual influence between partitions, make it easier to coordinate photovoltaic output, and be more conducive to meeting the requirements of reactive optimization accuracy and speed; in addition, it can also reduce the power flow on the branch and avoid power loss caused by long-distance transmission. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 A flow chart of a method for optimizing voltage regulation coefficient of a distributed photovoltaic distribution network based on cluster partitioning provided by the present invention;

[0065] Figure 2 A comparison diagram of grid node voltages of the simulation schemes S1 and S2 provided by the present invention;

[0066] Figure 3 A comparison diagram of reactive power loss in the lower branch of the simulation solutions S1 and S2 provided by the present invention;

[0067] Figure 4 A schematic diagram of the IEEE 33 node system partitioning provided by the present invention;

[0068] Figure 5 A comparison diagram of reactive power loss in the lower branches of the simulation schemes S2 and S3 provided by the present invention. DETAILED DESCRIPTION

[0069] Combination Figure 1 The present embodiment is described. The steps of the method for optimizing the voltage regulation coefficient of a distributed photovoltaic distribution network based on cluster partitioning described in the present embodiment include:

[0070] Step 1: In-depth analysis of the impact of large-scale distributed photovoltaic grid-connected capacity, access location, load capacity and other factors on the distribution network voltage;

[0071] In the power grid, common nodes are divided into three categories, namely PQ nodes, PV nodes and balance nodes. Since the reactive power-voltage relationship of photovoltaic power is determined by the set adjustment coefficient, the photovoltaic grid-connected point is abstractly defined as a new node type - adjustment coefficient node, that is, PX node. The known quantities of this node type in the power flow calculation are the active power P of the photovoltaic power source and the reactive voltage adjustment coefficient X, and the quantities to be calculated are the reactive power Q, the voltage amplitude V and the phase angle.

[0072] The known quantities and unknown quantities of each node type are shown in Table 1:

[0073] Table 1

[0074]

[0075] Step 2: Construct a power flow calculation model including voltage regulation coefficient nodes;

[0076] Step 2.1: Assume that there are n nodes in the system, among which the first non-PV grid-connected node is taken as the balancing node; x PV grid-connected nodes with differential regulation characteristics are included as PX nodes; nx-1 PQ nodes; the relationship between reactive power Q and voltage U shown by the PV grid-connected node is defined as the PX node, and the PX node is represented by polar coordinates:

[0077] Step 2.2: Derived power equation in polar coordinate form for PX node And the relationship between Q and U The nonlinear equation of the PX node is:

[0078]

[0079] Step 2.3: Get the active unbalance equation of the PX node:

[0080]

[0081] In the polar coordinate situation of the PX node, obtain the power flow correction equation containing the adjustment coefficient node:

[0082]

[0083] The elements in H, N, J, and L are the same as the elements of the Jacobian matrix in the traditional method, and R and S are the correction matrices of the node equilibrium equation of the adjustment coefficient. The values ​​of each element in the Jacobian matrix are:

[0084]

[0085] For the PX node, we can get and Will and The correction value is added into the power flow correction equation to solve it by repeated iteration.

[0086] The present invention controls the relationship between photovoltaic output and node voltage through the reactive-voltage adjustment coefficient, can make full use of the voltage adjustment capability of photovoltaic itself, give full play to the reactive power surplus of photovoltaic, and take into account the uncertainty of its operating power and load, so that photovoltaic can autonomously adjust reactive power according to external voltage fluctuations; optimize and select the voltage adjustment coefficient suitable for a variety of scenarios, avoid the frequent adjustment of the voltage adjustment coefficient in different scenarios, only need to adjust the reactive power output of the node based on the selected adjustment coefficient to control the node voltage, can solve the voltage limit problem in a variety of different scenarios.

[0087] Step 2.4: Use the modified Jacobian matrix to perform Newton-Raphson method power flow calculation. By constructing the PX node to simplify the calculation of the photovoltaic grid-connected node output reactive power and node voltage, the photovoltaic grid-connected node is added to the power flow calculation model. On the basis of selecting the differential coefficient, the photovoltaic output under different loads is obtained through power flow calculation to achieve the purpose of limiting the node voltage over-limit.

[0088] Step 3: The reactive voltage regulation coefficient setting model and algorithm of photovoltaic power source based on partition;

[0089] Since the source-load characteristics of different regions are different, and reactive voltage setting is a local adjustment problem, it is necessary to adjust the balance on the spot by layered partitions. The partition setting method can be adopted. Based on the node electrical distance partition method, the power constraint of the interval interconnection line is established in the constraint conditions to achieve the purpose of limiting the influence of mutual connection between partitions. In addition, the coupling constraints of different scenarios are established so that the selected adjustment coefficient is suitable for all scenarios.

