A rigid consensus method and system for multi-inverter coordination in power distribution networks

Abstract

The application discloses a rigid consensus method and system for multi-inverter coordination in a power distribution network, wherein for a Q(V) curve of traditional droop control, a correction coefficient of a slope is introduced for optimization; node voltage is estimated based on sensitivity analysis, constraints are established, an optimization model is defined and solved to obtain an optimized correction coefficient; a minimum rigid graph is determined, an expected distance between a current correction coefficient and a target correction coefficient is set, and then a master node is controlled; through iterative optimization, the master node and a slave node are converged to an optimization solution result, so that a new reactive power output vector is calculated, dynamic reactive power compensation is performed, and voltage regulation is completed. On the basis of rigid graph theory communication, the application regulates and controls voltage fluctuation of the power distribution network, maintains feeder voltage in a safe range, and ensures maximum total photovoltaic output power; while maintaining voltage stability of the power distribution network, the application also ensures output power of the photovoltaic power supply, and improves economic benefits of new energy power generation.

Description

Technical Field

[0001] This invention relates to the field of active distribution network voltage regulation technology, specifically to a rigid consensus method and system for coordinating multiple inverters in a distribution network, applicable to voltage stability control and improvement of renewable energy utilization efficiency in distribution networks containing high-penetration renewable energy. Background Technology

[0002] With the vigorous promotion of low-carbon energy development by countries around the world, distributed photovoltaic (PV) power generation has developed rapidly, with total installed capacity and scale continuously increasing. However, the increased PV penetration rate inevitably affects the reliability of the distribution network. Voltage violations and rapid fluctuations caused by PV grid connection have become major challenges restricting the safe and stable operation of the distribution network. The intermittent and random characteristics of distributed PV generators lead to voltage fluctuations at terminal nodes. When PV power generation exceeds load demand, reverse power flows from the terminal to the upstream node, causing a voltage rise. Conversely, when load demand exceeds PV output, the voltage at the terminal node drops due to line losses. In addition to the traditional requirements of stabilizing voltage fluctuations and maintaining operating voltage within a safe range, the need for rapid voltage control is gradually emerging.

[0003] Various regulating devices, such as on-load transformer tap changers, capacitor banks, static transformer compensators, and photovoltaic (PV) inverters, are coordinated for voltage control in distribution networks. Due to the flexible location and fast response of PV inverters, those with sensing, communication, and computing capabilities can be considered intelligent agents capable of coordinating voltage regulation. In traditional reactive power control, PV inverters follow local V(Q) curves to compensate for reactive power; however, the selection of the slope of the local V(Q) curve cannot be based on precise optimization calculations, making it difficult to alleviate voltage violations.

[0004] Some literature proposes a consensus-based leader-follower distributed control method, which guides photovoltaic inverters to achieve consensus while also addressing voltage violations. However, existing methods rely too heavily on a single agent and suffer from locality of reference, making them prone to single points of failure. Other literature considers an average consensus algorithm, but the partitioning involved hinders its adaptability to different scenarios and limits its flexibility. Further coordinated control is needed for cooperation among multiple inverters. Summary of the Invention

[0005] In view of this, in order to solve the above-mentioned problems in the prior art, this invention proposes a rigid consensus method and system for coordinating multiple inverters in a distribution network. Based on rigid graph theory communication, it coordinates the output relationship between distributed inverters on the feeder, regulates voltage fluctuations in the distribution network, and maximizes the utilization efficiency of renewable energy.

[0006] The present invention solves the above problems through the following technical means:

[0007] In a first aspect, the present invention provides a rigid consensus method for coordinating multiple inverters in a distribution network, comprising the following steps:

[0008] Step 1: Measure the voltage of the photovoltaic inverter node. The voltage value is fed back to the intelligent photovoltaic inverter of the control master node through the communication network. For the Q(V) curve of droop control, a slope correction coefficient ε is introduced for collaborative optimization.

[0009] Step 2: Estimate the node voltage based on sensitivity analysis, establish voltage constraints, and set constraints on current, inverter, transformer, and parameters. Define and solve the optimization model with the goal of maximizing the total power of the photovoltaic inverter.

