Photovoltaic energy storage system coordination control method

Abstract

The invention relates to the technical field of power grid data processing, in particular to a photovoltaic energy storage system coordination control method, which comprises the following steps of: acquiring power grid topology and line parameters, establishing a power grid admittance matrix, calculating voltage variation caused by unit reactive power injection between any two photovoltaic energy storage system nodes, and calculating the power grid admittance matrix. And grouping all photovoltaic energy storage system nodes to obtain a coordinated voltage regulation partition set. According to the method, the power grid admittance matrix is established by obtaining the power grid topology and the line parameters, and the voltage variation generated by unit reactive power injection between any two photovoltaic energy storage system nodes is calculated, so that the node voltage sensitivity is obtained, and the electrical coupling relation between the nodes is determined; therefore, the nodes are accurately grouped and a cooperative voltage regulation partition set is established, so that cooperative control among the nodes of the photovoltaic energy storage system is clearer.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid data processing, and in particular to a coordinated control method for a photovoltaic energy storage system. Background Art

[0002] The coordinated control of photovoltaic energy storage systems is mainly used to coordinate the real-time output of multiple photovoltaic energy storage systems and the charge status of energy storage units when power congestion or voltage fluctuations occur in the power grid.

[0003] Existing technologies directly coordinate the real-time output of each PV energy storage system and the state of charge of the energy storage unit, without establishing a voltage sensitivity model. This results in the actual electrical influence of the node being ignored, and the coordination ability between nodes is assessed only at a macro level, failing to quantify node responsibility. Furthermore, when power flow management tasks occur, only the overall available power of the energy storage equipment is used for rough scheduling, resulting in low scheduling accuracy and failure to fully tap the potential of power flow management, making it difficult to effectively resolve local line congestion in the power grid. Therefore, improvements are needed. Summary of the Invention

[0004] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a coordinated control method for a photovoltaic energy storage system.

[0005] To achieve the above objectives, the present invention adopts the following technical solution, a photovoltaic energy storage system coordinated control method, comprising the following steps: Obtain the grid topology and line parameters, establish the grid admittance matrix, calculate the voltage change caused by unit reactive power injection between any two PV energy storage system nodes, group all PV energy storage system nodes, and obtain a set of coordinated voltage regulation partitions; Based on the coordinated voltage regulation zone set, the real-time photovoltaic output fluctuation data and the energy storage unit charge state of each photovoltaic energy storage system in each zone are read to obtain the aggregate adjustable power range within the zone, and then the zone coordinated response weight set is established based on the aggregate adjustable power range within the zone; Obtain the power transfer distribution factor matrix and congested line identifiers from the grid dispatch center, extract the power transfer distribution factor values ​​of each PV energy storage system's partitioned grid connection point relative to the congested line, and obtain a set of power transfer distribution factors. This set of power transfer distribution factors is then calculated with the partitioned adjustable power to determine the maximum collaborative power flow diversion potential. Based on the maximum collaborative power flow diversion potential and the partition collaborative response weight set, an energy storage unit power instruction sequence is generated.

[0006] Preferably, the steps of obtaining the coordinated voltage regulation partition set are: Obtaining the topological connection relationship of all nodes in the power grid and the electrical parameters of the corresponding lines, calculating the self-admittance and mutual-admittance of the lines one by one based on the topological connection relationship and the electrical parameters of the lines, and constructing a power grid admittance matrix based on the self-admittance and mutual-admittance of the lines to generate a power grid admittance matrix; Based on the grid admittance matrix, inject unit reactive power into each photovoltaic energy storage system node in turn, calculate and record the voltage change amplitude of all other nodes in the grid when the node injects unit reactive power, and obtain the voltage change value between each node; Based on the voltage variation values ​​between the nodes, the voltage variation between the nodes of the photovoltaic energy storage system is counted node by node, a screening threshold for the node voltage variation is set, and the nodes whose voltage variation reaches or exceeds the screening threshold are classified into the same group to form a coordinated voltage regulation partition set.

[0007] Preferably, the step of obtaining the aggregated adjustable power range within the partition is: Based on the coordinated voltage regulation zone set, each photovoltaic energy storage system in each zone is visited one by one, and the current photovoltaic output fluctuation data and the state of charge value of the corresponding energy storage unit of each photovoltaic energy storage system are collected in real time to form a set of real-time photovoltaic output fluctuation data and energy storage unit state of charge value; Based on the real-time photovoltaic output fluctuation data and the energy storage unit state of charge value set, the maximum and minimum active power values, as well as the maximum and minimum reactive power values, that can be adjusted for each photovoltaic energy storage system in its current state are calculated one by one to form the adjustable active and reactive power limits of each photovoltaic energy storage system; Based on the adjustable active and reactive power limits of each photovoltaic energy storage system, the sum of the adjustable active power limits and the sum of the adjustable reactive power limits of all photovoltaic energy storage systems in each partition are aggregated and calculated to form an aggregated adjustable power range within the partition.

[0008] Preferably, the steps of obtaining the partition coordinated response weight set are: Based on the aggregated adjustable power range within the partition, the maximum active power, minimum active power, maximum reactive power, and minimum reactive power of each photovoltaic energy storage system in the partition are sequentially read, and the voltage change amplitude generated by the node when injecting unit reactive power to all other nodes in the partition is simultaneously retrieved, and the power regulation boundary and node voltage sensitivity value set of the photovoltaic energy storage system in the partition are integrated; Calculating a coordinated response weight of the photovoltaic energy storage system based on a power regulation boundary and a node voltage sensitivity value set of the photovoltaic energy storage system in the partition; Based on the coordinated response weights of the photovoltaic energy storage systems, the coordinated response weight results of all photovoltaic energy storage systems in the partition are summarized and sorted item by item according to the partition to which they belong, forming a partition coordinated response weight set.

[0009] Preferably, the steps of obtaining the power transmission distribution factor set are: The complete power transfer distribution factor matrix and the identifier of the target congested line in the control state are retrieved from the power grid dispatching center. The column vectors of the grid connection points of each photovoltaic energy storage system partition relative to the target congested line are screened to generate a set of power transfer distribution factors.

