Global optimization method, device and storage medium for grid-connected power

Abstract

The application discloses a global optimization method of grid-connected power, equipment and a storage medium, relates to the technical field of photovoltaic power stations, and discloses a global optimization method of grid-connected power, which comprises the following steps: acquiring circuit topology parameters of a transformer area power station and electrical parameters required by the transformer area power station; constructing an actual circuit topology according to the circuit topology parameters and the electrical parameters, wherein the actual circuit topology reflects the connection relationship of each cable line, each impedance and each inverter; deriving a global circuit model based on the actual circuit topology; solving the problem with the maximization of total power of a grid-connected point as the target, and constraining the global circuit model in the solving process to obtain optimal total power of the grid-connected point and optimal output currents of the inverters. Through the technical scheme, the global optimization of grid-connected power in a multi-inverter scene is realized, so that the overall power generation benefit of the photovoltaic power station is improved.

Description

Technical Field

[0001] This application relates to the field of photovoltaic power plant technology, and in particular to a method, device and storage medium for global optimization of grid-connected power. Background Technology

[0002] With the rapid development of photovoltaic power generation technology, the installed capacity of distributed photovoltaic power stations is increasing day by day, and multi-inverter grid connection has become the mainstream power station configuration. In actual projects, multiple inverters are connected in parallel to the grid connection point of the distribution area through main cables and branch cables, and complex electrical coupling relationships are formed between the inverters through line impedance.

[0003] Currently, traditional photovoltaic power plant control schemes generally adopt a single-unit independent modeling approach, where each inverter runs its own maximum power point tracking (MPPT) algorithm and is independently controlled to maximize its own output. However, this single-unit independent modeling approach makes it difficult to maximize the overall power generation efficiency of the power plant while ensuring equipment safety. Summary of the Invention

[0004] The main purpose of this application is to provide a method, device and storage medium for global optimization of grid-connected power, which aims to achieve global optimization of grid-connected power in multi-inverter scenarios, so as to improve the overall power generation efficiency of photovoltaic power plants.

[0005] To achieve the above objectives, this application proposes a global optimization method for grid-connected power, comprising: Obtain the circuit topology parameters and electrical parameters required for the operation of the power distribution substation; Based on the circuit topology parameters and electrical parameters, construct the actual circuit topology, which reflects the connection relationships of each cable line, each impedance, and each inverter. Based on the actual circuit topology, a global circuit model is derived, which is used to describe the electrical coupling relationship between each inverter in the transformer substation. The solution aims to maximize the total power at the grid connection point, and uses a global circuit model for constraints during the solution process to obtain the optimal total power at the grid connection point and the optimal output current of each inverter.

[0006] In one embodiment, the global circuit model derived based on the actual circuit topology includes: The actual circuit topology is converted into a directed graph topology, where nodes represent connection points with equal potential in the actual circuit topology, directional edges represent the current reference direction of each cable branch in the actual circuit topology, and edge weights represent the impedance value of each branch. Construct the correlation matrix, admittance matrix, voltage source matrix, current source matrix, current matrix, and voltage matrix based on the directed graph topology; Based on the correlation matrix, admittance matrix, voltage source matrix, current source matrix, current matrix, and voltage matrix, the nodal voltage matrix equation is derived using Kirchhoff's laws, serving as the global circuit model.

[0007] In one embodiment, converting a physical circuit topology into a directed graph topology includes: Each cable branch in the actual circuit topology is converted into a corresponding standard composite branch based on the standard composite branch model. The standard composite branch model includes a voltage source and an impedance connected in series, and a current source connected in parallel with the voltage source. Construct a directed graph topology based on each standard composite branch; The nodes of the directed graph topology include the grid connection point on the transformer substation side, the connection point between the main cable and the branch cable, the segment connection point on the branch cable, and the AC port access point of each inverter. The cable lines of the actual circuit topology include: main cable segment branches, branch cable segment branches, and inverter segment branches.

[0008] In one embodiment, constructing the correlation matrix, admittance matrix, voltage source matrix, current source matrix, current matrix, and voltage matrix based on the directed graph topology includes: Based on the connection relationships between nodes and branches in the directed graph topology and the current reference direction of the branches, construct the correlation matrix; Based on the impedance parameters of each branch in the directed graph topology, an admittance matrix is ​​constructed, where the voltage source matrix is ​​a column vector, and the elements in the column vector are the voltage values ​​of the voltage sources in the corresponding branches. Construct a current source matrix based on the current values ​​of the current sources contained in each branch of the directed graph topology; Construct a current matrix based on the current values ​​of each branch in the directed graph topology; Construct a voltage matrix based on the port voltage values ​​of each branch in the directed graph topology.

[0009] In one embodiment, the solution is performed with the objective of maximizing the total power at the grid connection point, and constraints are applied using a global circuit model during the solution process. The optimal total power at the grid connection point and the optimal output current of each inverter are obtained, including: With the goal of maximizing the total power at the grid connection point, the global circuit model is used as the equality constraint, the maximum output power and maximum output current of each inverter are used as the inequality constraint, and the preset voltage operating range of each inverter is used as the voltage constraint, a nonlinear programming model is constructed. The nonlinear programming model is solved iteratively to obtain the optimal output current of each inverter and the optimal total power at the grid connection point.

[0010] In one embodiment, the nonlinear programming model is iteratively solved to obtain the optimal output current of each inverter and the optimal total power at the grid connection point, including: The nonlinear programming model is solved, and the solution results are verified for business feasibility. The business feasibility verification is used to ensure that the solution results comply with the preset power grid dispatch rules and preset field operation and maintenance rules. If the solution passes the business feasibility verification, then the optimal total power of the grid connection point and the optimal output current of each inverter are obtained based on the solution. If the solution fails the business feasibility check, the nonlinear programming model is iterated again until the solution passes the business feasibility check. Based on the final solution that passes the check, the optimal total power at the grid connection point and the optimal output current of each inverter are obtained.

[0011] In one embodiment, the nonlinear programming model is iteratively solved to obtain the optimal output current of each inverter and the optimal total power at the grid connection point, including: The output current, output power, and voltage of each inverter are determined as optimization variables, and each constraint condition is taken as the feasible region of the optimization variables. Select a set of initial feasible points, wherein the output current of each inverter in the initial feasible points is less than or equal to the maximum output current of the corresponding inverter, the output power of each inverter is less than or equal to the maximum output power of the corresponding inverter, the voltage of each inverter is within the preset voltage operating range, and the global circuit model is satisfied. The global circuit model is incorporated into the objective function in the form of Lagrange multipliers, and the maximum output power, maximum output current and voltage constraints of each inverter are incorporated into the objective function in the form of logarithmic barrier functions to form an augmented objective function. Starting from the initial feasible point, the augmented objective function is iteratively searched along the Newton direction. After each iteration, the obstacle parameters are updated, and it is checked whether the current optimization variable is within the feasible region. If it is outside the feasible region, it is backed up to the boundary of the feasible region. When the set iteration termination condition is met, the iteration stops, and the output current of each inverter in the current optimization variables is taken as the optimal output current of each inverter. The optimal total power of the grid connection point is determined based on the output power of each inverter in the current optimization variables.

[0012] In one embodiment, after obtaining the optimal total power at the grid connection point and the optimal output current of each inverter, the method further includes: The output power of all inverters in pure maximum power point tracking mode is calculated and summed to obtain the total grid-connected power in pure maximum power point tracking mode. If the total grid connection power in pure maximum power point tracking mode is greater than or equal to the optimal grid connection power, and the current grid voltage and power fluctuations meet the preset compliance requirements, then each inverter will continue to operate in pure maximum power point tracking mode. If the total grid-connected power in pure maximum power point tracking mode is less than the optimal grid-connected power, or if the current grid voltage or power fluctuation does not meet the preset compliance requirements, then the optimal output current of each inverter will be sent to the corresponding inverter for execution.

[0013] In one embodiment, when issuing the optimal output current of each inverter to the corresponding inverter for execution, the method further includes: For each inverter, the inverter's optimal output current is compared with the maximum output current currently determined by the inverter's maximum power point tracking algorithm. If the optimal output current is less than or equal to the maximum output current, then the inverter is controlled to normally execute the corresponding optimal output current. If the optimal output current is greater than the maximum output current, the inverter's output current setting source is switched to maximum power point tracking loop operation to determine the inverter's maximum output current through the maximum power point tracking algorithm and control the inverter to execute the maximum output current.

[0014] In one embodiment, after iteratively solving the nonlinear programming model to obtain the optimal output current of each inverter and the optimal total power at the grid connection point, the method further includes: The optimal output current of each inverter is sent to the corresponding inverter for execution. Monitor the operating status of each inverter; When a fault is detected in any inverter causing it to disconnect from the grid, the global circuit model and the boundaries of other constraints are updated. Based on the updated global circuit model and other constraints, the new optimal total power at the grid connection point and the optimal output current are obtained by resolving the problem, and the optimal output current is then sent to the corresponding inverter in normal operating condition.