[0010] Step 3: Based on the correction coefficient ε and the target correction coefficient ε * Determine the minimum rigidity graph of communication, set the desired distance between the current correction coefficient and the target correction coefficient, and then control the master node with edge computing capabilities;

[0011] Step 4: After several iterations, both master and slave nodes converge to the optimized solution. Return to Step 1, and each inverter obtains a new reactive power output vector based on the optimized correction coefficient ε. Dynamic reactive power compensation is performed to achieve voltage regulation.

[0012] Preferably, in step 1, a slope correction coefficient ε is introduced for the Q(V) curve of droop control:

[0013] ε=[ε1,ε2,…,ε n ]

[0014] Where ε1, ε2, ..., ε n These represent the correction coefficients for the 1st, 2nd, ..., nth photovoltaic inverters, respectively. The Q(V) curve for the i-th inverter can be rewritten as:

[0015]

[0016] In the formula, This represents the reactive power output of the i-th inverter after correction, and the grid-connected node voltage of the i-th photovoltaic inverter is represented by V. i express, This is the maximum reactive power that the inverter can output. and ε represents the upper and lower limits of the dead zone. i It is the slope correction factor, k i This indicates the default slope of the Q(V) curve;

[0017] To maximize the utilization efficiency of grid-connected photovoltaic power, and setting the maximum total power of the photovoltaic inverter as the objective function, we have:

[0018]

[0019] Among them, P total Let S be the total active power output of the photovoltaic inverter, and let |S| be the current apparent power output of the i-th inverter. i | is determined by the result of MPPT calculation. Then it is determined by step 1.

[0020] Preferably, in step 2, in order to estimate the voltage value after the power change, it is necessary to use sensitivity analysis to determine the relationship between the voltage and power at each node. The calculation method is as follows:

[0021]

[0022]

[0023]

[0024] Among them, G ij B is the real part of the nodal admittance matrix. ij P represents the imaginary part of the nodal admittance matrix, indicating the mutual influence between nodes i and j. i Q i Represents the active and reactive power of a node, θ ij V represents the phase angle difference between node i and node j. i and V j Let ΔP, ΔQ, ΔV, and Δθ represent the voltages at nodes i and j, respectively. Let ΔP, ΔQ, ΔV, and Δθ be the system's active power increment matrix, reactive power increment matrix, voltage increment matrix, and phase angle increment matrix, respectively. J is the Jacobian matrix, where J... Pθ J PV J Qθ J QV These represent the relationships between active power and phase angle, active power and voltage, reactive power and phase angle, and reactive power and voltage in the Jacobian matrix, respectively, where n is the total number of system nodes;

[0025] The relationship between the system reactive power increment ΔQ and the node voltage increment ΔV can be derived from the above equation:

[0026] ΔV=S PV ΔP+S QV ΔQ

[0027] S PV and S QV This represents the voltage sensitivity matrix to active and reactive power; therefore, the rate of change ΔV of node j with respect to the voltage at node i is...i for:

[0028] ΔV i =S PV ΔP j +S QV ΔQ j

[0029] The estimated value of the voltage at node i at time t1 is:

[0030]

[0031] Where V i (t0) and V i (t1) represents the application of the correction coefficient ε. i The voltages before and after node i, ΔV i (t0) represents the voltage change at node i at time t0, S PV and S QV This represents the voltage sensitivity matrix to active and reactive power, ΔP. j and ΔQ j This represents the difference in power between the current moment and the previous moment; therefore, the boundary condition for voltage is:

[0032]

[0033] Among them, V low and V high This represents the lower and upper limits of the voltage allowed by the node;

[0034] Other constraints that need attention are:

[0035] Current constraints: In the process of power flow calculation in a power system, the following equation must be satisfied:

[0036] I≤I max

[0037] In the formula I max Let I be the maximum current carrying capacity of the conductor, and let I be the conductor current.

[0038] Inverter constraints: The power factor of a photovoltaic inverter is adjustable, therefore it must satisfy the following equation:

[0039]

[0040] In the formula P i Q i and S i Let be the active, reactive, and apparent power of the i-th inverter;

[0041] Parameter constraints: Since the correction factor changes the proportion of power output, it should satisfy the following formula:

[0042] ε i ≥0.

[0043] Preferably, the computational optimization model in step 2 is:

[0044]

[0045] V low ≤V i (t0)

[0046]

[0047] ε i ≥0

[0048] Among them, V high and V low These are the maximum and minimum values ​​of the voltage; I max P is the maximum current-carrying capacity of the conductor, where I is the conductor current; i Q i and S i These represent the active, reactive, and apparent power of the i-th inverter, respectively, ε i This is the correction coefficient for the i-th inverter.

