Rural power grid power distribution area electric energy quality optimization control method based on distributed optical energy storage system

Abstract

A rural power grid power distribution area electric energy quality optimization control method based on a distributed optical energy storage system belongs to the technical field of power systems, and comprises the following steps: firstly, according to different operation requirements of a power distribution network, realizing control targets of node voltage deviation, PV reduction amount and network loss minimization by adjusting active and reactive output of distributed PV; establishing a voltage optimization control objective function; secondly, the SOC relaxation technology is used for carrying out convex processing on the model, an original problem is converted into a convex problem, and due to the fact that the model is the convex model after relaxation, the optimality and the calculation efficiency of the solution can be guaranteed; and finally, constructing a distributed optimization control model based on the ADMM. According to the invention, active power distribution network voltage distributed optimization control containing distributed PV is realized, the voltage out-of-limit problem of the power distribution network can be effectively solved, the electric energy quality of the rural power distribution network can be improved, and the power supply reliability is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of power systems, and particularly relates to a method for optimizing and controlling the power quality of rural power distribution substations based on a distributed optical storage system. Background Technique

[0002] With the continuous advancement of the construction of a new power distribution system dominated by a high proportion of new energy, large-scale distributed renewable energy is connected to the rural power grid. The development of integrated energy and the development of rural power grids face opportunities and challenges respectively. Renewable energy in rural areas of our country is rich. A new rural energy system based on rooftop photovoltaic, combined with energy storage devices in the power distribution system, can explore a more efficient and high-quality power supply solution. Research on a new method for eliminating power quality problems at the end of the distribution network with large-scale distributed photovoltaics based on substation energy storage control, solving power quality problems such as bidirectional power fluctuations and imbalance caused by single-phase power generation and consumption, improving the power quality of rural distribution networks, and enhancing power supply reliability.

[0003] The rapid expansion of distributed power sources and the reduction of the proportion of non-clean energy have improved the green and low-carbon level of the energy industry. However, due to the intermittency and volatility of the output of renewable energy such as photovoltaics and wind power, and the spatio-temporal interleaving of renewable energy generation and load demand, new problems and challenges have been brought to the safe and reliable operation of the new power system. By reasonably controlling the energy storage system, the regulation potential of existing resources can be further improved, the dynamic regulation ability of the power grid can be enhanced, the power quality of rural power distribution substations can be improved, and the rural power grid can be better supported to adapt to the rapid development of distributed power sources. Therefore, the research on control strategies for energy storage grid-connected systems has become one of the key research fields for building a new power system.

[0004] In the field of control strategies for energy storage systems under the grid connection of distributed renewable energy, many scholars have carried out extensive and in-depth research. Among them, the virtual synchronous generator control of Hefei University of Technology realizes the bidirectional power conversion control of grid-connected inverters based on the traditional synchronous generator rotor equation, and has flexible adjustable inertia damping parameters [1] . For the control method of the energy storage system inverter, a virtual controller is proposed by combining virtual rotor, virtual primary, and virtual secondary control, and an extended inverter of the microgrid is introduced to stabilize / regulate the system frequency, but the influence of damping parameters on the transient stability of the system is ignored [2] . In order to ensure that the system has the active support ability of virtual inertia power under multi-disturbance operating conditions, a control algorithm for energy storage inverters with droop characteristics is designed, which can cope with sudden changes in various wind speeds and AC loads and suppress the frequency change on the AC side [3] . North China Electric Power University proposed a fuzzy inverter inertia adaptive control strategy that comprehensively considers fuzzy control and traditional inverter control. By analyzing the mechanism relationship between inertia and frequency change, fuzzy control rules are established [4]At present, most scholars conduct research on the inverter control strategy around the frequency regulation function of the energy storage inverter. However, few literatures attach importance to the regulatory effect of the operating state of the energy storage after grid connection on the inverter control strategy, and ignore the treatment effects of other power quality problems such as voltage fluctuation, three-phase imbalance, and voltage over-limit.

[0005] Therefore, the existing technology urgently needs a new technical solution to solve the above problems. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to provide a power quality optimization control method for rural power distribution substations based on a distributed optical energy storage system, which realizes the distributed optimization control of the voltage of the active distribution network containing distributed PV, can effectively solve the problem of voltage over-limit of the distribution network, and can improve the power quality of the rural distribution network and the power supply reliability.