[0015] In addition, to achieve the above objectives, this application also proposes a global optimization device for grid-connected power, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the global optimization method for grid-connected power as described above.

[0016] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the global optimization method for grid-connected power as described above.

[0017] The one or more technical solutions proposed in this application have at least the following technical effects: An actual circuit topology is constructed based on the circuit topology parameters and electrical parameters of the photovoltaic power station; a global circuit model is then derived from this actual circuit topology; finally, based on this global circuit model as constraints, the optimal total power at the grid connection point and the optimal output current of each inverter are obtained by solving the problem with the goal of maximizing the total power at the grid connection point. Since the actual circuit topology reflects the connection relationships of each cable line, impedance, and inverter, the electrical coupling relationship between each inverter in the photovoltaic power station can be derived from this actual circuit topology. Using this electrical coupling relationship as a constraint, and combining it with the goal of maximizing the total power at the grid connection point, the optimal output current of each inverter is the result of global collaborative optimization, thereby maximizing the sum of the output of each inverter as a whole, rather than simply pursuing the maximization of the output of a single inverter. Thus, without changing the hardware capabilities of any inverter, the optimal grid connection power in a multi-inverter scenario is achieved through global modeling and collaborative solving, thereby improving the overall power generation efficiency of the photovoltaic power station. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating the first embodiment of the global optimization method for grid-connected power in this application; Figure 2 This is a schematic diagram of the actual circuit topology in the embodiments of this application; Figure 3 This is another flowchart illustrating the first embodiment of the global optimization method for grid-connected power in this application; Figure 4 This is a schematic diagram of a standard composite branch model in an embodiment of this application; Figure 5 This is a schematic diagram of the directed graph topology in an embodiment of this application; Figure 6 A flowchart illustrating the second embodiment of the global optimization method for grid-connected power in this application; Figure 7 This is a flowchart illustrating the third embodiment of the global optimization method for grid-connected power in this application.

[0021] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0022] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0023] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0024] Currently, traditional photovoltaic power plant control schemes generally adopt a single-unit independent modeling approach, where each inverter operates its own Maximum Power Point Tracking (MPPT) algorithm, independently controlling itself to maximize its own output. However, because this single-unit independent modeling approach only focuses on the voltage and current states of each inverter, when its voltage exceeds limits, it can only passively take emergency measures such as reducing the power output of a single unit or shutting it down. This makes it difficult to maximize the overall power generation efficiency of the power plant while ensuring equipment safety. Furthermore, due to the line impedance of the main and branch cables, the terminal voltages of each inverter are not independent but mutually coupled. When the terminal inverter operates at maximum power, its injected current generates a voltage drop across the line impedance, causing the terminal voltage of the front-end inverter to rise. When the voltage exceeds the inverter's allowable operating range, the front-end inverter is forced to operate at reduced power, or even triggers overvoltage protection and disconnects from the grid, resulting in curtailment. Conversely, if the output of the terminal inverter is limited to reduce the front-end voltage, the terminal inverter cannot fully utilize its power generation capacity, also causing power generation losses. This multi-inverter voltage and current coupling effect caused by line voltage drops can lead to inherent power generation losses in the power plant.

[0025] To address the aforementioned issues, this application proposes a global optimization method for grid-connected power. The main technical solution includes obtaining the circuit topology parameters of the transformer substation and the electrical parameters required for the operation of the transformer substation. Based on the circuit topology parameters and electrical parameters, construct the actual circuit topology, which reflects the connection relationships of each cable line, each impedance, and each inverter. Based on the actual circuit topology, a global circuit model is derived, which is used to describe the electrical coupling relationship between each inverter in the transformer substation. The solution aims to maximize the total power at the grid connection point, and uses a global circuit model for constraints during the solution process to obtain the optimal total power at the grid connection point and the optimal output current of each inverter.

[0026] By constructing an actual circuit topology based on the circuit and electrical parameters of the photovoltaic power station, a global circuit model is derived. This model is then used as a constraint, and the optimal total grid-connected power and the optimal output current of each inverter are obtained by maximizing the total power at the grid connection point. Since the actual circuit topology reflects the connections of various cable lines, impedances, and inverters, the electrical coupling relationships between inverters in the photovoltaic power station can be derived from it. Using these electrical coupling relationships as constraints, and combining them with the goal of maximizing the total grid-connected power, the optimal output current of each inverter is the result of global collaborative optimization. This maximizes the overall output of each inverter, rather than simply maximizing the output of a single inverter. Thus, without changing any inverter hardware capabilities, optimal grid-connected power in multi-inverter scenarios is achieved through global modeling and collaborative solving, thereby improving the overall power generation efficiency of the photovoltaic power station.

[0027] It should be noted that the execution subject of this embodiment can be a global optimization device for grid-connected power with data processing, network communication and program running functions. The global optimization device for grid-connected power can be a computing service device or an electronic device that can realize the above functions.

[0028] Based on this, embodiments of this application provide a global optimization method for grid-connected power, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the global optimization method for grid-connected power according to this application. In this embodiment, the global optimization method for grid-connected power includes steps S10 to S40: Step S10: Obtain the circuit topology parameters of the transformer substation and the electrical parameters required for the operation of the transformer substation.

[0029] A transformer substation refers to a photovoltaic power generation system consisting of multiple inverters and their connecting cable lines. A transformer substation typically includes a distribution transformer, multiple branch cable lines, and multiple distributed inverters. These inverters are connected in parallel via cable lines and then fed into the low-voltage busbar of the distribution transformer. The following text uses this transformer substation as an example of a photovoltaic power station.

[0030] Circuit topology parameters can include information on the distribution of trunk cables, branch cables, and impedance. They can also include the number of branch cables, the connection location of each inverter on the branch circuit, the location distribution of each cable connection point, the connection sequence of each cable segment, and the impedance locations on the trunk cable and at the connection points of each branch circuit. Furthermore, these parameters can include the length and cross-sectional area of ​​the trunk cable, the length and cross-sectional area of ​​the branch cables, the material and laying method of each cable segment, and the distance between each inverter.

[0031] Electrical parameters refer to various electrical quantity data required for the normal operation of a photovoltaic power station. Specifically, electrical parameters include the grid voltage amplitude and frequency at the grid connection point on the transformer substation side, the maximum output power Pmax(M, Ni) of each inverter calculated in real time by the maximum power point tracking algorithm, and the maximum output current Imax(M, Ni) of each inverter. These electrical parameters may also include information such as the resistance per unit length of each cable segment, the equivalent impedance of each inverter output filter, and the system equivalent short-circuit impedance of the grid at the grid connection point.

[0032] The circuit topology parameters and electrical parameters required for the operation of the aforementioned power station can be obtained through on-site measurement, querying the system wiring diagram and design drawings of the photovoltaic power station.

[0033] Understandably, circuit topology parameters determine whether the subsequently constructed global circuit model can accurately reflect the physical connections of the actual power plant, while electrical parameters determine the output limits of each inverter and the boundary conditions of the power grid during subsequent optimization calculations. By employing the methods described above, circuit topology and electrical parameters can be accurately obtained, improving the accuracy of the subsequent actual circuit topology.

[0034] Step S20: Construct the actual circuit topology based on the circuit topology parameters and electrical parameters. The actual circuit topology reflects the connection relationships of each cable line, each impedance, and each inverter.

[0035] Actual circuit topology refers to the circuit connection diagram formed by abstracting the electrical components into circuit symbols according to the actual physical connection relationship of a photovoltaic power station. Actual circuit topology is not a mathematical abstract model, but rather an intermediate expression between physical devices and mathematical models. In actual circuit topology, the main cable is represented as a series-connected resistive element, each segment of the branch cable is represented as an independent resistive element, each inverter is represented as a current source element, each connection point is represented as a circuit node, and the grid connection point is represented as a voltage source element. Actual circuit topology completely retains all connection information and component parameters of the original circuit, but has not yet undergone graph theory and matrix representation mathematical processing. Each inverter has its own independent maximum output power and maximum output current, which can be calculated in real time by the inverter's internal MPPT algorithm based on the current light intensity and temperature conditions. Inverters are connected to the regional power grid in parallel via cable lines, and the inverters are electrically coupled to each other through a common cable network.

[0036] Combination Figure 2 , Figure 2The actual circuit topology diagram of this application shows that the resistance of each line segment can be defined as follows: one main cable corresponds to M branch cables, and the line resistance of the main cable and each branch cable is R1~RM respectively; Ni photovoltaic power stations are connected to each branch cable, and the line resistance of each segment of the corresponding branch cable is RM1~RMN. i The connection resistance of each photovoltaic power station's access branch cable is R(M, N). i The current and voltage definitions of each branch in the circuit correspond one-to-one with the resistance definitions, and all follow the above-mentioned marking rules.

[0037] Cable lines refer to the power cables used to transmit electrical energy in a photovoltaic power station. In a photovoltaic power station, cable lines are divided into main cable segments, branch cable segments, and inverter segments. Main cable segments originate from the grid connection point of the photovoltaic power station and are laid along the main path; they carry the total current flowing into all branch cables. Branch cable segments originate from various tap points on the main cable and connect to each inverter. Each branch cable segment is further divided into several sections by the connection points of each inverter, and each section has its own independent resistance value.