[0049] Preferably, in step 3, the communication network consisting of n photovoltaic grid-connected nodes is represented by an undirected graph G = (V, E), where V = {1, 2, ..., n} is the set of vertices. It is a set of undirected edges, and the number of edges is . like Let be the coordinates of vertices i and j. Then the frame F is (G, p), where... The framework is implemented using points on a plane; considering the order of the edges in E, the edge function φ(p) has the following definition:

[0050]

[0051] Where ||·||² is the Euclidean norm; the k-th component of φ(p) represents the k-th edge in E connecting vertices i and j; to reduce communication overhead, the graph rigid matrix method is introduced into undirected graph communication networks, and the rigid matrix R(p) of the frame F=(G,P) is defined as:

[0052]

[0053] If rank[R(p)] = 2n-3, i.e. l ​​= 2n-3, then the frame (G,p) has infinitesimal rigidity;

[0054] The optimized correction coefficient ε is represented by F using a minimum communication rigidity framework. * =(G* ,ε * ), where G * =(V * E * () represents a formation graphic. * indicates the target value of this variable when consensus is reached, and the expected distance d between the correction coefficient of inverter i and the correction coefficient of inverter j. ij for:

[0055]

[0056] The relative correction coefficients of the two inverters Defined as:

[0057]

[0058] and make It has the same order as the defined side function φ(p); the expected distance error e of the correction coefficient ij Defined as:

[0059]

[0060] Preferably, in step 3, the system with n adjustable correction coefficients is modeled by a single integrator.

[0061]

[0062] in It is the actual correction factor for the i-th inverter. It is the control input of the i-th inverter;

[0063] Expected distance between the correction factor and the target correction factor The adjustment direction of the correction coefficient is then adjusted in real time by estimating the correction as a function of time t:

[0064]

[0065] Where k1,k2>0 are user-defined control coefficients, τ is an intermediate quantity in the integration process, sgn(·) is the standard sign function, and e T =ε T -ε n This represents the interception error between the corrected value and the target value, where εT represents the target value of the correction coefficient, and ε n This indicates the correction value of the correction factor;

[0066] The control law u is defined as:

[0067] u = u a +h

[0068] in

[0069] u a =-R T (ε)z=(u a1 ,…,u an )

[0070]

[0071] Where k > 0 is the user-defined control gain, R T It is the transpose of the rigid matrix R. After calculating the new matrix, it can be written as (u a1 ,…,u an In the form of ) i,j)∈E * .

[0072] Preferably, in step 4, each inverter obtains a new correction coefficient ε. i Substituting the values ​​into the Q(V) equation from step 1, we obtain the new reactive power output vector Q. new ;

[0073]

[0074] in This represents the reactive power required to be output by the 1st, ..., nth photovoltaic inverter; the calculated... The voltage is fed into the corresponding inverter, which performs reactive power compensation based on the calculation results to complete voltage regulation.

[0075] Secondly, the present invention provides a rigid consensus system for coordinating multiple inverters in a distribution network, comprising:

[0076] The correction coefficient introduction module is used to measure the voltage of the photovoltaic inverter node. The voltage value is fed back to the intelligent photovoltaic inverter controlling the master node via the communication network. For the Q(V) curve of droop control, the slope correction coefficient ε is introduced for collaborative optimization.

[0077] The optimization model definition module is used to estimate node voltage based on sensitivity analysis, establish voltage constraints, and set constraints on current, inverter, transformer, and parameters. With the goal of maximizing the total power of the photovoltaic inverter, the optimization model is defined and solved.

[0078] The master node control module is used to adjust the correction coefficient ε and the target correction coefficient ε. * Determine the minimum rigidity graph of communication, set the desired distance between the current correction coefficient and the target correction coefficient, and then control the master node with edge computing capabilities;

[0079] The dynamic reactive power compensation module is used to ensure that after several iterations, both master and slave nodes converge to the optimized solution. Each inverter then obtains a new reactive power output vector based on the optimized correction coefficient ε. Dynamic reactive power compensation is performed to achieve voltage regulation.