[0010] Preferably, the steps for obtaining the maximum collaborative flow diversion potential are: Based on the power transfer distribution factor set, read the current active power value, maximum adjustable active power value, and minimum adjustable active power value of all members in each photovoltaic energy storage system partition, and sum the upward active power difference and downward active power difference within the same partition to form a partition directional adjustable active power pair; Based on the partitioned directional adjustable active power pairs, the maximum collaborative power flow diversion potential of the target line is calculated.

[0011] Preferably, the steps of acquiring the energy storage unit power instruction sequence are: The total amount of power flow grooming tasks currently required to be completed on the target line is obtained from the power grid dispatching center. The maximum coordinated power flow grooming potential is called, and the contribution value provided by each photovoltaic energy storage system partition in the calculation of the maximum coordinated power flow grooming potential is extracted one by one. The ratio of the partition's contribution value to the maximum coordinated power flow grooming potential is used as the partition's contribution ratio. The power grooming instructions allocated to each photovoltaic energy storage system partition are determined based on the contribution ratio, and the power grooming instructions for the partition are generated. Based on the partitioned power grooming instructions, the partitioned coordinated response weight set is called and the coordinated response weights of all photovoltaic energy storage systems in each partition are extracted one by one. The coordinated response weight of each photovoltaic energy storage system is divided by the sum of the coordinated response weights of all photovoltaic energy storage systems in the partition in turn to complete the normalization of the weights within the partition. The partitioned power grooming instructions are then subdivided into each photovoltaic energy storage system in the partition according to the normalized weight values ​​to generate an initial power instruction for each photovoltaic energy storage system.

[0012] Preferably, the step of acquiring the energy storage unit power instruction sequence further includes: based on the initial power instruction of each photovoltaic energy storage system, retrieving the real-time state of charge of the current energy storage unit of the corresponding photovoltaic energy storage system and the real-time adjustable active power range of the inverter one by one, comparing whether the initial power instruction falls within the real-time adjustable active power range of the inverter; if the initial power instruction exceeds the range, using the boundary value of the real-time adjustable active power range of the inverter as the corrected power instruction of the photovoltaic energy storage system; after all corrections are completed, the determined corrected power instruction is sent to the inverter of the corresponding photovoltaic energy storage system to generate the energy storage unit power instruction sequence.

[0013] Compared with the prior art, the advantages and positive effects of the present invention are: The present invention establishes a grid admittance matrix by acquiring grid topology and line parameters, and simultaneously calculates the voltage change caused by unit reactive power injection between any two photovoltaic energy storage system nodes, thereby obtaining node voltage sensitivity and clarifying the electrical coupling relationship between the nodes. This allows for precise grouping of nodes and establishment of a collaborative voltage regulation partition set, making the collaborative control between photovoltaic energy storage system nodes more explicit. Furthermore, the present invention combines real-time photovoltaic output fluctuation data within the partition with the charge state of the energy storage unit to determine the aggregate adjustable power range of each partition, and constructs a partition collaborative response weight set based on this, achieving quantification of the collaborative response capability of each node and eliminating the drawback of ambiguous regulation responsibility of nodes within the partition. The present invention also collaboratively calculates the power transmission distribution factor and the partition adjustable power, analyzes the actual congestion relief potential of each partition, and accurately allocates energy storage unit power instructions based on the maximum collaborative flow relief potential and the partition collaborative response weight, optimizing the power scheduling strategy and improving the grid flow relief efficiency. This achieves fine-grained control of photovoltaic energy storage system scheduling decisions between and within partitions, improving grid operation stability and renewable energy utilization efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 Schematic diagram of the steps of the present invention. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0016] See also Figure 1 The present invention provides a technical solution, a photovoltaic energy storage system coordinated control method, comprising the following steps: Obtain the grid topology and line parameters, establish the grid admittance matrix, calculate the voltage change caused by unit reactive power injection between any two PV energy storage system nodes, group all PV energy storage system nodes, and obtain a set of coordinated voltage regulation partitions; Based on the coordinated voltage regulation zone set, the real-time photovoltaic output fluctuation data and energy storage unit charge state of each photovoltaic energy storage system in each zone are read to obtain the aggregated adjustable power range within the zone. Then, based on the aggregated adjustable power range within the zone, the zone coordinated response weight set is established; The power transfer distribution factor matrix and congested line identifiers are obtained from the grid dispatch center. The power transfer distribution factor values ​​of each PV energy storage system's partitioned grid connection point relative to the congested line are extracted to obtain a set of power transfer distribution factors. This set of power transfer distribution factors is then calculated with the partitioned adjustable power to determine the maximum collaborative power flow diversion potential. Based on the maximum collaborative power flow diversion potential and the partition collaborative response weight set, the energy storage unit power command sequence is generated.

[0017] The steps to obtain the coordinated voltage regulation partition set are: Obtain the topological connection relationship of all nodes in the power grid and the electrical parameters of the corresponding lines. Based on the topological connection relationship and the electrical parameters of the lines, calculate the self-admittance and mutual-admittance of the lines one by one, and construct the power grid admittance matrix based on the line self-admittance and mutual-admittance to generate the power grid admittance matrix; Based on the grid admittance matrix, unit reactive power is injected into each PV energy storage system node in turn. The voltage change amplitude of all other nodes in the grid when the node injects unit reactive power is calculated and recorded point by point, and the voltage change value between each node is obtained. Based on the voltage variation values ​​between nodes, the voltage variation between nodes of each photovoltaic energy storage system is counted node by node, and a screening threshold for the node voltage variation is set. Nodes whose voltage variation reaches or exceeds the screening threshold are classified into the same group to form a coordinated voltage regulation partition set.