[0007] A power quality optimization control method for rural power distribution substations based on a distributed optical energy storage system includes the following steps, and the following steps are carried out in sequence:

[0008] Step 1: According to the grid operation requirements, perform minimum control of the node voltage deviation, PV curtailment amount, and network loss by adjusting the active and reactive power outputs of the distributed PV, and establish the objective function and constraint conditions of the voltage optimization control model;

[0009] Step 2: Use the SOC relaxation technology to convexify the voltage optimization control model described in Step 1;

[0010] Step 3: According to the decomposition and coordination principle, partition the distribution network and decouple adjacent sub-regions;

[0011] Step 4: Based on the ADMM general consistency optimization method, construct a voltage distributed optimization control model for the convexified model.

[0012] The objective function of the voltage optimization control model described in Step 1 is:

[0013]

[0014] In the formula: N bus is the set of nodes in the distribution network; U n is the voltage amplitude of node n; assuming that the voltage amplitude of node 1 is the voltage reference value, U1 = 1 p.u.; N PV is the set of nodes in the distribution network with PV connected; Pmax PV,n and P PV，n are the maximum active power output of PV and the PV active power output of node n respectively; k: n → k represents the set of the end nodes of the branch with node n as the head node; r nk and l nkThey are the resistance and the square of the current amplitude of branch n-k respectively; ω1-ω3 are the minimization weight coefficients, all of which are greater than or equal to 0 and ω1 + ω2 + ω3 = 1; ξ1-ξ3 are correction coefficients greater than 0 to ensure that the three terms in the formula are of the same order of magnitude; the first term in the formula is the node voltage deviation, the second term is the PV curtailment, and the third term is the network loss. The node voltage deviation is non-linear.

[0015] The constraint conditions described in step 1 include the distribution network power flow constraint, the distribution network operation safety constraint, and the PV inverter control constraint.

[0016] The model after the convexification process described in step 2 is:

[0017]

[0018] In the formula, u n is the square of the voltage amplitude of node n;

[0019] The voltage distributed optimization control model described in step 3 is:

[0020]

[0021] In the formula: both f(x) and g(y) are convex functions; A, B, and c are coefficient matrices; the constraint condition Ax + By = c of variables x and y constitutes the feasible region of the variables in the ADMM objective function.

[0022] Through the above design scheme, the present invention can bring the following beneficial effects:

[0023] 1. The present invention establishes a voltage optimization control model, introduces the SOC relaxation technology to convexify the model, and ensures the optimality of the solution and the calculation efficiency;

[0024] 2. The present invention proposes a distributed optimization control model based on ADMM, decomposes the original variables and the objective function into two parts, and ensures the solvability of the optimization process;

[0025] 3. The present invention can provide a more perfect and sufficient theoretical identification system for the distribution personnel in the operation of the power system, and helps to improve the power quality of rural distribution networks. Description of the Drawings

[0026] The following further describes the present invention in conjunction with the drawings and specific embodiments:

[0027] Figure 1 It is a flow chart of a method for optimizing the power quality of a rural power distribution substation area based on a distributed optical energy storage system of the present invention;

[0028] Figure 2Specific implementation manner of a power quality optimization control method for rural power distribution substations based on a distributed optical energy storage system of the present invention; diagram of power flow model of distribution network branches

[0029] Figure 3 Specific implementation manner of a power quality optimization control method for rural power distribution substations based on a distributed optical energy storage system of the present invention; structure diagram of a 6-node radial distribution network

[0030] Figure 4 Specific implementation manner of a power quality optimization control method for rural power distribution substations based on a distributed optical energy storage system of the present invention; process diagram of the partition of a 6-node radial distribution network

[0031] Figure 5 Specific implementation manner of a power quality optimization control method for rural power distribution substations based on a distributed optical energy storage system of the present invention; diagram of independent optimization of sub-regions and interaction of boundary variables between sub-regions

[0032] Figure 6 Specific implementation manner of a power quality optimization control method for rural power distribution substations based on a distributed optical energy storage system of the present invention; schematic diagram of IEEE 33-node test system

[0033] Figure 7 Specific implementation manner of a power quality optimization control method for rural power distribution substations based on a distributed optical energy storage system of the present invention; diagrams of system node voltage before and after control in Scenario 1