[0080] Thirdly, the present invention provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the rigid consensus method for coordinating multiple inverters in a distribution network as described in the first aspect of the present invention.

[0081] Fourthly, the present invention provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the rigid consensus method for coordinating multiple inverters in a distribution network as described in the first aspect of the present invention.

[0082] Compared with the prior art, the beneficial effects of the present invention include at least the following:

[0083] This invention addresses the voltage fluctuation problem in distribution networks for high-penetration renewable energy sources by employing a distributed control algorithm based on rigid consensus to coordinate and optimize the power output of multiple inverters. Compared to traditional local Q(V) control curves, this method applies a correction coefficient to adjust the curve's slope and estimates the impact of power changes on voltage using sensitivity matrix analysis. This maximizes the total active power output of the inverters while ensuring voltage stability. Extending the graphical rigidity-based method to the coordinated control of distributed photovoltaic inverters reduces communication requirements compared to traditional distributed communication, enhances the robustness of distributed collaboration, and effectively eliminates voltage violations. This method does not require each inverter to have computing power; only intelligent photovoltaic inverters with edge computing capabilities are needed for optimization calculations, followed by coordinated control via a rigid communication graph. This invention improves the utilization efficiency of renewable energy and offers significant economic benefits. Attached Figure Description

[0084] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0085] Figure 1 This is a schematic diagram of the system structure of the present invention;

[0086] Figure 2 This is a flowchart of the operation steps of the rigid consensus method for coordinating multiple inverters in a power distribution network according to the present invention.

[0087] Figure 3 This is a schematic diagram of the control curve of the dynamic correction coefficient Q(V) of the present invention;

[0088] Figure 4 This is a schematic diagram of the minimum rigidity communication of the present invention;

[0089] Figure 5 This is a schematic diagram of the optimization iteration of the present invention;

[0090] Figure 6 This is a schematic diagram of the rigid consensus system for coordinating multiple inverters in a power distribution network according to the present invention;

[0091] Figure 7 This is a block diagram of the electronic device structure of the present invention. Detailed Implementation

[0092] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, the technical solutions of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that the described embodiments are merely some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0093] Example 1

[0094] This invention proposes a rigid consensus method for coordinating multiple inverters in a distribution network. This method is applicable to active distribution network systems containing distributed photovoltaic (PV) power sources. The active distribution network system includes a PV array connected to a smart PV grid-connected inverter. The overall system structure of this method is illustrated below. Figure 1 As shown, the operation steps are combined Figure 2 The main points are as follows:

[0095] Step 1: Measure the voltage of the photovoltaic inverter node. The voltage value is fed back to the intelligent photovoltaic inverter controlling the master node via the communication network. For the Q(V) curve of traditional droop control, a slope correction coefficient ε is introduced for collaborative optimization.

[0096] Step 2: Estimate the node voltage based on sensitivity analysis, establish voltage constraints, and set other constraints such as current, inverter, transformer, and parameters. Define and solve the optimization model with the goal of maximizing the total power of the photovoltaic inverter.

[0097] Step 3: Based on the correction coefficient ε and the target correction coefficient ε * Determine the minimum rigidity graph of communication, set the desired distance between the current correction coefficient and the target correction coefficient, and then control the master node with edge computing capabilities;

[0098] Step 4: After several iterations (100-200 times), both master and slave nodes converge to the optimized solution. Return to Step 1, and each inverter obtains a new reactive power output vector based on the optimized correction coefficient ε. Dynamic reactive power compensation is performed to achieve voltage regulation.

[0099] Step 1 specifically includes the following steps:

[0100] In general, smart photovoltaic inverters with edge computing capabilities are defined as control master nodes, while other photovoltaic inverters on the feeder without edge computing capabilities are defined as follower nodes. The voltage V_i (i = 1, 2, ..., n) of each node is measured. The Q(V) curve for traditional droop control is shown below. Figure 3 As shown, a correction factor ε for the slope is introduced:

[0101] ε=[ε1,ε2,…,ε n ]

[0102] Where ε1, ε2, ..., ε n These represent the correction coefficients for the 1st, 2nd, ..., nth photovoltaic inverters, respectively. The Q(V) curve for the i-th inverter can be rewritten as:

[0103]

[0104]

[0105] In the formula, This represents the reactive power output of the i-th inverter after correction, and the grid-connected node voltage of the i-th photovoltaic inverter is represented by V. i express, This is the maximum reactive power that the inverter can output. and ε represents the upper and lower limits of the dead zone. i It is the slope correction factor, k i This indicates the slope of the default Q(V) curve.