[0090] Step 3.1: Construct the optimization model of reactive voltage regulation coefficient: In order to maintain the internal node voltage of the interval within the required range, as close as possible to the rated value of the grid voltage, and reduce the voltage fluctuation of the grid connection point, the objective function is set to optimize the internal node voltage of the interval;

[0091] The expression of the objective function is:

[0092]

[0093] In formula (5), i is the node number, n is the number of nodes in the interval, t is the operation scenario, and m is the total number of different operation scenarios. is the probability of the t-th operation mode occurring, is the node voltage value of node i in interval k under the tth operation scenario, U iref is the grid voltage reference value.

[0094] Step 3.2: Set constraints;

[0095] The coordinated setting model of the reactive power and voltage regulation system contains both equality constraints that describe the power balance relationship of the power grid and inequality constraints that characterize the operating requirements of power flow variables and control parameters. The constraints in different partitions are the same in different scenarios. The following describes a partition in a scenario:

[0096] Step 3.1: Equality constraints;

[0097] Assume that there are n nodes in the system, of which the first non-PV grid-connected node is used as a balancing node; x PV grid-connected nodes with differential regulation characteristics are included as PX nodes (i.e., differential regulation nodes), and nx-1 PQ nodes. The grid-side voltage U0 is used as the voltage reference value V ref ; X is the reactance that determines the reactive voltage droop characteristics, which is the adjustment coefficient; after sorting, the power balance equation constraint can be obtained as follows:

[0098]

[0099] In formula (6), P i , Q i are the injected active and reactive power of node i, U i , U j is the node voltage, G ij , B ij , δ ij is the conductance, susceptance and voltage phase difference between nodes i and j;

[0100] Step 3.2: Inequality constraints;

[0101] The constraint range of the regulation coefficient of the photovoltaic power source is:

[0102] X i.min ≤X i ≤X i.max (7);

[0103] Photovoltaic reactive power output is constrained by inverter capacity:

[0104]

[0105] In formula (8), S PV is the rated capacity of the photovoltaic inverter, P PV It is the active power output by the inverter at this moment; the reactive power that the photovoltaic inverter can generate is limited by the rated capacity.

[0106] When the reactive power generated by the power supply reaches its limit, it will be difficult for the power supply to continue to adjust the output reactive power based on the voltage feedback signal. The node type in the power flow calculation will be changed from a differential node to a PQ node. At this time, the reactive injection power of the node can be specified as the reactive upper or lower limit.

[0107] Node voltage constraints:

[0108] U i.min ≤U i ≤U i.max (9);

[0109] In order to minimize the mutual impact between the partitions, the transmission power constraints on the section interconnection lines are further restricted:

[0110] S ij.min ≤S ij ≤S ij.max (10);

[0111] In formula (10), S ij , S ij.min , S ij.max The two zone connection lines L ij The transmission power and amplitude of the line between the two end nodes i and j, where i and j are nodes in different intervals, and the line L ij It is the connecting line between the two intervals.

[0112] Step 3.3: The coupling constraints in different scenarios are:

[0113] X 1 =X 2 =...=X t =...=X m (11)

[0114] In formula (11), t is the tth operation scenario, m is the total number of scenarios with different operation conditions; X t is the adjustment coefficient under the tth operation scenario.

[0115] When adjusting and controlling based on partitions, the present invention can adjust reactive voltage based on partitions, narrow the control range, limit the mutual influence between partitions, make it easier to coordinate photovoltaic output, and be more conducive to meeting the requirements of reactive optimization accuracy and speed; in addition, it can reduce power flow on branches and avoid power loss caused by long-distance transmission.

[0116] Step 3.4: Optimization algorithm: The large-scale distributed photovoltaic grid-connected scenario constructed by formula (5) to formula (11) is a nonlinear optimization problem, which can be solved by the interior point method to obtain the optimization result of the voltage regulation coefficient of the distributed photovoltaic distribution network.

[0117] Example

[0118] Combination Figure 2-5 This embodiment is described. To verify the technical effect of the present invention, this embodiment performs the following example analysis:

[0119] Based on the IEEE E-33 node distribution network standard example, it is improved to build a large-scale photovoltaic grid-connected system.