[0034] Figure 8 Specific implementation manner of a power quality optimization control method for rural power distribution substations based on a distributed optical energy storage system of the present invention; diagrams of system node voltage before and after control in Scenario 2 Specific implementation manner

[0035] To make the objectives, technical solutions and advantages of the present invention clearer, the following further describes in detail the implementation manners of the present invention

[0036] To solve the defects and deficiencies existing in the control strategy of the energy storage system under the grid connection of distributed renewable energy in the background technology, the present invention proposes a power quality optimization control method for rural power distribution substations based on a distributed optical storage system. First, according to different operation requirements of the distribution network, the present invention intends to achieve the control objectives of minimum node voltage deviation, PV curtailment and network loss by adjusting the active and reactive power outputs of distributed PVs, and establishes a voltage optimization control objective function; then, uses the SOC relaxation technology to convexify the model and transforms the original problem into a convex problem. Since the relaxed model is a convex model, the optimality of the solution and the calculation efficiency can be guaranteed; finally, constructs a distributed optimization control model based on ADMM

[0037] A method for optimizing and controlling the power quality of a rural power distribution substation based on a distributed energy storage system, see Figure 1 , the method includes the following steps:

[0038] 101: According to different operating requirements of the distribution network, it is planned to adjust the active and reactive power outputs of the distributed PV to achieve the control objectives of minimum node voltage deviation, PV curtailment, and network loss, and establish a voltage optimization control objective function;

[0039] 102: Linearize the non-linear objective, introduce the equality relationship between the node voltage amplitude and its square, and use the SOC relaxation technique for convexification to ensure the optimality of the solution and the calculation efficiency;

[0040] Specifically, this step is: linearize the non-linear objective, introduce the equality relationship between the node voltage amplitude and its square, and use the SOC relaxation technique for convexification. The non-convex feasible region Coriginal of the original problem is relaxed into a convex cone feasible region CSOC, and the original problem is transformed into a convex problem;

[0041] The entire step 102 is the model convexification process. For the convexified model, when the original problem obtains the optimal solution, it can ensure that the relaxed equality is accurate enough to satisfy all the constraints of the original problem.

[0042] 103: According to the decomposition and coordination principle, partition the distribution network, and copy the sub-interval boundaries to adjacent sub-areas to achieve decoupling of adjacent sub-areas;

[0043] 104: Based on the ADMM general consistency optimization method, construct a voltage distributed optimization control model;

[0044] In summary, the embodiment of the present invention establishes a voltage optimization control model through the above steps 101-step 102, and uses the SOC relaxation technique to convexify the model to ensure the optimality of the solution; steps 103-step 104 construct a distributed optimization model based on ADMM.

[0045] The following combines Figures 2 - 8 and specific calculation formulas and examples to further introduce the solutions in the method of the present invention, as detailed in the following description:

[0046] 201: The established voltage optimization control objective function is:

[0047]

[0048] In the formula: N bus is the set of nodes in the distribution network; U n is the voltage amplitude of node n; it is assumed that the voltage amplitude of node 1 is the voltage reference value, that is, U1 = 1 p.u.; N PVis the set of nodes with PV connected in the distribution network; Pmax PV,n and P PV，n are the maximum active power output of PV and the active power output of PV at node n, respectively; k: n→k represents the set of end nodes of the branch with node n as the head node; r nk and l nk are the resistance and the square of the current amplitude of branch n-k, respectively; ω1—ω3 are the minimization weight coefficients, all greater than or equal to 0 and ω1 + ω2 + ω3 = 1; ξ1—ξ3 are correction coefficients greater than 0 to ensure that the three terms in the formula are of the same order of magnitude. The first term in the formula is the node voltage deviation, the second term is the PV curtailment, and the third term is the network loss. Obviously, the first term is non-linear.

[0049] Among them, the established constraint conditions include: distribution network power flow constraint, distribution network operation safety constraint, and PV inverter control constraint.

[0050] For the distribution network power flow model, see Figure 2 .

[0051] In the figure, u m , u n are the squares of the voltage amplitudes of nodes m and n, respectively; x mn is the reactance of branch m-n; S nk , P mn and Q mn are the apparent power, active power, and reactive power flowing from node m through branch m-n, respectively; P n and Q n are the net active load and reactive load injected into node n, respectively; S nk , P nk and Q nk are the apparent power, active power, and reactive power flowing from node n to the next node k, respectively. The current and active and reactive power constraints related to this branch can be expressed as:

[0052]

[0053]

[0054] The distribution network operation safety constraint is:

[0055]

[0056] In the formula: lmax mn is the maximum value of the square of the current amplitude of branch m-n; umin n and umax n are the minimum and maximum values of the square of the voltage amplitude of node n, respectively.