[0106] To maximize the utilization efficiency of grid-connected photovoltaic power, and setting the maximum total power of the photovoltaic inverter as the objective function, we have:

[0107]

[0108] Among them, P total Let S be the total active power output of the photovoltaic inverter, and let |S| be the current apparent power output of the i-th inverter. i | is determined by the result of MPPT calculation. Then it is determined by step 1.

[0109] Step 2 specifically includes the following steps:

[0110] To ensure maximum power output while maintaining the feeder voltage within the normal range, the corrected voltage is estimated using a voltage and power sensitivity matrix. To estimate the voltage value after power changes, sensitivity analysis is needed to determine the relationship between voltage and power at each node. The calculation method is as follows:

[0111]

[0112]

[0113]

[0114] Among them, G ij B is the real part of the nodal admittance matrix. ij P represents the imaginary part of the nodal admittance matrix, indicating the mutual influence between nodes i and j. i Q i Represents the active and reactive power of a node, θ ij V represents the phase angle difference between node i and node j. i and V j Let ΔP, ΔQ, ΔV, and Δθ represent the voltages at nodes i and j, respectively. Let ΔP, ΔQ, ΔV, and Δθ be the system's active power increment matrix, reactive power increment matrix, voltage increment matrix, and phase angle increment matrix, respectively. J is the Jacobian matrix, where J... Pθ J PV J Qθ J QV These represent the relationships between active power and phase angle, active power and voltage, reactive power and phase angle, and reactive power and voltage in the Jacobian matrix, respectively, where n is the total number of system nodes.

[0115] The relationship between the system reactive power increment ΔQ and the node voltage increment ΔV can be derived from the above formula.

[0116] ΔV=S PV ΔP+S QV ΔQ

[0117] S PV and S QV This represents the voltage sensitivity matrix to active and reactive power. Therefore, the rate of change ΔV of node j with respect to the voltage at node i is... i :

[0118] ΔV i =S PV ΔP j +S QV ΔQ j

[0119] The estimated value of the voltage at node i at time t1 can be obtained as follows:

[0120]

[0121] Where V i (t0) and V i (t1) represents the correction coefficient ε for the applied slope. i The voltages before and after node i, ΔV i (t0) represents the voltage change at node i at time t0, S PV and S QV This represents the voltage sensitivity matrix to active and reactive power, ΔP. j and ΔQ j This represents the difference in power between the current moment and the previous moment. Therefore, the boundary condition for voltage is...

[0122]

[0123] Among them, V low and V high This represents the lower and upper limits of the voltage allowed by the node.

[0124] Other constraints that need to be satisfied are:

[0125] Current constraints: In the process of power flow calculation in a power system, the following equation must be satisfied:

[0126] I≤I max

[0127] In the formula I max I is the maximum current carrying capacity of the conductor, and I is the conductor current.

[0128] Inverter constraints: The power factor of a photovoltaic inverter is adjustable, therefore it must satisfy the following equation:

[0129]

[0130] In the formula P i Q i and S i Let be the active, reactive, and apparent power of the i-th inverter.

[0131] Parameter constraints: Since the correction factor changes the proportion of power output, it should satisfy the following formula:

[0132] ε i ≥0

[0133] Furthermore, we define the computational optimization model:

[0134]

[0135]

[0136] Among them, V high and V low These are the maximum and minimum values ​​of the voltage; I max P is the maximum current-carrying capacity of the conductor, where I is the conductor current; i Q i and S i These represent the active, reactive, and apparent power of the i-th inverter, respectively, ε i This is the correction coefficient for the i-th inverter.

[0137] Step 3 specifically includes the following steps:

[0138] like Figure 4 As shown, a communication network consisting of n photovoltaic grid-connected nodes can be represented by an undirected graph G = (V, E), where V = {1, 2, ..., n} is the set of vertices. It is a set of undirected edges, and the number of edges is . like Let be the coordinates of vertices i and j. Then the frame F is (G, p), where... The framework is implemented using points on a plane. Considering the order of the edges in E, the edge function φ(p) is defined as follows:

[0139]

[0140] Where ||·||² is the Euclidean norm. The k-th component of φ(p) represents the k-th edge in E connecting vertices i and j. To reduce communication overhead, the graph rigid matrix method is introduced into traditional undirected graph communication networks. The rigid matrix R(p) of the frame F=(G,P) is defined as:

[0141]

[0142] If rank[R(p)] = 2n-3, i.e. l ​​= 2n-3, then the frame (G,p) is infinitely and minimally rigid.