[0018] Specifically, based on the obtained topological connection relationship of all nodes in the power grid and the electrical parameters of the corresponding lines, these basic data are first parsed to clarify the unique identifier of each node (such as power plants, substations, and load centers), as well as the start and end nodes connected to each transmission line. At the same time, the key electrical parameters of the line are extracted, including the resistance value, reactance value, and earth susceptance value per unit length of the line. These parameters are multiplied by the line length to obtain the total resistance of each line. , total reactance and total susceptance Then, we traverse each transmission line in the grid topology and calculate the series impedance of the line based on its total resistance and total reactance. , and find its reciprocal to obtain the series admittance of the line , for the earth susceptance of each line, it is equivalent to two parallel branches at both ends of the line in the π-type equivalent circuit, and each end is allocated half of the total line susceptance value, that is, After completing the calculation of all line parameters, we start to build a dimension of The grid admittance matrix is is the total number of nodes in the power grid, and each element of the matrix is ​​filled by the following rules: for the non-diagonal elements of the matrix ( ), whose value is the connection node With node The opposite of the sum of the admittances of all branches, that is, , if the node With node If there is no direct line connection between is 0, for the diagonal elements of the matrix , whose value is connected to the node The sum of the admittance of all branches and the parallel admittance of the node to the ground is calculated as follows: in, For nodes With node The branch admittance between Represents grounding. By repeating this calculation process for all nodes, the entire matrix is ​​filled and the grid admittance matrix is ​​finally obtained.

[0019] Based on the grid admittance matrix generated in the previous process, in order to determine the electrical coupling strength between the nodes of each photovoltaic energy storage system, it is necessary to calculate the voltage sensitivity. First, the grid admittance matrix is ​​inverted to obtain the grid impedance matrix corresponding to it. , each element of the impedance matrix The physical meaning is the current node When unit current is injected at node Then, all nodes in the grid that have photovoltaic energy storage systems deployed are identified and a node list is created. Then, the simulation calculation of unit reactive power injection is performed on each photovoltaic energy storage system node in the list. The specific operation is as follows: when it is the turn of the first When analyzing a photovoltaic energy storage system node, a reactive power increment with an amplitude of 1 per unit is set to be injected into the node, that is, In the power system flow calculation, the complex power injected by the node is and node voltage and injected current The relationship is , from which the change in node injection current can be derived and reactive power variation In the common case where the system voltage amplitude is close to 1 per unit, the relationship can be approximated as , so, to the node Injecting unit reactive power is approximately equivalent to injecting a current of 1 per unit and a phase of -90 degrees into the node. Then, the grid impedance matrix is ​​used to calculate the voltage change caused by this injection on all other nodes in the grid. For any node , the voltage change It can be calculated by the corresponding elements of the impedance matrix: Since the injected reactive power is unit, the calculated voltage change amplitude is That is the node For Node Voltage sensitivity coefficient ,Right now Repeat the above injection and calculation process for each photovoltaic energy storage system node in the list, and record the voltage change amplitude of all other nodes in the grid (especially other photovoltaic energy storage system nodes) during each injection, and compile all these calculation results into a The matrix of is the number of photovoltaic energy storage system nodes, is the total number of nodes in the power grid, and this matrix is ​​the voltage change value between each node.

[0020] Based on the voltage change values ​​between nodes obtained in the previous steps, i.e., the voltage sensitivity matrix, the photovoltaic energy storage system needs to be grouped to form partitions. First, from the complete voltage sensitivity matrix, a sub-matrix involving only the mutual influence between the photovoltaic energy storage system nodes is extracted. The dimension of this sub-matrix is , where the elements Indicates in The unit reactive power injected by each photovoltaic energy storage system node is The voltage variation amplitude caused by each photovoltaic energy storage system node is then divided into areas with close electrical coupling. A clear screening threshold for node voltage variation needs to be set. The setting of this threshold adopts statistical methods to ensure its rationality. Specifically, the voltage variation values ​​between all different photovoltaic energy storage system node pairs (i.e. All value), forming a numerical set, and calculating the arithmetic mean and standard deviation of all the values ​​in the set. For example, if the calculated mean is 0.045 and the standard deviation is 0.02, the screening threshold can be set to the mean plus a preset multiple of the standard deviation, such as "mean + 1.5 times the standard deviation", that is, , the threshold is used as the basis for judging whether two nodes belong to the same strong coupling region. Subsequently, the connected component search algorithm based on graph theory is used to group them, and each photovoltaic energy storage system node is regarded as a vertex of a graph. All photovoltaic energy storage system node pairs are traversed. , if the mutual voltage change between them or If at least one of them reaches or exceeds the set screening threshold of 0.075, an edge is added between the vertices representing the two nodes, indicating that there is a strong electrical coupling relationship between them. When all node pairs have completed the judgment and all edges have been constructed, the entire photovoltaic energy storage system network forms an undirected graph. Finally, algorithms such as depth-first search or breadth-first search are applied to find all connected components in the graph. Each independent connected component represents a collaborative voltage regulation partition, and all the nodes contained in it together constitute the members of a partition. All the found connected components are summarized to obtain the collaborative voltage regulation partition set.

[0021] The steps to obtain the aggregate adjustable power range within a partition are: Based on the coordinated voltage regulation zone set, each PV energy storage system in each zone is visited one by one, and the current PV output fluctuation data and the state of charge value of the corresponding energy storage unit of each PV energy storage system are collected in real time to form a set of real-time PV output fluctuation data and energy storage unit state of charge value; Based on real-time PV output fluctuation data and the energy storage unit state of charge value set, the maximum and minimum adjustable active power values, as well as the maximum and minimum reactive power values, of each PV energy storage system in its current state are calculated one by one, forming the adjustable active and reactive power limits of each PV energy storage system; Based on the adjustable active and reactive power limits of each photovoltaic energy storage system, the sum of the adjustable active power limits and the sum of the adjustable reactive power limits of all photovoltaic energy storage systems in each partition are aggregated and calculated to form the aggregated adjustable power range within the partition.