[0057] The PV inverter control constraint is:

[0058]

[0059] Where: Q PV，n is the reactive power output of the PV inverter at node n; S PV，n is the rated capacity of the PV inverter at node n; k f = cosθ, which is the minimum power factor of the PV inverter and is a given constant.

[0060] Obviously, the constraints in the above formula fully consider the limitations of the rated capacity and minimum power factor of the PV inverter. When the PV inverter operates on the boundary 0-1, it means that only the active power output of the PV is adjustable and the reactive power output is 0; when it operates on the boundary 1-2, it means that only the reactive power output of the PV is adjustable and the active power is output at the maximum power; when it operates on the boundary 2-3, it means that both the active and reactive power outputs of the PV are adjustable, but are limited by the rated capacity of the inverter; when it operates on the boundary 3-0, it means that both the active and reactive power outputs of the PV are adjustable, but are limited by the minimum power factor. The entire region Ⅰ satisfies the operating constraint conditions of the PV inverter, and the reactive power output by the inverter is inductive. Therefore, optimizing the power output of the PV inverter in this region is beneficial to the improvement of low voltage. The operating boundary of region Ⅱ has the same constraints as that of region Ⅰ, but the PV inverter in region Ⅱ outputs capacitive reactive power, that is, consumes inductive reactive power. Therefore, optimizing the power output of the PV in this region is beneficial to the recovery of overvoltage.

[0061] 202: Model convexification processing. First, linearize the non-linear objective, introduce the equality relationship between the node voltage amplitude and its square, and the equality relationship is:

[0062]

[0063] Use the SOC relaxation technology to convexify the distribution network power flow constraint as:

[0064]

[0065] The standard SOC form is:

[0066] ||[2P mn 2Q mn l mn -u m T ||≤l mn +u m (11)

[0067] ||[2U n u n -1] T ||≤u n +1 (12)

[0068] ​Using the SOC relaxation technique, the non-convex feasible region C_original of the original problem can be relaxed into a convex cone feasible region C_SOC, and then the original problem is transformed into a convex problem. Due to the introduction of SOC relaxation, the optimal solution S obtained by solving in C_SOC is the lower bound solution of the original problem. If the optimal solution S is a point in the original feasible region C_original, then the SOC relaxation is considered exact, that is, the optimal solution S is also the optimal solution of the original problem. The sufficient conditions for the exactness of the relaxation are strictly derived. Under certain conditions, that is, the objective function is an increasing function of the branch current, the network topology is a radial connected graph, etc., it is proved that when the original problem obtains the optimal solution, it can ensure that the equality after relaxation is accurate enough to satisfy all the constraints of the original problem. And the active distribution network voltage optimization problem proposed in this paper can meet the above conditions. Since the relaxed model is a convex model, the traditional branch and bound method and cutting plane method can also ensure the optimality and computational efficiency of the solution. After convexification, the model shown can be reformulated as:

[0069]

[0070] 203: Distribution network partitioning. The structure diagram of a 6-node radial distribution network is referred to Figure 3 , and according to the principle of decomposition and coordination, the distribution network is partitioned. Figure 3 In, the boundary between sub-region 1 and sub-region 2 is formed by node 2, node 4 and branch 2-4, and the boundary variables are the voltages of node 2 and node 4 and the transmission power flowing through branch 2-4 from node 2. When partitioning the distribution network, according to the principle of decomposition and coordination, the boundary between sub-regions needs to be copied into adjacent sub-regions to achieve decoupling of adjacent sub-regions. The specific distribution network partitioning process diagram is referred to Figure 4 , the dotted line represents the boundary branch of the adjacent sub-region copied; the hollow circle represents the boundary node of the adjacent sub-region copied; the superscript "+" represents that the sub-region retains the boundary variable; the superscript "-" represents the boundary variable of the adjacent sub-region copied. It can be seen that sub-region 1 copies node 4 and branch 2-4 and retains node 2; sub-region 2 copies node 2 and retains node 4 and branch 2-4. Define the boundary variables of sub-region 1 as Define the boundary variables of sub-region 2 as To decouple the problems of adjacent sub-regions after partitioning the distribution network and ensure the equivalence of the problems before and after partitioning, define the global variable to ensure that the boundary variables of sub-region 1 and sub-region 2 are correspondingly equal, that is, let