[0120] Through simulation analysis, the influence of factors such as grid-connected capacity, access location, photovoltaic output, etc. of a large number of distributed photovoltaic access on the voltage of the entire distribution network under different scenarios is analyzed; the influence of the regulation coefficient on the reactive voltage of the system is analyzed, and the feasibility of optimizing the regulation coefficient based on zoning conditions is analyzed. The optimized regulation coefficient of each photovoltaic node is compared with the simple setting.

[0121] 1. Example parameter setting;

[0122] Based on the standard IEEE-33 node system, photovoltaic power is connected at nodes 5, 8, 9, 10, 12, 13, 14, 16, 17, 18, 19, 20, 21, 22, 24, 25, 28, 29, 30, 32, and 33. The node reference voltage is taken as U B =12.66kV, three-phase power reference value S B =100MVA.

[0123] The total number of points of the constructed large-scale distributed photovoltaic grid-connected system is 33. Node 1 in the system is set as a balancing node, and the numbers of each power source and load are the original numbers of the standard example.

[0124] 2. Analysis of the impact of the regulation coefficient on the system reactive voltage;

[0125] In order to analyze the effectiveness of the voltage regulation method using the differential coefficient, simulation schemes S1 and S2 were designed and simulation calculations were carried out.

[0126] S1: Based on the IEEE33 node calculation data in 4.1, the voltage regulation coefficient of each node is not optimized and adjusted, and the regulation coefficient of each photovoltaic grid-connected node is uniformly set to X = -0.06;

[0127] S2: Based on the IEEE33 node calculation example data in 4.1, without partitioning, the voltage of each node is optimized according to the optimization model constructed by formulas (5) to (9) in the specific implementation method, and the appropriate adjustment coefficient value is obtained for calculation and analysis.

[0128] Figure 2The simulation results of node voltage in the S1 and S2 systems are shown in Figure 1. From the scatter distribution of the voltage amplitude of each node in the figure, it can be seen that when the regulation coefficient is not optimized, the fluctuation of the voltage of each node is more obvious; when the optimized regulation coefficient is used for power flow calculation, the voltage per unit value of each node is closer to 1. By comparison, it can be seen that by optimizing the node voltage regulation coefficient, the voltage fluctuation of most nodes in the system can be effectively controlled.

[0129] Figure 3 The results of the per-unit reactive power loss of the system branches of simulation schemes S1 and S2 are shown in Figure 2. It is obvious that the system branch loss after optimizing the adjustment coefficient is significantly lower than that of simulation scheme S1, all of which are lower than 0.02, and most of which are close to zero, which proves that adjusting the adjustment coefficient control can effectively reduce the system power loss. This is because the photovoltaic surplus capacity is effectively used to adjust the voltage of the photovoltaic node and the nearby nodes, reducing the line power transmission.

[0130] Table 2 compares the output of the balancing node in the simulation results of the two cases. When the optimized adjustment coefficient is used for power flow calculation, the output of the balancing node is smaller. This proves that by optimizing the adjustment coefficient, the photovoltaic residual capacity can be maximized for reactive voltage control, limiting the voltage fluctuation of the photovoltaic grid-connected node and nearby nodes; thereby reducing the output of the balancing node.

[0131] Table 2

[0132]

[0133] 3. Analysis of setting effect of reactive voltage regulation coefficient;

[0134] In order to analyze the impact and feasibility of the proposed partition-based regulation coefficient optimization method on node voltage and line power loss, a simulation scheme S3 was designed, and the simulation calculation results of schemes S1, S2 and S3 were compared.

[0135] S3: Based on the IEEE33-node example data in 4.1, the partition-based reactive power optimization method proposed in this chapter is adopted. The partition results are as follows: Figure 4 As shown, on this basis, an optimization model is constructed according to formulas (5) to (11) in the specific implementation method. The simulation uses per-unit value for calculation, and sets the voltage constraint of formula (9) to 0.95-1.05 to limit the node voltage fluctuation range; the transmission power limit on the interconnection line between partitions in formula (10) is set as small as possible to limit the connection and influence between partitions, highlighting the partition-based optimization model. The scenario probability is calculated based on the K-means clustering algorithm, and the optimization calculation is performed to select the optimal adjustment coefficient for each photovoltaic grid-connected point to cope with a variety of different operating conditions.