[0038] Impedance refers to the electrical parameters in a circuit that impede the flow of current. In this embodiment, impedance mainly includes three categories. The first category is the resistance of the branch cable segment, which is determined by the cable's material, length, and cross-sectional area, and is the main cause of line voltage drop. The second category is the equivalent impedance located in the inverter segment branch, which exists inside the inverter and affects the relationship between the inverter port voltage and output current. The third category is the system equivalent impedance of the power grid at the grid connection point of the main cable segment branch, which represents the equivalent internal resistance from the grid connection point to the grid side, affecting the degree to which the grid connection point voltage changes with injected power.

[0039] In one alternative approach, after obtaining the circuit topology parameters and electrical parameters, the actual circuit topology of the photovoltaic power station can be drawn in computer-aided design software or circuit simulation software.

[0040] Another alternative approach involves constructing the actual circuit topology using tables and parameter lists, without relying on software. Specifically, this can be achieved by creating a data structure containing a node list and a branch list. Each row in the node list records the node number and node type, including grid connection point type, trunk branch connection point type, branch segment point type, and inverter port type. Each row in the branch list records the branch number, branch type, branch start node number, branch end node number, and branch electrical parameters. Branch types include trunk cable segment branches, branch cable segment branches, and inverter segment branches. Branch electrical parameters include the branch's resistance, reactance, embedded voltage source value, and embedded current source value. By filling in the data in the node and branch lists, the obtained circuit topology parameters and electrical parameters are structurally stored in memory, forming an actual circuit topology that can be used in subsequent steps.

[0041] Step S30: Based on the actual circuit topology, a global circuit model is derived, wherein the global circuit model is used to describe the electrical coupling relationship between each inverter in the power distribution area.

[0042] A global circuit model is a model that describes the electrical characteristics of an entire power distribution substation in the form of mathematical equations. It is not a simple circuit diagram, but a set of simultaneous matrix equations. The global circuit model connects the relationships between all node voltages and all branch currents in the actual circuit topology using a unified matrix expression. By solving this matrix equation, the voltage values ​​of all nodes and the current values ​​of all branches can be calculated under any set of inverter output currents.

[0043] Electrical coupling refers to the mutual influence between inverters in a power distribution substation via shared cables. Specifically, when an inverter increases its output current, the current flowing through the cable creates an additional voltage drop. This voltage drop propagates along the cable, causing changes in the voltage at the connection points of other inverters. For example, when an inverter at the end of a branch cable increases its output, the voltage drop generated by its injected current on the branch cable will increase the voltage at the node closer to the main branch, thus affecting the terminal voltage of other inverters connected to the same branch, and even impacting inverters on other branches via the main cable.

[0044] In one alternative approach, directed graph modeling from graph theory can be used to transform the actual circuit topology into a directed graph topology, and then a global circuit model can be derived based on this directed graph topology. This directed graph topology consists of nodes, edges, and edge weights. Each connection point in the actual circuit topology can be abstracted as a node in the directed graph, with node numbers corresponding one-to-one with node numbers in the actual circuit topology. Each branch in the actual circuit topology is abstracted as a directed edge in the directed graph, with the edge direction set as the reference direction of the branch current. By constructing a global circuit model, the multi-inverter voltage coupling relationships, which are difficult to quantify in traditional schemes, are transformed into mathematical equations, providing a clear computational basis for subsequent optimization solutions.

[0045] Step S40: Solve the problem with the goal of maximizing the total power at the grid connection point, and constrain the solution using a global circuit model to obtain the optimal total power at the grid connection point and the optimal output current of each inverter.

[0046] The total grid-connected power refers to the sum of the output power injected into the grid by all inverters at the grid-connected point of the distribution area. The total grid-connected power can be calculated as the sum of the output power of each inverter, or as the sum of the output power of each inverter minus the power consumed in all cable lines. Since power losses in cable lines are proportional to the square of the current, the magnitude of the total grid-connected power depends not only on the total output of each inverter but also on how that output is distributed among the inverters. With the same total output, allocating more output to inverters closer to the grid-connected point reduces line losses, thereby increasing the total grid-connected power.

[0047] Optimal grid-connected total power refers to the maximum possible value of the total grid-connected power obtained through optimization calculations, under the premise of satisfying all constraints. Optimal grid-connected total power is a theoretical maximum value, representing the highest power generation level achievable through optimal output allocation under the current power plant topology. Optimal output current refers to the current value that each inverter should output to maximize the total grid-connected power. Optimal output current is the final output result of the global optimization calculation, telling each inverter how much current it should generate under the current operating conditions.

[0048] In one alternative approach, the interior-point method can be used to solve the optimization problem. Specifically, the objective function for maximizing the total power at the grid connection point can be expressed as a mathematical expression for the sum of the output power of each inverter. The global circuit model is incorporated as equality constraints into the optimization model. The maximum output power and maximum output current of each inverter are incorporated as inequality constraints into the optimization model. The output current and output power of each inverter are set as the optimization variables to be solved. Initial parameters for the interior-point method are set, including the initial penalty factor and the iteration convergence accuracy. The inequality constraints are incorporated into the objective function in the form of a logarithmic barrier function to form an augmented objective function. Starting from the initial feasible point, the augmented objective function is iteratively searched along the Newton direction. After each iteration, the barrier parameters are updated, and it is checked whether each optimization variable to be solved satisfies all constraints. The iteration stops when the preset maximum number of iterations is reached. The optimal total power at the grid connection point is calculated based on the optimal output power of each inverter, and the current iteration result is taken as the optimal output current of each inverter.

[0049] Another alternative approach is to use a particle swarm optimization (PSO) algorithm to solve the optimization problem. Specifically, this involves: encoding the output current of each inverter into a position vector for each particle in the swarm, where the dimension of each particle equals the total number of inverters. Parameters such as the swarm size, maximum number of iterations, inertia weight, and learning factor are set. The positions and velocities of all particles are randomly initialized, ensuring that each particle's position is within the range of the maximum output current of each inverter. For each particle, the output current of each inverter represented by its position vector is substituted into the global circuit model to calculate the voltage value of each inverter, and it is checked whether any inverter's voltage value exceeds the preset operating voltage range. For particles that satisfy all constraints, the total power of their corresponding grid connection point is calculated as the particle's fitness value. The individual optimal position of each particle and the global optimal position of the entire swarm are updated based on the fitness value. The velocity and position of all particles are updated according to the velocity and position update formulas of the PSO algorithm. These steps are repeated until the maximum number of iterations is reached or the fitness value of the global optimal position no longer improves after multiple consecutive iterations. Finally, the optimal total power at the grid connection point is calculated based on the optimal output power of each inverter, and the current iteration result is used as the optimal output current of each inverter.

[0050] In this embodiment, an actual circuit topology is constructed based on the circuit topology and electrical parameters of the power station. A global circuit model is then derived from this actual circuit topology. Finally, based on this global circuit model as constraints, the optimal total power at the grid connection point and the optimal output current of each inverter are obtained by solving the problem with the objective of maximizing the total power at the grid connection point. Since the actual circuit topology reflects the connection relationships of each cable line, impedance, and inverter, the electrical coupling relationships between the inverters in the power station can be derived from this actual circuit topology. Using this electrical coupling relationship as a constraint, and combining it with the objective of maximizing the total power at the grid connection point, the optimal output current of each inverter is the result of global collaborative optimization. This maximizes the sum of the output of each inverter as a whole, rather than simply pursuing the maximization of the output of a single inverter. Thus, without changing any inverter hardware capabilities, the optimal grid connection power in a multi-inverter scenario is achieved through global modeling and collaborative solving, thereby improving the overall power generation efficiency of the photovoltaic power station.

[0051] In one feasible implementation, refer to Figure 3 Step S30 may include steps S31 to S33: Step S31: Convert the actual circuit topology into a directed graph topology.

[0052] Directed graph topology is a mathematical structure used to abstract the connections in physical circuits into a directed graph form in graph theory. In directed graph topology, each electrical connection point in the physical circuit is abstracted as a node, and each electrical path is abstracted as an edge with a direction indicator. The direction of each edge represents the preset reference direction of the current in that branch, rather than the actual current flow direction. That is, directed graph topology uses nodes to represent connection points with equal potential in the actual circuit topology, uses directed edges to represent the current reference direction of each cable branch in the actual circuit topology, and uses edge weights to represent the impedance value of each branch.