[0022] Specifically, based on the coordinated voltage regulation zone set, the system first establishes a connection with the local monitoring unit or energy management system (EMS) of each photovoltaic energy storage system in the zone according to the zone identifier through a preset communication network address list. The connection uses the Modbus or IEC 61850 MMS protocol based on TCP / IP to ensure the real-time and reliability of data transmission. For each photovoltaic energy storage system, the data collection task is triggered and executed at a fixed time cycle, for example, every 5 seconds. In each collection cycle, the system requests two types of core data from the target photovoltaic energy storage system: the first type is the real-time power output of the photovoltaic array, by querying the DC side or AC side power metering point of its grid-connected inverter, and continuously collecting five instantaneous active power values ​​within a 1-second window period. For example, [510kW, 512kW, 509kW, 513kW, 511kW], the standard deviation of these five sample points is calculated, and the resulting value (approximately 1.58kW in this example) is defined as the PV output fluctuation data at the current moment; the second category is the state of charge (SOC) value of the energy storage unit, which is obtained by directly querying the corresponding data register of the battery management system (BMS), for example, reading 85.5%. These collected raw data are appended with a timestamp and a unique device identification code before transmission and aggregated to the central computing node via a communication link. After receiving the data, the central node classifies it into the corresponding PV energy storage system according to the unique device identification code, and performs preliminary sorting based on the partition information. The PV output fluctuation data and SOC values ​​obtained by all members in the same partition during the same collection cycle are compiled to form a set of real-time PV output fluctuation data and energy storage unit SOC values.

[0023] Based on the real-time PV output fluctuation data and the energy storage unit charge state value set, the system calls the pre-configured device file parameters for each PV energy storage system in the set. The file contains the inverter rated apparent power , Rated capacity of energy storage unit , Maximum charge and discharge rate , and upper and lower state-of-charge limits for safe operation (e.g. , ), first calculate the adjustment range of active power, the maximum adjustable active power Depends on current photovoltaic output and the maximum discharge power of the energy storage unit The sum of, is the rated power of the energy storage unit ( The smaller value of the current state of charge and the allowable discharge power is to prevent the SOC from falling below the limit within the next scheduling period (e.g. 15 minutes). The calculated power, minimum adjustable active power Photovoltaic power output Subtract the maximum charging power of the energy storage unit It is concluded that Also limited by the rated charging power and the need to prevent SOC from exceeding The power limit is then calculated, and the adjustment range of reactive power is then calculated. This range is closely related to the active power output of the current system. The maximum adjustable reactive power With minimum adjustable reactive power Determined by the PQ capability curve of the inverter, the calculation formula is: in, is the current total active power output of the photovoltaic energy storage system, that is ,therefore, and For example, if the rated apparent power of a photovoltaic energy storage system inverter is 1.2MVA, the current photovoltaic output is 0.8MW, the energy storage unit SOC is 60%, and the maximum charge and discharge power is 0.5MW, then its maximum active power limit is However, due to the rated power limit of the inverter, the actual output cannot exceed 1.2MW, and its minimum active power limit is , its maximum reactive power limit is , the minimum reactive power limit is -0.89MVar. The system performs this set of calculations on each photovoltaic energy storage system, and finally obtains the adjustable active and reactive power limits of each photovoltaic energy storage system.

[0024] Based on the adjustable active and reactive power limits of each photovoltaic energy storage system, the system calls the coordinated voltage regulation partition set generated in the previous step and performs power aggregation calculations on a partition basis. This process is performed by traversing each partition in the coordinated voltage regulation partition set. For the currently processed For each partition, the system first extracts a list of unique identifiers of all PV energy storage system members contained in the partition from the partition definition. Then, based on this list, it queries and extracts the unique identifiers of all these members from the set of adjustable active and reactive power limits of each PV energy storage system. 、 、 and Four values, then, perform the aggregation calculation at the partition level, the aggregated maximum adjustable active power of the partition By adding all members in the partition Performing algebraic summation, we get Similarly, the aggregate minimum adjustable active power of the partition , aggregate maximum adjustable reactive power and aggregate minimum adjustable reactive power It is also obtained by summing the limits corresponding to all members in a partition. For example, a partition contains two PV energy storage systems A and B. System A's adjustable active power limit is [0.3MW, 1.2MW], and its adjustable reactive power limit is [-0.89MVar, 0.89MVar]. System B's adjustable active power limit is [0.2MW, 1.0MW], and its adjustable reactive power limit is [-0.6MVar, 0.6MVar]. Then, the aggregated adjustable active power limit sum of the partition is [0.5MW, 2.2MW], and the aggregated adjustable reactive power limit sum is [-1.49MVar, 1.49MVar]. The system completes this aggregation calculation for all partitions in turn and associates the four aggregate limit results of each partition with the partition identifier to form the aggregated adjustable power range within the partition.

[0025] The steps to obtain the partition collaborative response weight set are: Based on the aggregated adjustable power range within the partition, the maximum active power, minimum active power, maximum reactive power, and minimum reactive power of each PV energy storage system in the partition are read in sequence. The voltage change amplitude generated by the node when injecting unit reactive power to all other nodes in the partition is also retrieved simultaneously. This is integrated to form the power regulation boundary of the PV energy storage system in the partition and the node voltage sensitivity value set. Based on the power regulation boundary and node voltage sensitivity value set of the photovoltaic energy storage system within the partition, the coordinated response weight of the photovoltaic energy storage system is calculated using the following formula: ; in, For the The coordinated response weight of the photovoltaic energy storage system, For the The maximum active power limit of a photovoltaic energy storage system, For the The minimum active power limit of a photovoltaic energy storage system, For the The maximum reactive power limit of a photovoltaic energy storage system, For the The minimum reactive power limit of a photovoltaic energy storage system, For the The photovoltaic energy storage system node The voltage change amplitude generated when a node injects unit reactive power, For the The photovoltaic energy storage system node The voltage change amplitude generated when a node injects unit reactive power, is the total number of photovoltaic energy storage systems in the zone, is the power regulation capability evaluation index, It is the electrical impact assessment index; Based on the coordinated response weights of the photovoltaic energy storage systems, the coordinated response weight results of all photovoltaic energy storage systems in the partition are summarized and sorted item by item according to the partition to which they belong, forming a partition coordinated response weight set.