[0071] 204: The specific steps to construct a distributed optimization control model based on ADMM are as follows. The original variables are decomposed into different variables x and y, and the objective function is also decomposed into two parts to ensure the decomposability of the optimization process. The standard form of the algorithm is as follows:

[0072]

[0073] Where: both f(x) and g(y) are convex functions; A, B, and c are coefficient matrices. The constraint condition Ax + By = c for variables x and y constitutes the feasible region of the variables in the ADMM objective function.

[0074] Using the general consistency optimization method of ADMM, the present invention uses f(x) in Equation (12) to represent the independent optimization objective of each sub-region, and constructs a voltage distribution optimization control model based on the general consistency optimization method of ADMM as follows:

[0075]

[0076] Where: N is the number of distribution network sub-regions; fj(xj) is a convex function, representing the voltage optimization control objective function corresponding to sub-region j in Equation (13); the sub-region variable xj consists of internal variables and boundary variables of the sub-region; Xj is the feasible region of the sub-region variable xj, that is, the constraint condition corresponding to sub-region j in Equation (11); is the boundary variable and its global variable constitute a consistency constraint to ensure that the boundary node voltages and boundary branch transmission powers are equal when adjacent sub-regions are solved independently. The augmented Lagrangian function form of (15) is:

[0077]

[0078] Where: ρ j > 0 is the penalty parameter of sub-region j; λ j is the dual variable of sub-region j. To facilitate the update of the global variable, the dual variable λ j is scaled to μ j =(1 / ρ j )λ j , then Equation (17) is equivalent to Equation (19).

[0079]

[0080] The iterative calculation process of ADMM is as follows:

[0081]

[0082] Where: k is the number of iterations; k g is the number of boundary variables connected to ; G(j, i)=g is the mapping relationship between the i-th element in the boundary variable and the g-th element in the global variable .

[0083] For the independent optimization of distribution network sub-regions and the interaction process of boundary variables between sub-regions, refer toFigure 5 Taking sub-region 1 as an example, the sub-region variable x1 in sub-region 1 = [u1, P 12 , Q 12 , …, u + 2, u - 4, P - 24, Q - 24], where are boundary variables, and the rest are internal variables; the global variables Sub-region 1 and sub-region 2 are independently and parallelly optimized using Equation (18) to obtain the sub-region variables x1 and x2; sub-region 1 and sub-region 2 exchange their boundary variables and and use Equation (19) to update the global variables as shown in Figure 5 ; update the dual variables μ1 and μ2 using Equation (20); loop Figure 3 - 3 the optimization calculation within the 5 sub-regions and the boundary variable interaction process between sub-intervals until the convergence condition Equation (22) is satisfied, and stop the iteration.

[0084]

[0085] In the formula: the initial residual rk+1j and the dual residual sk+1j are the distances from the current iteration solution to the optimal solution. When the consistency constraint condition is exactly satisfied, rk+1j tends to 0. When the objective function value tends to the minimum value, sk+1j tends to 0; εk+1pri,j and εk+1dua,j are the convergence thresholds of the initial residual and the dual residual in this iteration respectively; is the number of elements of the boundary variable of the distribution network sub-region; ∈ abs and ∈ rel are the absolute tolerance and relative tolerance reference values respectively. In this article, ∈ abs = 10 -6 , ∈ rel = 5 × 10 -5

[0086] In summary, the example of the present invention realizes the power quality optimization control of the rural power distribution substation based on the distributed optical storage system through the above steps 201 - 204. On the one hand, it can ensure the optimality of the solution and the calculation efficiency. On the other hand, it realizes the distributed power quality optimization control, which has wide engineering application value.