[0053] In a directed graph topology, a node is an abstract representation of an electrical connection point with equal potential in the actual circuit topology. Each node corresponds to a physical location in the actual circuit, where all connected wires have the same potential. Nodes form the row dimension basis for constructing the correlation matrix, and each node corresponds to an unknown node voltage. A directed edge is an abstract representation of each branch in the actual circuit topology. Each edge connects two nodes and has a predefined directional arrow. This direction does not represent the actual direction of current flow, but rather provides a unified positive or negative sign reference for subsequently writing Kirchhoff's current law equations. When the actual current direction is consistent with the predefined reference direction, the branch current takes a positive value; when the actual current direction is opposite to the predefined reference direction, the branch current takes a negative value. The current reference direction is a manually defined reference direction used to determine the sign of the current. When constructing a directed graph topology, a reference direction needs to be specified for each edge, for example, by agreeing to point from a node with a smaller number to a node with a larger number. Once the current reference direction is set, it remains fixed throughout the modeling process and does not change with the actual current. The direction of the actual current is determined by the sign of the current in that branch in the solution results. Edge weight is a numerical attribute assigned to each edge of the directed graph topology, used to characterize the electrical parameters of the physical branch corresponding to that edge. In this embodiment, the edge weight mainly refers to the impedance value of the branch, including the resistance value of the cable line, the equivalent impedance value of the inverter output filter, and the system equivalent impedance value of the power grid at the grid connection point.

[0054] In one alternative approach, each cable branch in the actual circuit topology can be converted into a corresponding standard composite branch based on a standard composite branch model. The standard composite branch model includes a voltage source and an impedance connected in series, and a current source connected in parallel with the voltage source. A directed graph topology is constructed based on each standard composite branch. The nodes of the directed graph topology include the grid connection point on the transformer substation side, the connection point between the main cable and the branch cable, the segment connection point on the branch cable, and the AC port access point of each inverter. The cable lines in the actual circuit topology include: main cable segment branches, branch cable segment branches, and inverter segment branches.

[0055] The standard composite branch model is a general equivalent representation of circuit branches, uniformly representing any type of electrical branch as a combination of three basic components: a voltage source connected in series, an impedance connected in series, and a current source connected in parallel. The standard composite branch model is as follows: Figure 4 As shown. Based on this standard composite branch model, the final directed graph topology can be obtained as follows. Figure 5 As shown.

[0056] A standard composite branch is a branch representation obtained by equivalently representing a specific branch in an actual circuit according to a standard composite branch model. Each standard composite branch has three defined parameters: the value of the series voltage source, the value of the series impedance, and the value of the parallel current source.

[0057] For main cable segment branches, the series voltage source of the standard composite branch is the grid connection point voltage on the transformer substation side, the impedance is the equivalent impedance of the power grid system, and the parallel current source is zero. For branch cable segment branches, the series voltage source of the standard composite branch is zero, the parallel current source is zero, and the impedance is the resistance value of the cable line. For inverter segment branches, the series voltage source of the standard composite branch is zero, the impedance is the equivalent impedance value of the inverter output filter, and the parallel current source is the equivalent output current of the inverter.

[0058] After obtaining the standard composite branches using the above method, a node list and edge list of the directed graph topology are constructed based on the connection relationships of the generated standard composite branches. First, the node list is constructed: traverse the endpoints of all standard composite branches, collect all unique endpoints, and assign a unique node number to each endpoint. The node numbering order is as follows: first, assign the numbers of the grid connection points on the transformer substation side; then, assign the numbers of the connection points between the main trunk cable and the branch cable in order from the main trunk to the branch; next, assign the numbers of the segment connection points on the branch cable in order from closest to furthest from the main trunk; finally, assign the numbers of the AC port access points of each inverter according to the relationship corresponding to the segment connection points of the branch cable. Then, the edge list is constructed: traverse each standard composite branch, search for the corresponding node number in the node list based on the starting and ending endpoints of the branch, and convert the branch into an edge in the directed graph topology. The starting node number of an edge is the node number corresponding to the starting endpoint of the branch, and the ending node number is the node number corresponding to the ending endpoint of the branch. The direction of the edge is set from the starting node to the ending node, and the impedance value of the edge is taken from the impedance field value of the standard composite branch. After the node list and edge list are constructed, they are stored in memory to form a complete directed graph topology data structure, which facilitates the derivation of the subsequent global circuit model.

[0059] In another alternative approach, a directed graph data structure is created in memory, containing a list of nodes and a list of edges. Each row in the node list records a node's ID, node type, and node name. Node types include grid connection point type, trunk branch connection point type, branch segment point type, and inverter port type. Each row in the edge list records an edge's ID, edge type, start node ID, end node ID, reference direction flag, and impedance value. Edge types include trunk cable type, branch cable type, and inverter equivalent type. When populating the node list, nodes are created sequentially according to the connection point order in the actual circuit topology, and each node is assigned a unique ID. When populating the edge list, edges are created sequentially according to the branch connection relationships in the actual circuit topology, specifying a start node and an end node for each edge, and setting the reference direction of the edge to point from the start node to the end node. Simultaneously, the impedance values ​​of the corresponding branches are filled into the edge's impedance value field. After the node list and edge list are filled, the directed graph data structure completely stores the topological connections and electrical parameters of the actual circuit topology.

[0060] Step S32: Construct the correlation matrix, admittance matrix, voltage source matrix, current source matrix, current matrix, and voltage matrix based on the directed graph topology.

[0061] The incidence matrix is ​​a two-dimensional matrix used to describe the connections between nodes and edges in a directed graph topology, as well as the reference directions of the edges. The number of rows in the incidence matrix equals the total number of nodes in the directed graph, and the number of columns equals the total number of edges. Each element in the incidence matrix is ​​assigned a value according to the following rules: if the reference direction of an edge is from a node, the elements in the corresponding row and column of that edge are both 1; if the reference direction of an edge is into a node, the elements in the corresponding row and column of that edge are both -1; if an edge has no direct connection to a node, the elements in the corresponding row and column of that edge are both 0.

[0062] The admittance matrix is ​​a diagonal matrix used to store the admittance values ​​of each edge in a directed graph topology. The number of rows and columns of the admittance matrix equals the total number of edges in the directed graph. The elements on the diagonal are non-zero, and the elements off-diagonal are all zero. The k-th element on the diagonal of the admittance matrix is ​​equal to the admittance value of the k-th edge, which is the reciprocal of the impedance value of that edge.

[0063] The voltage source matrix is ​​a column vector used to store the voltage source values ​​embedded in each edge of a directed graph topology. The number of rows in the voltage source matrix equals the total number of edges in the directed graph, and each element corresponds to the voltage value of a series voltage source embedded in an edge. If an edge does not contain any embedded voltage sources, the value of the corresponding element is zero.

[0064] The current source matrix is ​​a column vector used to store the values ​​of the current sources embedded in each edge of the directed graph topology. The number of rows in the current source matrix is ​​equal to the total number of edges in the directed graph, and each element corresponds to the current value of the parallel current source embedded in one edge. If an edge does not contain any embedded current sources, the value of the element corresponding to that edge is zero. In this embodiment, the standard composite branch of the inverter segment contains non-zero embedded current sources, the value of which is the equivalent output current of the inverter, while the embedded current sources of the other branches are all zero.

[0065] The current matrix is ​​a column vector used to store the branch current values ​​of each edge in a directed graph topology. The number of rows in the current matrix equals the total number of edges in the directed graph, and each element corresponds to the branch current of one edge. The elements in the current matrix are the variables to be solved, and they are continuously updated during the optimization process. The current matrix elements in the final optimal solution are the current values ​​of each branch.

[0066] The voltage matrix is ​​a column vector used to store the branch port voltage values ​​of each edge in a directed graph topology. The number of rows in the voltage matrix equals the total number of edges in the directed graph, and each element corresponds to the port voltage of one edge. The elements in the voltage matrix can be obtained by multiplying the node voltage matrix by the transpose of the incidence matrix; that is, the port voltage of each edge is equal to the difference between the voltages of the two nodes at the two ends of the edge.

[0067] In one optional approach, an affinity matrix is ​​constructed based on the connection relationships between nodes and branches in the directed graph topology and the current reference direction of the branches. In this matrix, rows correspond to nodes, and columns correspond to branches. If the current reference direction of a branch flows out of a node, the corresponding element takes a first preset value; if the current reference direction of a branch flows into a node, the corresponding element takes a second preset value; if the branch and node are not associated, the corresponding element takes a third preset value. An admittance matrix is ​​constructed based on the impedance parameters of each branch in the directed graph topology. This admittance matrix is ​​a diagonal matrix, and the diagonal elements are the reciprocals of the impedances of the corresponding branches. The matrix is ​​further constructed based on the impedance parameters of each branch in the directed graph topology. Based on the voltage values ​​of the voltage sources, construct a voltage source matrix, where the voltage source matrix is ​​a column vector and the elements in the column vector are the voltage values ​​of the voltage sources in the corresponding branches; based on the current values ​​of the current sources contained in each branch of the directed graph topology, construct a current source matrix, where the current source matrix is ​​a column vector and the elements in the column vector are the current values ​​of the current sources in the corresponding branches; based on the current values ​​of each branch of the directed graph topology, construct a current matrix, where the current matrix is ​​a column vector and the elements in the column vector are the current values ​​of the corresponding branches; based on the port voltage values ​​of each branch of the directed graph topology, construct a voltage matrix, where the voltage matrix is ​​a column vector and the elements in the column vector are the port voltage values ​​of the corresponding branches.