[0136] Table 3 compares the node voltages of nodes (nodes 7-12) in the selected partition 4 under the three simulation schemes S1, S2 and S3. By comparing the per-unit values ​​of the node voltages in Table 3, it can be seen that under the simulation scheme S3, the number of nodes with a per-unit voltage value of 1 is the largest. The adjustment coefficient after partition optimization can most effectively limit the problem of node voltage fluctuations, better coordinate the relationship between photovoltaic output and node voltage, and control the node voltage near the voltage reference value.

[0137] Table 3

[0138]

[0139] according to Figure 5 The results of the two simulation schemes are compared and analyzed. Under the two simulation schemes, the losses of most branches of the system are close to zero; however, in comparison, the number of lossy branches and the loss of active and reactive power under the control of simulation scheme S2 are greater than those under S3. Therefore, the optimization of the adjustment coefficient based on the partition can more effectively reduce the power loss of the branch. This is because the partition control reduces the long-distance transmission of power, the nodes within the interval can coordinate and control each other, and the mutual influence between the partitions is small, thereby reducing the power loss of the interval connection line and the internal branch of the partition.

[0140] 4. Applicability of optimization results under various operating conditions;

[0141] Since there are a large number of analysis results considering multiple operating conditions, this embodiment selects simulation results of internal nodes of partition 4 under two different operating conditions for analysis.

[0142] It can be seen from Table 4 that under simulation scheme S3, the voltage amplitude per unit value of each node in the partition under different operating conditions deviates less from 1. Therefore, the optimization of the regulation coefficient based on the partition can more effectively select the regulation coefficient suitable for different operating conditions, more effectively control the photovoltaic output, use the photovoltaic remaining capacity to deal with a variety of different operating conditions, and limit the voltage fluctuation of each node in the partition.

[0143] Table 4

[0144]

[0145] The above is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technician familiar with this profession can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent replacement and improvement made to the above embodiments without departing from the content of the technical solution of the present invention, based on the technical essence of the present invention, within the spirit and principles of the present invention, still fall within the protection scope of the technical solution of the present invention.

Claims

1. A method for optimizing voltage regulation coefficient of distributed photovoltaic distribution network based on cluster partitioning, characterized in that: The steps of the distributed photovoltaic distribution network voltage regulation coefficient optimization method based on cluster partitioning include: Step 1: Analyze the voltage regulation capability of photovoltaics on grid nodes; Step 2: construct a power flow calculation model including a voltage differential coefficient node according to the analysis results, and calculate the photovoltaic output under different loads based on the power flow calculation model including the voltage differential coefficient node; Step 3: Construct a distributed photovoltaic voltage regulation coefficient optimization setting model. Combined with the photovoltaic output under different loads, the distributed photovoltaic voltage regulation coefficient optimization setting model is solved using the interior point method to obtain the optimization result of the distributed photovoltaic distribution network voltage regulation coefficient.

2. The method for optimizing voltage regulation coefficient of distributed photovoltaic distribution network based on cluster partitioning according to claim 1 is characterized in that: The analysis of the voltage regulation capability of photovoltaic power to grid nodes in step 1 specifically includes: The photovoltaic grid-connected point is defined as the PX node, and the PX node is used as the adjustment coefficient node to obtain the known quantities and unknown quantities of the PQ node, PV node, balancing node and PX node in the power flow calculation; among them, the known quantities of the PQ node are active power and reactive power, and the unknown quantities are voltage amplitude and phase angle, the known quantities of the PV node are active power and voltage amplitude, and the unknown quantities are reactive power and voltage phase angle, the known quantities of the balancing node are voltage amplitude and phase angle, and the unknown quantities are active power and reactive power, the known quantities of the PX node are active power and reactive voltage adjustment coefficient, and the unknown quantities are reactive power, voltage amplitude and phase angle.

3. The method for optimizing voltage regulation coefficient of distributed photovoltaic distribution network based on cluster partitioning according to claim 1 is characterized in that: The steps for calculating the photovoltaic output under different loads in step 2 include: Step 2.1: The first non-PV grid-connected node in the power grid is taken as a balancing node, where the power grid includes x PX nodes and nx-1 PQ nodes; Step 2.2: Define the PX node according to the relationship between the reactive power Q and the voltage U of the photovoltaic grid-connected point, and obtain the polar coordinate form of the PX node; Step 2.3: In the polar coordinate form of the X node, calculate the power flow correction equation containing the node with the regulation coefficient. On the basis of the selected regulation coefficient, obtain the photovoltaic output under different loads through power flow calculation, and perform voltage crossing of the limit node of the power grid.