[0087] Next, combined with a specific example, for a power quality optimization control method of a rural power distribution substation based on a distributed optical storage system proposed in the embodiment of the present invention, this example takes the IEEE - 33 bus system as an example for case analysis. The topology diagram of the IEEE - 33 bus system is as Figure 6 shown, and the details are described below:

[0088] The distributed PVs in the embodiments are numbered PV1 - PV9 and are installed at nodes 5, 8, 11, 15, 18, 21, 25, 29, and 33 in sequence, with the minimum power factor k f = 0.95; Based on the proposed method of guiding system partitioning by the number of global variables formed by system partitioning, considering the distribution network topology, geographical area, PV distribution, etc. comprehensively, sub-regions A1 and A2 are decomposed with nodes 5, 6, and branch 4 - 6 as the boundaries, and sub-regions A2 and A3 are decomposed with nodes 8, 9, and branch 8 - 9 as the boundaries; The distributed PV power sources installed in sub-regions A1 - A3 are (PV1, PV6, PV7), (PV2, PV8, PV9), and (PV3, PV4, PV5) respectively. In the numerical example of this paper, the voltage reference value is set to 12.66 kV, the reference load is 3715 + j2300 kV·A, the power reference value is 10 MW, and the safe operating range of the node voltage is [0.95, 1.05] p.u.

[0089] To study the influence of the proposed voltage optimization control strategy on the distribution network voltage, PV curtailment, and network loss, the following two scenarios are set respectively: Scenario 1 is the scenario where the PV output is large and the load level is low, resulting in the system node voltage exceeding the upper limit; Scenario 2 is the scenario where the PV output is small and the load level is high, resulting in the system node voltage falling below the lower limit.

[0090] In the above two scenarios, the values of each weight coefficient in formula (1) are ω1 = 0.4, ω2 = 0.3, ω3 = 0.3, and the calculation results of each scenario are described as follows.

[0091] Scenario 1: On a sunny day, at a certain moment between 10:00 - 14:00, the light is sufficient and the load level is low, resulting in an excess of PV output, which causes the voltage of some nodes to exceed the limit. At this time, the output power before the PV output power control is shown, and the load is 50% of the reference load. The proposed control strategy is used to perform active and reactive power regulation on the distributed PV, and the reference of the system node voltage distribution before and after the optimal control Figure 7 In the figure, the node voltage is in per-unit value, and the same applies hereinafter.

[0092] Figure 7Among them, before the optimization control, due to the excess PV output, the voltage of nodes 10 - 18 exceeded the upper limit, and the voltage of node 18 exceeded the limit to 1.08 p.u. To keep the node voltages of the distribution network within the safe operation range, a positive value indicates that the PV increases active / reactive power, and a negative value indicates reducing active power / generating capacitive reactive power. According to the algorithm proposed in the present invention, PV2 - PV5, PV8, and PV9 all generate capacitive reactive power to suppress the increase in node voltage. Since the voltage of node 18 exceeds the limit most severely, PV5 needs to reduce 64 kW of active power output to make up for the deficiency of reactive power regulation, so as to meet the voltage control requirements. By adjusting the active and reactive power outputs of the PVs, the voltages of all nodes in the system operate between 1 and 1.0479 p.u., meeting the voltage safety constraints. At the same time, since the reactive power regulation of PV1, PV6, and PV7 has little impact on the voltage - exceeding nodes, the inductive reactive power they generate is mainly used to optimize the system power flow distribution and compensate for reactive power loads to reduce network losses. However, to eliminate voltage over - limits, the capacitive reactive power increased by the PVs will cause the system's net reactive power load to increase by 641 kvar, and thus the network loss increases by 54.8 kW.

[0093] Scenario 2: Select a certain moment in the evening (17:00 - 19:00). At this time, the light intensity decreases and the load increases, resulting in some end - node voltages falling below the lower limit, and the load is 1.2 times the base load. Refer to the voltage distribution of the distribution network nodes before and after the distributed PV control using the method proposed in this paper. Figure 8 .

[0094] Figure 8 Among them, due to the decrease in PV output and the increase in load, before the voltage optimization control, the voltage level of the system nodes is low, and the voltages of end - nodes 29 - 33 are lower than the lower limit value. For example, the voltage of node 32 is as low as 0.942 p.u. After adopting the proposed optimization control strategy, the PV output power and its change amount are as shown in the output power after control in Table B2 of Appendix B. At this time, the change amounts of the active power outputs of PV1 - PV9 are all 0, that is, the active power output of the system PV does not change. The reactive power output of each PV is limited by the minimum power factor, and the inductive reactive power of 789 kvar is increased by making full use of the reactive power output capacity of the PV inverter to reduce the system's net reactive power load, thereby improving the voltage level of the system nodes. The system node voltages operate between 0.9521 and 1 p.u., and the network loss is reduced by 45.7 kW.