[0026] Specifically, based on the coordinated voltage regulation partition set, the data integration process is started. First, each partition identifier in the partition set is traversed. For each partition, the system obtains the unique device addresses of all photovoltaic energy storage systems from its member list. Then, the system uses these addresses to extract the maximum active power of each member system from the "adjustable active and reactive power limits of each photovoltaic energy storage system" data set calculated and stored in the previous step. , minimum active power , maximum reactive power and minimum reactive power These four key power regulation boundary parameters, while extracting the power boundary parameters, the system performs a parallel query operation to retrieve the "voltage change value between each node" matrix calculated and stored at an earlier stage, for any photovoltaic energy storage system in the partition , the system will traverse all other member systems in the partition (in ), and find and extract the corresponding voltage sensitivity value from the voltage change value matrix , that is, node When unit reactive power is injected at the node This process ensures that only the electrical coupling relationship between the internal members of the partition is considered, and the impact on the nodes outside the partition is ignored. All the power regulation boundary parameters and internal node voltage sensitivity values ​​extracted for the photovoltaic energy storage systems in the same partition are organized together to form a temporary data structure. For example, for partition 1, which contains systems A, B, and C, a data structure containing four power limits for A, B, and C will be formed, as well as 、 、 、 、 、 By repeatedly executing this data extraction and pairing process for all partitions, the data packets of all internal sensitivity values ​​are finally integrated to form the power regulation boundary and node voltage sensitivity value set of the photovoltaic energy storage system in the partition.

[0027] formula: ; The usefulness of the formula is that by calculating the modulus of the two-dimensional vector consisting of the active and reactive power regulation ranges, that is, , comprehensively measures the comprehensive regulation margin of each photovoltaic energy storage system in the power plane, avoiding the one-sidedness of evaluating the active or reactive capacity alone. Secondly, by calculating the average voltage sensitivity of the photovoltaic energy storage system to all other nodes in the partition, that is, , characterizing its electrical influence within the partition, and thirdly, introducing an adjustable evaluation index and This allows grid dispatchers to flexibly adjust the weights of the two indicators based on the current grid operation objectives (for example, prioritizing power balance or maintaining voltage stability). Finally, through normalization (dividing by the sum of the evaluation values ​​of all members in the partition), the relativity and fairness of the weights are ensured. 、 、 、 The steps for obtaining are as follows: The adjustable power boundary of each photovoltaic energy storage system is not set directly, but is the result of dynamic calculation based on the aforementioned step "calculating the maximum and minimum active power values, as well as the maximum and minimum reactive power values, of each photovoltaic energy storage system in its current state based on the real-time photovoltaic output fluctuation data and the energy storage unit charge state value set". These values ​​are directly obtained from the "adjustable active and reactive power limits of each photovoltaic energy storage system" data set generated in this step for the photovoltaic energy storage system currently being processed. For example, in the calculation example, for the photovoltaic energy storage system A in a certain zone, the current adjustable power limits obtained by query are: maximum active power , minimum active power , maximum reactive power , minimum reactive power ,These values ​​reflect the real regulation potential of the equipment under the current working conditions; The steps to obtain the parameter are as follows: The photovoltaic energy storage system node The voltage change amplitude generated when a node injects unit reactive power, that is, the voltage sensitivity, comes from the "voltage change value matrix between nodes" generated by the early step of "based on the grid admittance matrix, injecting unit reactive power into each photovoltaic energy storage system node in turn, calculating and recording the voltage change amplitude of all other nodes in the grid when the node injects unit reactive power point by point". When calculating the weight of a specific partition, the system will extract the voltage sensitivity values ​​between all members within the partition from the matrix based on the partition member list. For example, when calculating the weight of system A in a partition containing three systems A, B, and C, its voltage sensitivity to B and C is required, which can be obtained by querying , ; The steps for obtaining are as follows: this parameter represents the total number of PV energy storage systems in the current calculation zone. It is an integer obtained by counting the member list of a specific zone in the coordinated voltage regulation zone set. For example, if the currently processed zone contains PV energy storage systems A, B, and C, the system obtains by reading the member list of the zone and counting its elements. ; and The acquisition steps are as follows: the two parameters are the power regulation capability evaluation index and the electrical influence evaluation index, whose values ​​are determined by the grid operation strategy and optimized and calibrated through offline simulation analysis. The specific setting process is: construct a series of typical grid operation scenarios (such as load peak, photovoltaic power generation, line light load, etc.), and in each scenario, set a set of and The initial value of , ), run the coordinated control simulation, and evaluate the control effect. The evaluation indicators include voltage over-limit rate, power fluctuation smoothing effect, regulation response speed, etc. Then, use optimization algorithms such as grid search or gradient descent to adjust the and Repeat the simulation and evaluation until a set of optimal comprehensive evaluation indicators is found. and In this example, we set , ,Representatives believe that power regulation capability and electrical influence are equally important; Calculation process: Taking a partition containing three photovoltaic energy storage systems A, B, and C as an example, calculate the coordinated response weight of each system , the known parameters are as follows: System A: , , , ; System B: , , , ; System C: , , , ; Voltage sensitivity: , , , , , ; Total number of systems in the partition , evaluation index , ; First, calculate the unnormalized evaluation value (i.e., the numerator of the formula) for each system separately: For System A: Power regulation capability items: ; Electrical impact items: ; Evaluation value of system A: ; For System B: Power regulation capability items: ; Evaluation value of system B: ; For System C: Power regulation capability items: ; Electrical impact items: ; Evaluation value of system C: ; Then, calculate the normalized denominator (i.e. the sum of all system evaluation values): Denominator = ; Finally, the collaborative response weight of each system is calculated: ; ; ; The results show that the coordinated response weights of systems A, B, and C in this partition are 0.3297, 0.2638, and 0.4065, respectively. These values ​​are dimensionless relative weights, and their sum is 1. Among them, system C has the highest weight, which means that its comprehensive regulation capability and electrical influence are the best among the three. It should undertake the largest proportion of tasks in the subsequent power command allocation. System B has the lowest weight and undertakes the smallest proportion of regulation tasks. These calculated weight values ​​are the coordinated response weights of the photovoltaic energy storage system.