4. The method for optimizing voltage regulation coefficient of distributed photovoltaic distribution network based on cluster partitioning according to claim 3 is characterized in that: The steps of obtaining the photovoltaic output under different loads through power flow calculation in step 2.3 include: Step 2.3.1: Calculate the power equation of the PX node in polar coordinates, derive the relationship between the power equation of the PX node in polar coordinates and the reactive power Q and voltage U, and obtain the nonlinear equation of the PX node; Step 2.3.3: Obtain the active unbalance equation of the PX node, and construct the power flow correction equation containing the adjustment coefficient node by combining the nonlinear equation of the PX node; Step 2.3.3: Solve the nonlinear equation of the PX node to obtain and Step 2.3.4: and It is used as an element of the power flow correction equation containing the node with the adjustment coefficient to iteratively solve and obtain the photovoltaic output under different loads; The power equation in polar coordinate form at the PX node is: The relationship between reactive power Q and voltage U is: The nonlinear equation of the PX node is: The active unbalance equation of the PX node is: The power flow correction equation containing the adjustment coefficient node is: The expression of the elements of the power flow correction equation containing the adjustment coefficient node is:

5. The method for optimizing voltage regulation coefficient of distributed photovoltaic distribution network based on cluster partitioning according to claim 1 is characterized in that: The steps of obtaining the optimization result of the voltage regulation coefficient of the distributed photovoltaic distribution network in step 3 include: Step 3.1: Based on the power flow calculation model containing the voltage regulation coefficient node, the photovoltaic output under different loads is obtained, the objective function is set, and the voltage of the internal nodes of the power grid under different operation scenarios is optimized so that the voltage of the internal nodes of the power grid under all operation scenarios is close to the rated value of the power grid voltage; Step 3.2: Set the constraints for different partitions in different scenarios of the power grid; Step 3.3: Set coupling constraints under different scenarios based on the constraints of different partitions in different grid scenarios to build large-scale distributed photovoltaic grid-connected scenarios; Step 3.4: Use the interior point method to solve the nonlinear optimization of the large-scale distributed photovoltaic grid-connected scenario and obtain the optimization result of the voltage regulation coefficient of the distributed photovoltaic distribution network; The expression of the objective function is: In formula (7), i is the node number, n is the number of nodes in the interval, t is the operation scenario, and m is the total number of different operation scenarios. is the probability of the tth operation mode occurring, is the node voltage value of node i in interval k under the tth operation scenario, U iref is the grid voltage reference value; The expressions of coupling constraints in different scenarios are: X 1 =X 2 =...=X t =...=X m (8); In formula (8), t is the tth operation scenario, m is the total number of scenarios with different operation conditions; X t is the adjustment coefficient under the tth operation scenario.

6. The method for optimizing voltage regulation coefficient of distributed photovoltaic distribution network based on cluster partitioning according to claim 5 is characterized in that: In step 3.2, the constraints for different partitions in different power grid scenarios include equality constraints and inequality constraints; The equality constraints specifically include: Taking the grid-side voltage U0 as the voltage reference value V ref , X is used as the adjustment coefficient to calculate the power balance equation constraint; The inequality constraints specifically include: photovoltaic power supply adjustment coefficient constraint, photovoltaic reactive power output subject to inverter capacity constraint, node voltage constraint and line transmission power constraint on the section interconnection line; The expression of the power balance equality constraint is: The expression of photovoltaic power source regulation coefficient constraint is: X i.min ≤X i ≤X i.max (10); The expression of photovoltaic reactive output constrained by inverter capacity is: In formula (11), S PV is the rated capacity of the photovoltaic inverter, P PV is the active power output by the inverter at this moment; The expression of node voltage constraint is: IN i.min ≤U i ≤U i.max (12); The expression of line transmission power constraint on the section interconnection line is: S ij.min ≤S ij ≤S ij.max (13); In formula (13), S ij , S ij.min , S ij.max The two zone connection lines L ij The transmission power and amplitude of the line between the two end nodes i and j, i and j are nodes in different intervals, and the line L ij It is the connecting line between the two areas.

7. The method for optimizing voltage regulation coefficient of distributed photovoltaic distribution network based on cluster partitioning according to claim 6 is characterized in that: Obtaining node voltage constraints specifically includes: When the reactive power of the power source reaches the limit, the node type in the power flow calculation is converted from PX node to PQ node, the reactive injection power of the node is specified as the reactive upper limit value, and the node voltage constraint is calculated based on the reactive upper limit value.

Images