[0143] The optimized correction coefficient ε is represented by F using a minimum communication rigidity framework. * =(G * ,ε * ), where G * =(V * E * () represents a formation graphic. * indicates the target value of this variable when consensus is reached, and the expected distance d between the correction coefficient of inverter i and the correction coefficient of inverter j. ij for:

[0144]

[0145] The relative correction coefficients of the two inverters Defined as:

[0146]

[0147] and make It has the same order as the side function φ(p). The expected distance error e of the correction coefficient. ij Defined as:

[0148]

[0149] A system with n adjustable correction coefficients is modeled by a single integrator:

[0150]

[0151] in It is the actual correction factor for the i-th inverter. It is the control input of the i-th inverter.

[0152] Expected distance between the correction factor and the target correction factor The adjustment direction of the correction coefficient is then adjusted in real time by estimating the correction as a function of time t:

[0153]

[0154] Where k1,k2>0 are user-defined control coefficients, τ is an intermediate quantity in the integration process, sgn(·) is the standard sign function, and e T =ε T -ε n ε represents the interception error between the corrected value and the target value. T ε represents the target value of the correction factor. n This indicates the correction value of the correction factor;

[0155] The control law u is defined as:

[0156] u = u a +h

[0157] in

[0158] u a =-kR T (ε)z=(u a1 ,…,u an )

[0159]

[0160] Where k > 0 is the user-defined control gain, RT It is the transpose of the rigid matrix R. After calculating the new matrix, it can be written as (u a1 ,…,u an In the form of ) (i,j)∈E * .

[0161] Step 4 specifically includes the following steps:

[0162] like Figure 5 As shown, through continuous optimization and iteration, each inverter obtains a new correction coefficient ε. i Substituting the values ​​into the Q(V) equation from step 1, we obtain the new reactive power output vector Q. new .

[0163]

[0164] in This represents the reactive power output required by the 1st, ..., nth photovoltaic inverter. The calculated... The voltage is fed into the corresponding inverter, which performs reactive power compensation based on the calculation results to complete voltage regulation.

[0165] This invention improves the utilization efficiency of renewable energy and has high economic benefits by using intelligent photovoltaic inverters with edge computing capabilities for optimization calculations, followed by coordinated control via a rigid communication graph.

[0166] Example 2

[0167] like Figure 6 As shown, the present invention provides a rigid consensus system for coordinating multiple inverters in a distribution network, including a correction coefficient introduction module, an optimization model definition module, a master node control module, and a dynamic reactive power compensation module.

[0168] The correction coefficient module shown is used to measure the voltage of the photovoltaic inverter node. The voltage value is fed back to the intelligent photovoltaic inverter controlling the master node via the communication network. For the Q(V) curve of traditional droop control, the slope correction coefficient ε is introduced for collaborative optimization.

[0169] The optimization model definition module shown is used to estimate node voltage based on sensitivity analysis, establish voltage constraints, and set constraints on current, inverter, transformer, and parameters. With the goal of maximizing the total power of the photovoltaic inverter, the optimization model is defined and solved.

[0170] The master node control module shown is used to adjust the correction coefficient ε and the target correction coefficient ε. * Determine the minimum rigidity graph of communication, set the desired distance between the current correction coefficient and the target correction coefficient, and then control the master node with edge computing capabilities;

[0171] The dynamic reactive power compensation module shown is used to achieve convergence of the master and slave nodes to the optimized solution after several iterations. Each inverter then obtains a new reactive power output vector based on the optimized correction coefficient ε. Dynamic reactive power compensation is performed to achieve voltage regulation.

[0172] Based on rigid graph theory communication, this invention coordinates the output relationship between distributed inverters on the feeder line, regulates voltage fluctuations in the distribution network, and maintains the feeder voltage within a safe range while maximizing the total photovoltaic output power. This invention maintains stable distribution network voltage while ensuring the output power of the photovoltaic power source, thus improving the economic efficiency of new energy power generation.