[0068] Step S33: Based on the correlation matrix, admittance matrix, voltage source matrix, current source matrix, current matrix, and voltage matrix, the node voltage matrix equation is derived using Kirchhoff's laws, and used as the global circuit model.

[0069] Kirchhoff's laws are two fundamental laws in circuit analysis, including Kirchhoff's current law and Kirchhoff's voltage law. Kirchhoff's current law states that at any given moment, the sum of the currents flowing into any node in a circuit is equal to the sum of the currents flowing out of that node; that is, the algebraic sum of the currents at a node is zero. Kirchhoff's voltage law states that at any given moment, the algebraic sum of the voltages along any closed loop in a circuit is zero. In this embodiment, Kirchhoff's current law is primarily used to establish the node voltage matrix equations.

[0070] The node voltage matrix equation is the mathematical expression of the global circuit model. It expresses the relationship between all node voltages, all branch currents, and all excitation sources in matrix form. This equation simplifies the complex circuit network into a unified linear algebraic system. By solving this equation, the voltage values ​​of all nodes can be obtained under a given inverter output current.

[0071] In one alternative approach, the first step is to apply Kirchhoff's Current Law: the correlation matrix multiplied by the current matrix equals a zero matrix, meaning the algebraic sum of the currents flowing into all nodes is zero. This is the matrix representation of Kirchhoff's Current Law. The second step is to apply the voltage-current relationship equation for a composite branch: the current matrix equals the admittance matrix multiplied by the voltage matrix minus the voltage source matrix, plus the current source matrix. This equation expresses the relationship between the current in each branch and the branch's port voltage, embedded voltage sources, and embedded current sources. The third step is to apply the relationship between node voltages and branch voltages: the voltage matrix equals the transpose of the correlation matrix multiplied by the node voltage matrix, meaning the port voltage of each branch equals the difference between the voltages at the two nodes of that branch. The fourth step is to substitute the voltage matrix expression from the third step into the branch current expression from the second step, obtaining the current matrix equal to the admittance matrix multiplied by the transpose of the correlation matrix multiplied by the node voltage matrix minus the admittance matrix multiplied by the voltage source matrix, plus the current source matrix. Fifth, substitute the current matrix expression from step four into the Kirchhoff's current law expression from step one, resulting in the equation: the correlation matrix multiplied by the admittance matrix multiplied by the transpose of the correlation matrix multiplied by the node voltage matrix. Subtracting the correlation matrix multiplied by the admittance matrix multiplied by the voltage source matrix and adding the correlation matrix multiplied by the current source matrix equals zero. Sixth, rearrange the expression from step five, resulting in the equation: the correlation matrix multiplied by the admittance matrix multiplied by the transpose of the correlation matrix multiplied by the node voltage matrix equals the correlation matrix multiplied by the current source matrix minus the correlation matrix multiplied by the admittance matrix multiplied by the voltage source matrix. This expression is the node voltage matrix equation.

[0072] For example, the number of nodes in a global circuit model is The number of branches is Construct an association matrix (A), where the association matrix represents the number of nodes. Number of branches. If the current in a branch flows out of the node, use 1; if the current in a branch flows into the node, use -1; if the branch and node are not related, use 0. Assign nodes U0, U1…UM, U… Z U11, U12…UMN M ,U(1,1),U(1,2)…U(M,N M Sort the branches according to i1, i2…iM, i11, i12…iMN. M i R (1,1),i R (1,2)...i R (M, N) M ),i(1,1),i(1,2)…i(M,N M Sort the data. Construct the association matrix (A) as follows: A= .

[0073] Construct a current matrix (I) based on the current of each branch, I = [i1, i2…iM, i11, i12…iMN]. M i R (1,1),i R (1,2)...i R (M, N) M ),i(1,1),i(1,2)…i(M,N M )] T According to Kirchhoff's current law, AI = 0. Combining this with the node voltage equation, a mapping relationship can be established between the inverter output current and the currents of each branch and the voltages of each node.

[0074] Construct a voltage matrix (U) for each branch, U = [u1, u2…uM, u11, u12…uMN]. M u R (1,1),u R (1,2)...u R (M, N) M ),u(1,1),u(1,2)…u(M,N M )] T The voltages of each node are used to construct a voltage matrix (U). n ), U n =[U0, U1…UM, U Z U11, U12…UMN M ,U(1,1),U(1,2)…U(M,N M )]T, then U=AT U n .

[0075] The voltage sources of each branch form a voltage source matrix (U s ), Us=[us1, us2…usM, us11, us12…usMN M us R (1,1),us R (1,2)...us R (M, N) M ),us(1,1),us(1,2)…us(M,N M )] T The current sources of each branch form a current source matrix (I). s ), I s =[is1, is2…isM, is11, is12…isMN M is R (1,1), is R (1,2)...is R (M, N) M ), is (1, 1), is (1, 2)...is (M, N M )] T .

[0076] The admittances of each branch constitute the admittance matrix (Y), Y = diag[Y1, Y2…YM, Y11, Y12…YMN]. M Y R (1,1),Y R (1,2)...Y R (M, NM), Y (1, 1), Y (1, 2)…Y (M, N M )].

[0077] From the voltage and current equations of a standard composite branch, we can solve the system of equations: AI = 0, U = A T U n Therefore, the relationship of the nodal voltage matrix equation Un is: AYA T U n =AI s -AYU s The node voltage matrix equation can be used as the global circuit model in this embodiment.

[0078] In this embodiment, during the derivation of the global circuit model based on the actual circuit topology, the actual circuit topology is converted into a directed graph topology. Based on the directed graph topology, an association matrix, admittance matrix, voltage source matrix, current source matrix, current matrix, and voltage matrix are constructed. Finally, based on the association matrix, admittance matrix, voltage source matrix, current source matrix, current matrix, and voltage matrix, the node voltage matrix equation is derived using Kirchhoff's laws, serving as the global circuit model. This method transforms the simulated actual circuit topology into a computer-processable discretized directed graph structure. The matrices constructed from this directed graph structure are then organically combined into a unified node voltage matrix equation, completing the transformation from scattered matrix data to a standard circuit model. This node voltage matrix equation enables subsequent optimization solutions to mathematically accurately describe the coupling relationship between the output of each inverter and the voltage of each node.

[0079] In one feasible implementation, step S40 may include steps S41-S42: Step S41: With the goal of maximizing the total power at the grid connection point, the global circuit model is used as the equality constraint, the maximum output power and maximum output current of each inverter are used as the inequality constraint, and the preset voltage operating range of each inverter is used as the voltage constraint, a nonlinear programming model is constructed.

[0080] During the solution process, the global circuit model can serve as an equality constraint. Other constraints can also be combined, including the maximum output power and maximum output current of each inverter, and the preset voltage operating range of each inverter, to constrain the solution process. The following will introduce each constraint: Maximum output current constraints for each inverter: Each current source inverter is limited by the device's MPPT and hardware rating parameters; the maximum output current is Imax(M, N). i Meanwhile, considering the scenario where the output current of the current source inverter is 0 (fault disconnection), it can be analyzed into the equation I(M, N). i ) (I(M, N) i )-Imax(M, N i ))≤0.

[0081] Maximum output power constraints for each inverter: In actual operation of a photovoltaic power station, the real-time maximum output power Pmax(M, N) of each current source inverter is obtained using the MPPT algorithm. i ), which can be analyzed into the equation P(M, N) i )-Pmax(M, N i )≤0.

[0082] Preset voltage operating range constraints for each inverter: According to the technical regulations for photovoltaic power generation systems connected to the distribution network, the preset voltage operating range for inverters connected to the distribution network is 0.85U. N ~1.1U N Considering the possibility of the inverter disconnecting from the grid, the equation is analytically expressed as U(M, N). i ) (U(M, N) i )-1.1U N ) (U(M, N) i -0.85U N )≤0.

[0083] Besides the constraints mentioned above, the photovoltaic power station's grid-connected circuit is a symmetrical three-phase circuit, and the inverter operates with unity power factor. This embodiment can uniformly use line voltage and line current for calculations. Based on the power relationship of the three-phase circuit, the active power satisfies the formula. The analytical expression is P(M, N) i )- U(M,N i ) I(M, N) i )=0, and this is also used as a constraint in the solution process.

[0084] Step S42: Iteratively solve the nonlinear programming model to obtain the optimal output current of each inverter and the optimal total power at the grid connection point.

[0085] In one optional approach, a nonlinear programming model can be solved, and the solution results can be verified for business feasibility. If the solution passes the business feasibility verification, the optimal total power at the grid connection point and the optimal output current of each inverter are obtained based on the solution results. If the solution fails the business feasibility verification, the nonlinear programming model is iteratively solved again until the solution passes the business feasibility verification. Based on the final verified solution results, the optimal total power at the grid connection point and the optimal output current of each inverter are obtained. Through business feasibility verification, solutions that are mathematically optimal but not engineering-feasible can be automatically identified and eliminated, ensuring that the final issued instructions are optimal, compliant, safe, and executable, thereby improving the practicality and reliability of the optimal output current of each inverter.