[0028] Based on the coordinated response weights of the PV energy storage system calculated in the previous step, the system initiates the weight set generation process. This process aims to create a structured data set to store the weight distribution results for all partitions. The process first initializes an empty data structure, the partition coordinated response weight set, as the final output. Then, the process begins to traverse each partition one by one, using the coordinated voltage regulation partition set as the index. For the specific partition currently being processed, the system has already obtained the coordinated response weight values ​​of all PV energy storage system members within that partition. For example, for partition 1, the weights of its members A, B, and C have been calculated to be 0.3297, 0.2638, and 0.4065, respectively. The process uses the partition's unique identifier (e.g., "Partition_1") as the primary key and pairs the "device ID-weight value" of all members within that partition (e.g., {"System_A": 0.3297, "System_B": 0.2638, "System_C": 0.4065}) as the value associated with the primary key, and then store this complete "partition-weight list" entry into the previously initialized partition collaborative response weight set. The process is repeated continuously, and the program will then process partition 2 and partition 3 until all partitions in the collaborative voltage regulation partition set have been processed, and their respective weight allocation results have been summarized and organized into the final set in a unified format to form a partition collaborative response weight set.

[0029] The steps for obtaining the power transfer distribution factor set are: The complete power transfer distribution factor matrix and the identifier of the target congested line in the control state are retrieved from the power grid dispatching center. The column vectors of the grid connection points of each photovoltaic energy storage system partition relative to the target congested line are screened to generate a set of power transfer distribution factors.

[0030] Specifically, a data request is initiated through the MMS secure communication link based on the IEC 61850 standard established with the energy management system (EMS) of the power grid dispatching center to retrieve the complete power transfer distribution factor (PTDF) matrix stored on the dispatching center server and the target congested line identifier that is currently under real-time monitoring and regulation. The complete power transfer distribution factor matrix is ​​a large, pre-calculated offline sparse matrix, whose row index corresponds to the unique number of all transmission lines in the power grid, the column index corresponds to the unique number of all power grid nodes, and the matrix elements are Representation node Injection of unit active power to the line The contribution of the power flow, the target congestion line identifier is generated by the line overload monitoring module of the dispatching center. For example, when the real-time power transmission of line "L-45-78" (connecting nodes 45 and 78) reaches 95% of its thermal stability limit, the system will issue "L-45-78" as the target congestion line identifier. After obtaining these two data, the system first locates the corresponding row in the complete power transmission distribution factor matrix according to the target congestion line identifier "L-45-78". Then, the system needs to identify the grid connection point node number of each photovoltaic energy storage system partition. This information is read from the pre-configured "partition-grid connection point mapping table", which lists in detail the unique number of the aggregated equivalent grid connection node of each coordinated voltage regulation partition in the power grid topology. For example, the grid connection point of partition 1 is node 23, and the grid connection point of partition 2 is node 56. Then, the system uses these grid connection point node numbers as indexes to filter out the corresponding column vector elements in the "L-45-78" row of the complete power transmission distribution factor matrix, that is, and ,These filtered “partition identifier-PTDF value” pairs are summarized to generate a set of power transmission distribution factors.

[0031] The steps to obtain the maximum collaborative flow diversion potential are: Based on the power transfer distribution factor set, the current active power value, maximum adjustable active power value, and minimum adjustable active power value of all members in each PV energy storage system partition are read. The upward active power difference and downward active power difference within the same partition are summed to form a partition directional adjustable active power pair. Based on the partitioned directional adjustable active power pairs, the maximum cooperative power flow diversion potential of the target line is calculated using the following formula: ; in, The maximum collaborative flow diversion potential of the target line, is the number of photovoltaic energy storage system partitions, For the The power transfer distribution factor of each partitioned grid connection point relative to the target congested line, For the The total amount of active power that can be adjusted upwards in each partition, For the The total amount of active power that can be adjusted downward in each partition, For the The collaborative response quality factor of each partition is is the quality factor impact index, For Sign function that takes a sign value.

[0032] Specifically, based on the power transmission distribution factor set generated in the previous process, the system starts the calculation process of the partition adjustable power. The process is iterated based on the coordinated voltage regulation partition set. For each partition in the partition set, for example, The system first identifies a list of all PV energy storage system members in the partition, and then, for each member in the list (e.g., member ), the system reads three key status values ​​from the "adjustable active and reactive power limits of each photovoltaic energy storage system" data set through the internal data interface: current active power value This value is obtained by collecting the readings of the power meter at the grid connection point of the photovoltaic energy storage system in real time; the maximum adjustable active power value and the minimum adjustable active power value These two values ​​are previously calculated dynamically based on the photovoltaic output, energy storage charge state and equipment physical constraints. Next, the system calculates the value for each member. Calculate its upward and downward active power regulation margins, and the upward regulated active power difference is defined as , the downward regulated active power difference is defined as After completing the difference calculation of all members in the partition, the system aggregates these differences at the partition level, that is, summing up the difference in active power adjustment of all members in the same partition to obtain the total adjustment capacity of the partition. Similarly, sum up the difference in active power reduction of all members to get the total reduction capacity , this pair of aggregated values ​​( , ) with the partition identifier Association,Finally, by performing this calculation for all partitions, partition-wise adjustable active power pairs are formed.