[0173] Other features in this embodiment are the same as those in Embodiment 1, and therefore will not be repeated here.

[0174] Example 3

[0175] Based on the same concept, the present invention also provides a schematic diagram of a physical structure, such as... Figure 7 As shown, the server may include a processor 810, a communications interface 820, a memory 830, and a communication bus 840, wherein the processor 810, communications interface 820, and memory 830 communicate with each other via the communication bus 840. The processor 810 can call logical instructions in the memory 830 to execute the steps of the rigid consensus method for multi-inverter coordination in a power distribution network.

[0176] Furthermore, the logical instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0177] Example 4

[0178] Based on the same concept, the present invention also provides a non-transitory computer-readable storage medium storing a computer program containing at least one piece of code that can be executed by a master control device to control the master control device to implement the steps of the rigid consensus method for coordinating multiple inverters in a distribution network.

[0179] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive).

[0180] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

[0181] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A rigid consensus method for multi-inverter coordination in a power distribution network, characterized in that, Comprising the following steps: Step 1: Measure the voltage at the photovoltaic inverter node. The voltage value is fed back to the intelligent photovoltaic inverter controlling the master node via the communication network. (For droop control...) Curve, introducing a slope correction factor Perform collaborative optimization; Step 2: Estimate node voltage based on sensitivity analysis, establish voltage constraints, and set constraints on current, inverter, transformer, and parameters. Define and solve the optimization model with the goal of maximizing the total power of the photovoltaic inverter; Step 3: Based on the correction coefficient... With target correction coefficient Determine the minimum rigidity graph for communication, set the desired distance between the current correction coefficient and the target correction coefficient, and then control the master node with edge computing capabilities; Step 4: After several iterations, both master and slave nodes converge to the optimized solution, return to Step 1, and each inverter adjusts its operation according to the optimized correction coefficient. Obtain the new reactive power output vector Dynamic reactive power compensation is performed to complete voltage regulation; In step 3, an undirected graph is used to represent the communication network consisting of n photovoltaic grid-connected nodes. To indicate, among which It is a set of vertices. It is a set of undirected edges, and the number of edges is . ;like It is the vertex and vertex If the coordinates are given, then the frame F is... ,in This indicates that the framework is implemented using points on a plane; considering the order of the edges in E, the edge function... The following definitions apply: in It is a Euclidean norm; The Middle Each component represents Connecting vertices and The Edges; to reduce communication overhead, the graph rigid matrix method is introduced into undirected graph communication networks, framework. rigid matrix Defined as: If there is ,Right now So, the framework It possesses infinitesimal rigidity; the optimized correction coefficient Represented by a minimum communication rigidity framework ,in For formation graphics, * indicates the target value of this variable when consensus is reached, inverter Correction factor and inverter Expected distance between correction factors for: The relative correction coefficients of the two inverters Defined as: and make With the defined edge function Expected distance error with the same sorting and correction coefficient Defined as: ; In step 3, the system of n adjustable correction factors is modeled by a single integrator where is the actual correction factor of the n th inverter, is the control input of the n th inverter; Expected distance between correction coefficient and target correction coefficient The adjustment direction of the correction coefficient is adjusted in real time by estimating the following function of time t: wherein, is a user-defined control coefficient, and τ is an intermediate quantity in the integral process, is a standard symbol function, represents the intercept error between the correction value and the target value, wherein represents the target value of the correction coefficient, represents the correction value of the correction coefficient; control law is defined as: wherein wherein, is a user-defined control gain, is the transpose matrix of the rigid matrix After a new matrix is calculated, it can be written in the form of , , .

2. The rigid consensus method for multi-inverter coordination in a power distribution network of claim 1, wherein, In step 1, the drooping control Curve, introducing a slope correction factor : in Representing the first The correction factor for the first photovoltaic inverter, for the first... one inverter The curve is rewritten as: In the formula, represents the corrected reactive power of the nth inverter output, the grid-connected node voltage of the nth photovoltaic inverter is represented by , represents the corrected reactive power of the nth inverter output, the grid-connected node voltage of the nth photovoltaic inverter is represented by , is the maximum reactive power that can be output by the inverter, and are the upper and lower limits of the dead zone, is the correction coefficient of the slope, represents the default curve slope; in order to maximize the utilization efficiency of photovoltaic grid connection, the total power of the photovoltaic inverter is set as the objective function, that is, wherein, is the total active power output by the photovoltaic inverter, the current apparent power output of the nth inverter is , which is determined by the operation result of MPPT, which is determined by step 1.