[0086] Business feasibility verification, in particular, is a check of the engineering practicality of the solution results after the mathematical optimization solution is completed. The purpose of business feasibility verification is to ensure that the solution results are not only mathematically feasible, but also executable, compliant, and safe in engineering practice. The content of business feasibility verification includes, but is not limited to, two aspects: verification of power grid dispatching rules and verification of on-site operation and maintenance rules.

[0087] The preset grid dispatch rules are technical specifications and requirements that photovoltaic power plants must comply with during grid-connected operation, formulated by the power grid company or power dispatching agency. These rules specifically include limitations on power fluctuations at the grid connection point; allowable voltage deviation ranges at the grid connection point; power factor requirements at the grid connection point; and functional requirements for inverters to participate in grid frequency and voltage regulation. Verification of the preset grid dispatch rules ensures that the optimization results will not trigger grid alarms or protection actions. The preset on-site operation and maintenance rules are technical specifications and requirements formulated by the photovoltaic power plant's operation and management managers to guide the daily operation and maintenance of the power plant. These rules specifically include current-carrying capacity limits for each cable section to ensure that cables do not overheat and become damaged due to overcurrent; inverter load balancing requirements to prevent individual inverters from operating at extremely low loads for extended periods, thus affecting equipment lifespan; and limits on the number of inverter start-ups and shutdowns to prevent frequent start-ups and shutdowns from impacting the equipment. Verification of the preset on-site operation and maintenance rules ensures that the optimization results are safe and reliable during long-term operation.

[0088] The aforementioned business feasibility verification of the solution results includes: reading data such as the optimal output current of each inverter, the voltage of each node, and the total power of the grid connection point from the solution results. Then, the module checks each item according to preset grid dispatch rules. The first check is the grid connection point power fluctuation verification. This module compares the total grid connection point power in the current solution result with the total grid connection point power in the previous solution result, calculates the power change rate, and if the power change rate is greater than the preset change rate, the verification fails. The second check is the grid connection point voltage verification. This module checks whether the grid connection point voltage is within the preset range of the rated voltage. If it exceeds this range, the verification fails. The third check is the cable current carrying capacity verification. This module compares the current value of each branch with the rated current carrying capacity of the cable. If the current of any cable exceeds the preset current carrying capacity, the verification fails. The fourth check is the inverter load rate verification. This module checks the ratio of the output power of each inverter to its rated power. If the load rate of any inverter is less than the preset load rate for a duration exceeding the preset time, the verification is deemed to have failed. After all verification items are completed, the business feasibility verification module outputs a verification result flag. A true flag indicates that all verification items have passed, while a false flag indicates that at least one verification has failed.

[0089] In another alternative approach, the output current, output power, and voltage of each inverter can be defined as optimization variables, and the constraints can be considered as the feasible region of the optimization variables. A set of initial feasible points is selected, where the output current of each inverter is less than or equal to its maximum output current, the output power of each inverter is less than or equal to its maximum output power, and the voltage of each inverter is within a preset voltage operating range, satisfying the global circuit model. The global circuit model is then incorporated into the objective function in the form of Lagrange multipliers, and the maximum output power, maximum output power, and maximum output power of each inverter are considered as the feasible region of the optimization variables. The large output current and voltage constraints are incorporated into the objective function in the form of a logarithmic barrier function, forming an augmented objective function. Starting from the initial feasible point, the augmented objective function is iteratively searched along the Newton direction. After each iteration, the barrier parameters are updated, and it is checked whether the current optimization variable is within the feasible region. If it is outside the region, the iteration is backtracked to the boundary of the feasible region. When the set iteration termination condition is met, the iteration stops, and the output current of each inverter in the current optimization variable is taken as the optimal output current of each inverter. The optimal total grid-connected power is determined based on the output power of each inverter in the current optimization variable. This optional method uses the interior-point method to solve for the specific process of obtaining the optimal total grid-connected power and the optimal output current of each inverter. This makes the solution of the nonlinear programming model no longer a black box operation, realizes the feasibility of the algorithm solution process, and improves the accuracy of the solution process.

[0090] The above-mentioned approach incorporates the global circuit model into the objective function in the form of Lagrange multipliers, and incorporates the maximum output power, maximum output current, and voltage constraints of each inverter into the objective function in the form of logarithmic barrier functions, forming an augmented objective function. This includes creating a computational function for the augmented objective function. This function receives the value of the current optimization variable as input, calculates, and returns the value of the augmented objective function. Internally, this function first calculates the value of the original objective function, which is the negative of the sum of the products of all inverter terminal voltages and output currents. Since the optimization problem is to maximize the total power at the grid connection point, while standard optimization algorithms typically handle minimization problems, taking the negative transforms the maximization problem into a minimization problem. Then, it iterates through all equality constraints. For the node voltage matrix equation of the global circuit model, it calculates the difference between the left and right sides of the equation, multiplies this difference by the corresponding Lagrange multiplier, and adds it to the augmented objective function. Next, all inequality constraints are iterated. For the maximum output current constraint of each inverter, the difference between the current output current and the maximum output current is calculated. If the difference is positive, it indicates that the limit has been exceeded, and a large positive number is returned as a penalty. If the difference is negative, the negative of the natural logarithm of the difference is calculated, multiplied by the current barrier parameter, and added to the augmented objective function. The same processing method is used for the maximum output power constraint and voltage constraint. Finally, the original objective function value, the values ​​of all Lagrange multipliers, and the values ​​of all logarithmic barrier function terms are added to obtain the final value of the augmented objective function, which is then returned.

[0091] The aforementioned barrier parameters are used in the interior-point method to control the strength of the logarithmic barrier function. In the early stages of iteration, the barrier parameters are set to large values, resulting in a significant penalty effect from the logarithmic barrier function within the feasible region, thus pushing the optimization variables towards the center of the feasible region. As iteration progresses, the barrier parameters gradually decrease, the penalty effect of the logarithmic barrier function gradually weakens, and the optimization variables can gradually approach the constraint boundary, eventually converging to the optimal solution of the original problem. The barrier parameters are typically updated by multiplying them by a decay factor less than one.

[0092] The above-described process, starting from an initial feasible point and iteratively searching the augmented objective function along the Newton direction, updates the obstacle parameters after each iteration and checks whether the current optimization variable is within the feasible region. If it is outside, it backtracks to the feasible region boundary. This includes: at the beginning of the iteration loop, setting the value of the current optimization variable to the value of the initial feasible point. Then, entering the iteration loop, the following operations are performed in each iteration: First, calculate the gradient vector and Hessian matrix of the augmented objective function at the current point. Each element of the gradient vector is the partial derivative of the augmented objective function with respect to the corresponding optimization variable, and each element of the Hessian matrix is ​​the second-order partial derivative of the augmented objective function with respect to the two optimization variables. Second, solve the system of linear equations. Multiply the Hessian matrix by the negative gradient vector to obtain the Newton direction vector. Third, perform a one-dimensional search to determine the step size. Starting from the initial step size, move along the Newton direction, calculate the value of the optimization variable after the move, and check whether all constraints are satisfied. If satisfied, accept the step size; if not, shorten the step size and recheck until a step size that satisfies all constraints is found. Fourth, update the values ​​of the optimization variables according to the determined step size. Fifth, multiply the obstacle parameters by the decay factor to obtain the new obstacle parameters. Sixth, check whether the updated optimization variables are within the feasible region. If an optimization variable exceeds the upper limit value corresponding to the constraint, set the value of that optimization variable to the upper limit value corresponding to the constraint. Seventh, calculate the updated augmented objective function value and compare it with the objective function value of the previous iteration to determine whether the termination condition is met. If not, proceed to the next iteration.

[0093] The aforementioned iteration termination conditions are a set of conditions used to determine when the iterative algorithm stops. In this embodiment, the iteration termination conditions may include three: the first condition is that the difference between the augmented objective function values ​​of two adjacent iterations is less than a preset difference, indicating that the objective function has no longer improved significantly; the second condition is that the change in the optimization variable is less than a preset change, indicating that the solution has tended to stabilize; the third condition is that the number of iterations has reached a preset maximum number of iterations, which is to prevent the algorithm from falling into an infinite loop. The iteration will stop when any one of the three conditions is met.

[0094] In this embodiment, a framework for optimization is defined: maximizing the total power at the grid connection point is the objective; the global circuit model serves as equality constraints; the maximum output power and current of the inverters are used as inequality constraints; and the voltage operating range is used as voltage constraints. This framework is used to construct and solve a nonlinear programming model. The multi-inverter collaborative optimization problem is transformed into a standard nonlinear programming problem. By clearly defining the objective function and various constraints, it is ensured that the optimization results maximize power generation efficiency while also meeting equipment safety and grid compliance requirements.