[0033] formula: , the formula is useful in that, through the symbolic function As an intelligent switch, it automatically selects the optimal regulation direction for each partition. If the power injection in a partition will aggravate the congestion ( ), the formula will automatically call its power reduction On the contrary, if its power injection can relieve congestion ( ), then call its power increase This design makes the formulation of power flow diversion strategy adaptive at the physical level. In addition, the formula combines the adjustment amount with the power transmission distribution factor. The absolute value of is multiplied to ensure that the partition with greater influence on the line flow has a greater weight in the total potential. More importantly, the coordinated response quality factor is introduced. and impact index , taking into account soft indicators such as the internal coordination of the partition and communication reliability, so that the evaluation results reflect not only "how much can be adjusted" but also "how well it is adjusted"; The steps for obtaining are as follows: this parameter is the total number of PV energy storage system partitions. Its value is determined by querying and counting the number of independent partitions contained in the "cooperative voltage regulation partition set" generated in the initial step of "a method for coordinated control of PV energy storage systems". This process is automatically counted by the software. For example, if the system divides all PV energy storage systems into 3 independent cooperative voltage regulation partitions after voltage sensitivity cluster analysis, then The value of is 3; The steps to obtain the parameter are as follows: The power transfer distribution factor of each partition grid connection point relative to the target congested line is directly read from the "power transfer distribution factor set" generated in the previous step of this method. This set has been filtered according to the target congested line identifier issued by the power grid dispatching center, and stores the key-value pairs of each partition and its corresponding power transfer distribution factor value. For example, for the congested line L-45-78, the factor of partition 1 is queried from the set. , the factor of partition 2 , the factor of partition 3 , these dimensionless values ​​are directly used in subsequent calculations; and The steps to obtain are: The total amount of active power that can be adjusted upward and downward for each partition is derived from the "partition directional adjustable active power pair" data set generated immediately before this step. This set stores the aggregated upward and downward power margins for each partition. For example, the system queries this set and obtains: the adjustable power pair for partition 1 is ( , ), the adjustable power pair of partition 2 is ( , ), the adjustable power pair of partition 3 is ( , ); The steps to obtain the parameter are as follows: The coordinated response quality factor of each partition is used to quantify the coordination performance within the partition. Its calculation comprehensively considers the stability of communication within the partition and the consistency of energy storage state. The calculation formula is: ,in, is the normalized value of the average communication delay of all members in the partition, is the normalized value of the standard deviation of the energy storage state of charge of all members in the partition, and the weight and Set according to operation and maintenance experience, for example 、 , indicating that both are equally important. For example, the average communication delays of the three partitions monitored are [80ms, 120ms, 70ms], and the SOC standard deviations are [4%, 9%, 3%], respectively. The reasonable range of communication delay is set to [50ms, 150ms], and the reasonable range of SOC standard deviation is set to [2%, 10%]. The maximum and minimum normalization method is used to calculate: , calculated similarly , ; The acquisition steps are as follows: This parameter is the quality factor influence index, which is pre-set by the grid dispatch strategy and is used to adjust the influence of the coordinated response quality factor in the calculation. Its value is determined by regression analysis of historical dispatch data. The success rate of flow diversion and the cost of regulation are selected from the following values ​​(e.g., from 0.5 to 2.0, with a step size of 0.1), and the value with the best comprehensive benefit is selected. For example, the analysis shows that when When , the system can more effectively utilize high-quality resources and avoid secondary adjustments caused by poor response of low-quality resources, so in this calculation, ; Calculation process: Based on the results obtained from the above parameters, the maximum coordinated flow diversion potential is calculated: Partition 1 ( ): , the power should be adjusted up.

[0034] Contribution value = ; Partition 2 ( ): , the power should be reduced.

[0035] Contribution value = ; Partition 3 ( ): , the power should be adjusted up.

[0036] Contribution value = ; Sum the contribution of each partition: ; The results show that the maximum power flow reduction that can be achieved on the target congested line through the coordinated action of all PV energy storage systems is 1.57149 MW. This value is the maximum coordinated power flow diversion potential of the target line. It provides a direct and quantitative basis for the power grid dispatching center to determine whether the current line congestion problem can be solved by relying solely on these distributed resources. If the power flow reduction required by the dispatching center is less than 1.57149 MW, the task can be executed; otherwise, other backup measures need to be activated.

[0037] The steps for obtaining the power instruction sequence of the energy storage unit are: The total amount of power flow grooming tasks currently required on the target line is obtained from the grid dispatch center. The maximum coordinated power flow grooming potential is then called. The contribution value provided by each PV energy storage system partition in the calculation of the maximum coordinated power flow grooming potential is extracted one by one. The ratio of the partition's contribution value to the maximum coordinated power flow grooming potential is used as the partition's contribution ratio. Based on the contribution ratio, the power grooming instructions allocated to each PV energy storage system partition are determined, and the power grooming instructions for the partition are generated. Based on the partitioned power grooming instructions, the partitioned coordinated response weight set is called and the coordinated response weights of all PV energy storage systems in each partition are extracted one by one. The coordinated response weight of each PV energy storage system is divided by the sum of the coordinated response weights of all PV energy storage systems in the partition to complete the normalization of the weights within the partition. The partitioned power grooming instructions are then subdivided into each PV energy storage system in the partition based on the normalized weight values ​​to generate the initial power instruction for each PV energy storage system. Based on the initial power command of each photovoltaic energy storage system, the real-time state of charge of the current energy storage unit of the corresponding photovoltaic energy storage system and the real-time adjustable active power range of the inverter are retrieved one by one to compare whether the initial power command falls within the real-time adjustable active power range of the inverter. If the initial power command exceeds the range, the boundary value of the real-time adjustable active power range of the inverter is used as the corrected power command of the photovoltaic energy storage system. After all corrections are completed, the determined corrected power command is sent to the inverter of the corresponding photovoltaic energy storage system to generate a power command sequence for the energy storage unit.