3. The rigid consensus method for multi-inverter coordination in a power distribution network of claim 1, wherein, In step 2, in order to estimate the voltage value after the power change, the change relationship between the voltage of each node and the power is analyzed by using the sensitivity, and the operation method is as follows: Wherein, is the real part of the node admittance matrix, is the imaginary part of the node admittance matrix, representing the mutual influence of the nodes and the node , , represent the active power and the reactive power of the node, represent the phase angle difference between the nodes and the node , and represent the voltage of the nodes and the node , , , , are the active power increment matrix, the reactive power increment matrix, the voltage increment matrix, and the phase angle increment matrix of the system respectively, is the Jacobian matrix, wherein , , , are the relationships between the active power and the phase angle, the active power and the voltage, the reactive power and the phase angle, and the reactive power and the voltage in the Jacobian matrix respectively, is the total number of system nodes; the relationship between the reactive power increment of the system and the voltage increment of the node is obtained by the above formula: and represent the sensitivity matrix of the voltage to the active power and the reactive power; therefore, the change rate of the node to the voltage of the node is: The estimated value of the voltage of the node at the time is: Wherein and represent the voltage of the node before and after applying the correction coefficient , represents the voltage change amount of the node at the time , and represent the sensitivity matrix of the voltage to the active power and the reactive power, and represents the difference between the current and the previous time step; the boundary conditions for the voltage are thus: wherein, and represent the lower and upper voltage limits allowed at the nodes. Other constraints to be noted are: Current constraint: In the process of power system flow calculation, the following formula must be satisfied: wherein is the maximum current carrying capacity of the conductor, is the current of the conductor; Inverter constraints: The photovoltaic inverter power factor is adjustable, so it must satisfy the following equation: where , and are the active, reactive and apparent power of the th inverter. Parameter constraint: Since the correction coefficient is a proportion to change the power output, the following equation should be satisfied: .

4. The rigid consensus method for multi-inverter coordination in a power distribution network according to claim 3, wherein, The calculation optimization model in step 2 is: wherein, and are the maximum and minimum values of the voltage; is the maximum current-carrying capacity of the conductor, is the current of the conductor; , and are the active, reactive and apparent power of the th inverter, respectively, is the correction coefficient of the th inverter.

5. The rigid consensus method for multi-inverter coordination in a power distribution network of claim 1, wherein, In step 4, each inverter obtains a new correction coefficient. Substitute into step 1 The formula is used to calculate and obtain the new reactive power output vector. ; in Indicates the first The reactive power output required by each photovoltaic inverter; the calculated The voltage is fed into the corresponding inverter, which performs reactive power compensation based on the calculation results to complete voltage regulation.

6. A system implementing the rigid consensus method for multi-inverter coordination in a power distribution network according to any one of claims 1-5, characterized in that, Comprising: The correction coefficient introduction module is used for measuring the photovoltaic inverter node voltage, and the voltage value is fed back to the intelligent photovoltaic inverter of the control master node through the communication network. The correction coefficient of the slope is introduced for the droop control curve The collaborative optimization is performed; The optimization model definition module is used for estimating node voltage based on sensitivity analysis, establishing voltage constraints, and setting constraints of current, inverter, transformer and parameters, taking the maximum total power of the photovoltaic inverter as the target, defining the optimization model and solving it; The master node control module is used to adjust the parameters according to the correction coefficient. With target correction coefficient Determine the minimum rigidity graph of communication, set the desired distance between the current correction coefficient and the target correction coefficient, and then control the master node with edge computing capabilities; The dynamic reactive compensation module is used for converging the master node and the slave node to the optimization solution result through several iterations, and each inverter is based on the optimized correction coefficient to obtain a new reactive output vector , performing dynamic reactive compensation, and completing voltage regulation.

7. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor implements the steps of the rigid consensus method for multi-inverter coordination in a power distribution network according to any one of claims 1-5 when executing the program.

8. A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the rigid consensus method for multi-inverter coordination in a power distribution network according to any one of claims 1-5.

Images