[0095] Based on the first embodiment of this application, in the second embodiment of this application, the same or similar content as the above embodiment can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 6 Following step S40, the global optimization method for grid-connected power also includes steps S50 to S70: Step S50: Calculate the output power of all inverters in pure maximum power point tracking mode, and sum them up to obtain the total grid-connected power in pure maximum power point tracking mode.

[0096] Pure MAX Point Tracking (MPT) mode refers to a working mode in which the inverter does not receive any external power commands and operates independently entirely based on its built-in MPT algorithm. In this mode, each inverter autonomously searches for and tracks the optimal operating point of the photovoltaic modules based on the current light intensity and ambient temperature, maximizing its output power under the current environmental conditions. Pure MPT mode is the default operating mode for photovoltaic inverters and is the control mode commonly used in traditional photovoltaic power plants. The output power in pure MPT mode refers to the active power value calculated and output in real time by the inverter's own MPT algorithm when operating in this mode. This power value represents the maximum power that the inverter can generate under the current light and temperature conditions, but it is derived under the assumption that other inverters have no impact on its terminal voltage. The total grid-connected power in pure MPT mode is the sum of the output power of all inverters in pure MPT mode. This total grid-connected power is a theoretical reference value. It represents the total power that the grid-connected point can achieve if all inverters can operate at full capacity in maximum power point tracking mode without voltage overruns. However, in actual operation, due to voltage coupling effects, this total grid-connected power often cannot be fully realized, as some inverters may be forced to reduce power or disconnect from the grid due to voltage overruns.

[0097] Step S60: If the total grid-connected power in pure maximum power point tracking mode is greater than or equal to the optimal grid-connected power, and the current grid voltage and power fluctuations meet the preset compliance requirements, then each inverter will continue to operate in pure maximum power point tracking mode.

[0098] Pre-defined compliance requirements refer to the technical indicator limits stipulated by power grid companies and power dispatching agencies for the grid-connected operation of photovoltaic power plants. These requirements mainly include two aspects: First, voltage compliance requirements, which require the grid connection point voltage and the voltage at each inverter terminal to be within the allowable range specified by national standards. Second, power fluctuation compliance requirements, which require the rate of change of active power at the grid connection point to not exceed the prescribed limits.

[0099] Maintaining each inverter in pure maximum power point tracking (MPPT) mode means that the controller does not issue any power or current commands to the inverters, allowing them to continue their default autonomous operation. In this state, each inverter independently runs its own MPPT algorithm, autonomously searching for the optimal operating point, unaffected by external commands. Maintaining pure MPPT mode means that global optimization is not required under current operating conditions, as pure MPPT mode already achieves sufficient power generation and meets compliance requirements.

[0100] Step S70: If the total grid connection power in pure maximum power point tracking mode is less than the optimal grid connection power, or if the current grid voltage or power fluctuation does not meet the preset compliance requirements, then the optimal output current of each inverter will be sent to the corresponding inverter for execution.

[0101] Sending the optimal output current of each inverter to its corresponding inverter for execution means that the controller sends the calculated optimal output current value of each inverter to the corresponding inverter via the communication interface. Upon receiving the optimal output current, the inverter uses it as the setpoint for its output current and adjusts the on-time of its power switches to ensure the actual output current tracks this setpoint. While executing the optimal output current, the inverter's internal maximum power point tracking algorithm continues to run in the background, continuously calculating the maximum output current under the current environmental conditions as an upper limit protection for the output current.

[0102] In one alternative approach, when the optimal output current of each inverter is sent to the corresponding inverter for execution, the method further includes: for each inverter, comparing the optimal output current of the inverter with the maximum output current determined by the maximum power point tracking algorithm of the inverter; if the optimal output current is less than or equal to the maximum output current, controlling the inverter to normally execute the corresponding optimal output current, and controlling the maximum power point tracking loop of the inverter to continue working, so as to regulate the output current of the inverter based on the maximum output current of the maximum power point tracking loop; if the optimal output current is greater than the maximum output current, switching the output current setting source of the inverter to the maximum power point tracking loop to determine the maximum output current of the inverter through the maximum power point tracking algorithm, and controlling the inverter to execute the maximum output current. After receiving the optimal output current, the inverter needs to compare the optimal output current with the maximum output current determined by the local maximum power point tracking algorithm. Based on the comparison result, it decides whether to execute the optimal output current or fall back to the maximum power point tracking mode. It establishes a dual safety guarantee mechanism, which ensures both the effect of global optimization and the safe operation of the equipment.

[0103] In this embodiment, a complete logic for secondary judgment of grid connection point power is added. That is, after obtaining the optimal solution, it is not executed directly. Instead, it is first compared with the power generation level in pure maximum power point tracking mode, and combined with compliance checks on voltage and power fluctuations, it decides whether to execute the optimal output current command obtained through global optimization. This avoids blind optimization and ensures that intervention is only performed when it truly brings benefits or is absolutely necessary. This on-demand optimization strategy ensures both improved power generation efficiency and reduced unnecessary system overhead.

[0104] Based on the above embodiments of this application, in the third embodiment of this application, the same or similar content as the above embodiments can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 7 Following step S40, the global optimization method for grid-connected power further includes steps S80~S110: Step S80: Send the optimal output current of each inverter to the corresponding inverter for execution.

[0105] Step S90: Monitor the operating status of each inverter.

[0106] Monitoring the operating status of each inverter refers to the controller periodically collecting operating data from each inverter through the communication interface, including output current, output voltage, output power, operating temperature, fault codes, and other information. This data is then analyzed and processed to determine whether the inverter is operating normally. The purpose of monitoring is to promptly detect abnormal conditions of the inverter, especially grid disconnection events, so as to respond in a timely manner.

[0107] Grid disconnection due to fault refers to the state in which an inverter stops transmitting power to the grid due to its own fault or external causes. Causes of grid disconnection may include damage to internal power devices, DC-side insulation faults, AC-side overvoltage or undervoltage protection issues, and abnormal grid frequency. When an inverter experiences grid disconnection due to fault, its output current becomes zero, and it no longer injects power into the grid. Grid disconnection affects the total power generation of a photovoltaic power plant and also alters the current distribution in the circuit topology, thereby affecting the terminal voltage of other normally operating inverters.

[0108] Step S100: When a fault is detected in any inverter and the grid disconnects, update the global circuit model and the boundaries of other constraints.

[0109] Detecting grid disconnection due to a fault in any inverter means that the controller, through continuous monitoring of the operating status, discovers that an inverter has stopped transmitting power to the grid due to its own fault or external causes. The criteria for determining grid disconnection are usually that the inverter's output current is zero for several consecutive monitoring cycles, and its operating status indicator shows a fault state or offline state.

[0110] Updating the global circuit model refers to modifying the global circuit model according to the actual circuit topology changes after a fault disconnection. When an inverter fails and disconnects from the grid, it no longer injects current into the grid, which is equivalent to the current source branch corresponding to that inverter being disconnected from the circuit. Therefore, it is necessary to remove the branch corresponding to that inverter from the global circuit model, including deleting the column corresponding to the branch from the correlation matrix, the row and column corresponding to the branch from the admittance matrix, the element corresponding to the branch from the current source matrix, and the element corresponding to the branch from the voltage source matrix. Simultaneously, it is also necessary to update the coefficients in the node voltage matrix equations to reflect the new circuit topology.

[0111] The boundaries of other constraints refer to the value ranges of various constraints other than those in the global circuit model. These other constraints mainly include the maximum output power constraint, the maximum output current constraint, and the preset voltage operating range constraint for each inverter. When an inverter fails and disconnects from the grid, it no longer participates in power generation; therefore, its corresponding maximum output power and maximum output current constraints are no longer applicable and need to be removed from the optimization model. Simultaneously, since the current and voltage distributions of other normally operating inverters will change after the faulty inverter disconnects from the grid, their voltage operating range constraints may need to be appropriately adjusted according to the new operating conditions to ensure that the terminal voltages of all inverters remain within a safe range under the new topology.

[0112] Step S110: Based on the updated global circuit model and other constraints, the new optimal total power at the grid connection point and the optimal output current are solved again, and the optimal output current is sent to the corresponding inverter in normal operation.

[0113] The new optimal total power and optimal output current at the grid connection point can be obtained by resolving using the interior point method. The solution process is the same as described in the above embodiment and will not be repeated here.

[0114] In this embodiment, when the inverter fails and disconnects from the grid, it has autonomous fault tolerance and model reconstruction capabilities. When the inverter fails and disconnects from the grid, it can automatically update the global circuit model and constraints, dynamically complete the optimal allocation of power to the remaining equipment, without manual intervention, adapt to the trend of less manned operation and maintenance of power plants, and effectively reduce operation and maintenance costs.

[0115] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the global optimization method of grid-connected power of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0116] Based on the same inventive concept, this application provides a global optimization device for grid-connected power, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to execute the global optimization method for grid-connected power in the above embodiments.