[0095] The analysis results of Scenarios 1 and 2 show that the proposed voltage distributed optimization control strategy for active distribution networks with distributed PV can effectively solve the problem of voltage over - limits in distribution networks. In Scenario 1, the PV can eliminate the phenomenon of voltage exceeding the upper limit by increasing capacitive reactive power and appropriately reducing active power. In Scenario 2, the PV can eliminate the phenomenon of voltage falling below the lower limit by increasing inductive reactive power to compensate for reactive power loads, and at the same time, the network loss can be reduced.

[0096] The above results show that by adjusting the active and reactive power outputs of distributed PV, it helps to improve the system voltage security and make the system operation safer and more economical.

[0097] Those skilled in the art can understand that the attached drawings are only schematic diagrams of a preferred embodiment, and the above serial numbers of the embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0098] The above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

[0099] References

[0100] [1] Ye Wanjia. Research on Key Technologies of Virtual Synchronous Generator Control for Energy Storage Subsystem in Wind-Energy Storage System [D]. Hefei University of Technology, 2023.

[0101] [2] Chai Xiuhui, Zhang Chunjiang, Zhao Xiaojun, etc. Research on Hybrid VSG Control Strategy for Grid-Connected Energy Storage Inverters [J]. Power Electronics, 2023, 57(11): 74 - 76 + 97.

[0102] [3] Guo Dongxu, Zhao Pengxiang, Yang Jialin, etc. Suppression of Active-Frequency Oscillation in Parallel System of Grid-Forming Energy Storage Inverters [J]. Power Electronics, 2024, 58(09): 70 - 73.

[0103] [4] Li Dandan. Research on Voltage Source Inverter Based on Fuzzy Control [D]. North China Electric Power University, 2013.

Claims

1. A method for optimizing and controlling power quality in a rural power distribution area based on a distributed solar energy storage system, characterized by: The process includes the following steps, which are performed in sequence: Step 1: According to the grid operation requirements, the node voltage deviation, PV reduction and network loss are minimized by adjusting the active and reactive output of distributed PV, and the voltage optimization control model objective function and constraint conditions are established; Step 2: Using SOC relaxation technology to convexify the voltage optimization control model described in step 1; Step 3: According to the decomposition and coordination principle, the distribution network is divided into zones and adjacent sub-zones are decoupled; Step 4: For the convex model, a voltage distributed optimization control model is constructed based on the ADMM general consistency optimization method.

2. According to claim 1, a method for optimizing and controlling power quality in a rural power distribution area based on a distributed photovoltaic energy storage system is characterized by: The objective function of the voltage optimization control model in step 1 is: Where: N bus is the node set in the distribution network; U n is the voltage amplitude of node n; let the voltage amplitude of node 1 be the voltage reference value, U1=1p.u.; N PV is the set of nodes connected to PV in the distribution network; Pmax PV, n and P PV，n are the maximum PV active output power and the PV active output power of node n respectively; k: n→k represents the set of branch end nodes with node n as the head node; r nk and l nk are the resistance and current amplitude square of branch nk respectively; ω1-ω3 are minimization weight coefficients, all greater than or equal to 0 and ω1+ω2+ω3=1; ξ1-ξ3 are correction coefficients greater than 0 to ensure that the values ​​of the three items in the formula are of the same order of magnitude; the first item in the formula is the node voltage deviation, the second item is the PV reduction, and the third item is the network loss. The node voltage deviation is nonlinear.

3. According to claim 1, a method for optimizing and controlling power quality in a rural power distribution area based on a distributed photovoltaic energy storage system is characterized by: The constraints in step 1 include distribution network power flow constraints, distribution network operation safety constraints, and PV inverter control constraints.

4. According to claim 1, a method for optimizing and controlling power quality in a rural power distribution area based on a distributed photovoltaic energy storage system is characterized by: The model after convexification processing in step 2 is: Where u n is the square of the voltage amplitude at node n.

5. According to claim 1, a method for optimizing and controlling power quality in a rural power distribution area based on a distributed photovoltaic energy storage system is characterized by: The voltage distributed optimization control model in step 3 is: Where: f(x) and g(y) are convex functions; A, B, c are coefficient matrices; the constraints Ax+By=c of variables x and y constitute the feasible domain of variables in the ADMM objective function.

Images