[0117] The grid-connected power global optimization device provided in this application, employing the grid-connected power global optimization method in the above embodiments, can achieve global optimization of grid-connected power in multi-inverter scenarios, thereby improving the overall power generation efficiency of photovoltaic power plants. Compared with the prior art, the beneficial effects of the grid-connected power global optimization device provided in this application are the same as those of the grid-connected power global optimization method provided in the above embodiments, and other technical features in the grid-connected power global optimization device are the same as those disclosed in the method of the previous embodiment, and will not be repeated here.

[0118] Based on the same inventive concept, this application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, which are used to execute the global optimization method for grid-connected power in the above embodiments.

[0119] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory (EPROM), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, radio frequency (RF), etc., or any suitable combination thereof.

[0120] The aforementioned computer-readable storage medium may be included in a global optimization device for grid-connected power; or it may exist independently and not be assembled into a global optimization device for grid-connected power.

[0121] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the grid-connected power global optimization device, cause the grid-connected power global optimization device to implement the steps of the aforementioned method.

[0122] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0123] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0124] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0125] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described global optimization method for grid-connected power. This enables global optimization of grid-connected power in multi-inverter scenarios, thereby improving the overall power generation efficiency of photovoltaic power plants. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the global optimization method for grid-connected power provided in the above embodiments, and will not be elaborated upon here.

[0126] The above are only some embodiments of this application and do not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A global optimization method for grid-connected power, characterized in that, The method includes: Obtain the circuit topology parameters of the power station in the distribution area and the electrical parameters required for the operation of the power station in the distribution area; Based on the circuit topology parameters and the electrical parameters, an actual circuit topology is constructed, wherein the actual circuit topology reflects the connection relationships of each cable line, each impedance, and each inverter. Based on the actual circuit topology, a global circuit model is derived, wherein the global circuit model is used to describe the electrical coupling relationship between each inverter in the power distribution area. The solution is performed with the goal of maximizing the total power at the grid connection point, and the global circuit model is used as a constraint during the solution process to obtain the optimal total power at the grid connection point and the optimal output current of each inverter.

2. The global optimization method for grid-connected power as described in claim 1, characterized in that, The global circuit model derived based on the actual circuit topology includes: The actual circuit topology is converted into a directed graph topology, wherein the directed graph topology uses nodes to represent connection points with equal potential in the actual circuit topology, uses directional edges to represent the current reference direction of each cable branch in the actual circuit topology, and uses edge weights to represent the impedance value of each branch. Construct the correlation matrix, admittance matrix, voltage source matrix, current source matrix, current matrix, and voltage matrix based on the directed graph topology; Based on the correlation matrix, the admittance matrix, the voltage source matrix, the current source matrix, the current matrix, and the voltage matrix, the node voltage matrix equation is derived using Kirchhoff's laws and serves as the global circuit model.

3. The global optimization method for grid-connected power as described in claim 2, characterized in that, The process of converting the actual circuit topology into a directed graph topology includes: Each cable branch in the actual circuit topology is converted into a corresponding standard composite branch based on the standard composite branch model. The standard composite branch model includes a voltage source and an impedance connected in series, and a current source connected in parallel with the voltage source. Construct the directed graph topology based on each of the standard composite branches; The nodes of the directed graph topology include the grid connection point on the transformer substation side, the connection point between the main cable and the branch cable, the segment connection point on the branch cable, and the AC port access point of each inverter. The cable lines of the actual circuit topology include: main cable segment branches, branch cable segment branches, and inverter segment branches.

4. The global optimization method for grid-connected power as described in claim 2, characterized in that, The construction of the correlation matrix, admittance matrix, voltage source matrix, current source matrix, current matrix, and voltage matrix based on the directed graph topology includes: Based on the connection relationships between nodes and branches in the directed graph topology and the current reference direction of the branches, an association matrix is ​​constructed. Based on the impedance parameters of each branch in the directed graph topology, an admittance matrix is ​​constructed, wherein the voltage source matrix is ​​a column vector, and the elements in the column vector are the voltage values ​​of the voltage sources in the corresponding branches. Construct a current source matrix based on the current values ​​of the current sources contained in each branch of the directed graph topology; Construct a current matrix based on the current values ​​of each branch in the directed graph topology; A voltage matrix is ​​constructed based on the port voltage values ​​of each branch in the directed graph topology.

5. The global optimization method for grid-connected power as described in claim 1, characterized in that, The process of solving for the optimal total power at the grid connection point, with constraints imposed by the global circuit model, to obtain the optimal total power at the grid connection point and the optimal output current of each inverter includes: With the goal of maximizing the total power at the grid connection point, the global circuit model is used as the equality constraint, the maximum output power and maximum output current of each inverter are used as the inequality constraint, and the preset voltage operating range of each inverter is used as the voltage constraint, a nonlinear programming model is constructed. The nonlinear programming model is iteratively solved to obtain the optimal output current of each inverter and the optimal total power at the grid connection point.

6. The global optimization method for grid-connected power as described in claim 5, characterized in that, The iterative solution of the nonlinear programming model to obtain the optimal output current of each inverter and the optimal total power at the grid connection point includes: The nonlinear programming model is solved, and the solution results are verified for business feasibility. The business feasibility verification is used to ensure that the solution results comply with preset power grid dispatch rules and preset field operation and maintenance rules. If the solution result passes the business feasibility verification, then based on the solution result, the optimal total power of the grid connection point and the optimal output current of each inverter are obtained; If the solution result fails the business feasibility verification, the nonlinear programming model is iteratively solved again until the solution result passes the business feasibility verification. Based on the final solution result that passes the verification, the optimal total power of the grid connection point and the optimal output current of each inverter are obtained.

7. The global optimization method for grid-connected power as described in claim 5, characterized in that, The iterative solution of the nonlinear programming model to obtain the optimal output current of each inverter and the optimal total power at the grid connection point includes: The output current, output power, and voltage of each inverter are determined as optimization variables, and the constraints are taken as the feasible region of the optimization variables. Select a set of initial feasible points, wherein the output current of each inverter in the initial feasible points is less than or equal to the maximum output current of the corresponding inverter, the output power of each inverter is less than or equal to the maximum output power of the corresponding inverter, the voltage of each inverter is within the preset voltage operating range, and satisfies the global circuit model. The global circuit model is incorporated into the objective function in the form of Lagrange multipliers, and the maximum output power, maximum output current and voltage constraints of each inverter are incorporated into the objective function in the form of logarithmic barrier functions to form an augmented objective function. Starting from the initial feasible point, the augmented objective function is iteratively searched along the Newton direction. After each iteration, the obstacle parameters are updated, and it is checked whether the current optimization variable is within the feasible region. If it is outside the feasible region, the search returns to the boundary of the feasible region. When the set iteration termination condition is met, the iteration stops, and the output current of each inverter in the current optimization variables is taken as the optimal output current of each inverter. The optimal total power of the grid connection point is determined based on the output power of each inverter in the current optimization variables.

8. The global optimization method for grid-connected power as described in any one of claims 1 to 7, characterized in that, After obtaining the optimal total power at the grid connection point and the optimal output current of each inverter, the following is also included: The output power of all inverters in pure maximum power point tracking mode is calculated and summed to obtain the total grid-connected power in pure maximum power point tracking mode. If the total grid-connected power in the pure maximum power point tracking mode is greater than or equal to the optimal grid-connected power, and the current grid voltage and power fluctuations meet the preset compliance requirements, then each inverter will continue to operate in the pure maximum power point tracking mode. If the total grid-connected power in the pure maximum power point tracking mode is less than the optimal grid-connected power, or if the current grid voltage or power fluctuation does not meet the preset compliance requirements, then the optimal output current of each inverter will be sent to the corresponding inverter for execution.

9. The global optimization method for grid-connected power as described in claim 8, characterized in that, When distributing the optimal output current of each inverter to the corresponding inverter for execution, the method further includes: For each inverter, the inverter's optimal output current is compared with the maximum output current currently determined by the inverter's maximum power point tracking algorithm. If the optimal output current is less than or equal to the maximum output current, then the inverter is controlled to normally execute the corresponding optimal output current. If the optimal output current is greater than the maximum output current, the output current setting source of the inverter is switched to maximum power point tracking loop operation, so as to determine the maximum output current of the inverter through the maximum power point tracking algorithm, and control the inverter to execute the maximum output current.

10. The global optimization method for grid-connected power as described in claim 5, characterized in that, After iteratively solving the nonlinear programming model to obtain the optimal output current of each inverter and the optimal total power at the grid connection point, the method further includes: The optimal output current of each inverter is sent to the corresponding inverter for execution. Monitor the operating status of each inverter; When a fault is detected in any inverter causing it to disconnect from the grid, the global circuit model and the boundaries of other constraints are updated. Based on the updated global circuit model and other constraints, the new optimal total power at the grid connection point and the optimal output current are obtained by resolving the problem, and the optimal output current is then sent to the corresponding inverter in normal operating condition.

11. A global optimization device for grid-connected power, characterized in that, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the global optimization method for grid-connected power as described in any one of claims 1 to 10.

12. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the global optimization method for grid-connected power as described in any one of claims 1 to 10.

Images