[0038] Specifically, first, through the preset communication interface with the power grid dispatching center, the total amount of current flow diversion tasks currently required to be completed by the target line issued by it is received. For example, the dispatching center specifies that a 1.2MW flow reduction is required on the congested line L-45-78. This value is the total task amount. At the same time, the system calls the "maximum collaborative flow diversion potential" calculated in the previous key step, whose value is 1.57149MW. Since the total task amount is less than the maximum potential, the system confirms the feasibility of the task and continues to execute it. Then, the system traces back the calculation process of the maximum collaborative flow diversion potential and extracts the contribution value provided by each photovoltaic energy storage system partition in the calculation one by one. These contribution values ​​are intermediate results that have comprehensively considered the regulation capacity, electrical impact and collaborative response quality of each partition. For example, the contribution values ​​extracted from partitions 1, 2, and 3 are 0.8034MW, 0.13725MW and 0.63084MW respectively. Then, the system calculates the contribution ratio of each partition, that is, divides the contribution value of a single partition by the maximum collaborative flow diversion potential, and obtains the contribution ratio of partition 1. , the proportion of partition 2 is , the proportion of partition 3 is Based on this contribution ratio, the system distributes the total power flow diversion task in proportion and determines the power regulation amount that each partition needs to bear. This regulation amount refers to the amount of active power injection that needs to be increased or decreased at the partition grid connection point. The calculation method is to first calculate a global regulation ratio, that is, the ratio of the total task amount to the maximum potential ( ), and then apply this ratio to the original adjustable power amount of each zone when calculating the potential. Specifically, for zone 1 that needs to increase power (its original increase potential is 5.2MW), its power regulation instruction is For zone 2 (whose original reduction potential is 6.1MW) that needs to reduce power, the instruction is: , while for zone 3 which needs to increase power (its original increase potential is 7.0MW), the instruction is , where the positive sign indicates an increase in active output and the negative sign indicates a decrease. Finally, these calculated power adjustment values ​​are bound to the corresponding partition identifiers to generate power grooming instructions for the partition.

[0039] Based on the power grooming instructions for the partitions generated in the previous process, the system starts the instruction refinement program and processes each partition one by one. For any partition, such as partition 1, the system first reads its power grooming instructions from the instruction set, which means that a total of 3.97MW of active power needs to be increased. Next, the system calls the previously calculated and stored "partition coordinated response weight set" to extract the coordinated response weights of all PV energy storage system members within partition 1. For example, partition 1 contains systems A, B, and C, and their corresponding coordinated response weights are 0.3297, 0.2638, and 0.4065, respectively. Since these weights have been normalized by dividing by the sum of the evaluation values ​​of all members in the partition during calculation, the sum is exactly 1, so there is no need to perform normalization again. The system directly uses these weight values ​​as the basis for power allocation, and subdivides the total power grooming instructions of the partition to each PV energy storage system in the partition according to the weight ratio. For system A, the power regulation it should bear is the total instructions of the partition multiplied by its own weight, that is, Similarly, the regulation amount that system B should bear is , the regulation amount that system C should bear is These calculated values ​​represent the ideal power change requirements for each PV energy storage system. The system repeats this allocation process for all partitions and their internal members, decomposing the macro power instructions of each partition into specific power adjustment values ​​for each independent PV energy storage system. Finally, these values ​​are summarized to generate the initial power instructions for each PV energy storage system.

[0040] Based on the initial power command of each photovoltaic energy storage system, the system enters the final command verification and issuance stage. This process traverses each photovoltaic energy storage system to ensure the executable nature of the command. For a single photovoltaic energy storage system, such as system A, the system first reads its initial power command as an increase of 1.31MW. Then, the system retrieves the system's current operating status data, including the current active power value of its grid connection point. (e.g. 0.8MW), and the previously calculated real-time adjustable active power range of the inverter that has fully considered the physical limitations of the equipment and the energy storage charge state, for example, the range of system A is [0.3MW, 1.2MW]. The system calculates the target power setting value of system A based on the initial instruction, that is, , then, the target value is compared with the adjustable range, and it is found that 2.11MW exceeds the upper limit of its maximum adjustable active power of 1.2MW. Therefore, the initial power instruction is not feasible for system A. The system immediately starts the correction program to correct the power instruction of the system to the boundary value of the adjustable range, that is, the corrected target power setting value is 1.2MW. This 1.2MW is the corrected power instruction of the photovoltaic energy storage system. After completing the instruction verification and necessary correction of all photovoltaic energy storage systems, the system converts these finalized and executable corrected power instructions into specific operation instructions for the energy storage unit. Specifically, the power instruction of the energy storage unit is obtained by subtracting the current real-time output of the photovoltaic array from the corrected total power instruction. For example, if the real-time photovoltaic output of system A is 0.8MW and its corrected power instruction is 1.2MW, the instruction sent to its energy storage unit is , indicating that the energy storage unit needs to discharge 0.4MW. Finally, the system encapsulates these calculated energy storage unit power command values ​​into a message that complies with the device communication protocol (such as Modbus or IEC 61850) and sends it to the inverter or energy management system of the corresponding photovoltaic energy storage system through the communication network for execution, thereby generating an energy storage unit power command sequence.

[0041] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A photovoltaic energy storage system coordinated control method, characterized in that: The following steps are involved: Obtain the grid topology and line parameters, establish the grid admittance matrix, calculate the voltage change caused by unit reactive power injection between any two PV energy storage system nodes, group all PV energy storage system nodes, and obtain a set of coordinated voltage regulation partitions; Based on the coordinated voltage regulation zone set, the real-time photovoltaic output fluctuation data and the energy storage unit charge state of each photovoltaic energy storage system in each zone are read to obtain the aggregate adjustable power range within the zone, and then the zone coordinated response weight set is established based on the aggregate adjustable power range within the zone; Obtain the power transfer distribution factor matrix and congested line identifiers from the grid dispatch center, extract the power transfer distribution factor values ​​of each PV energy storage system's partitioned grid connection point relative to the congested line, and obtain a set of power transfer distribution factors. This set of power transfer distribution factors is then calculated with the partitioned adjustable power to determine the maximum collaborative power flow diversion potential. Based on the maximum collaborative power flow diversion potential and the partition collaborative response weight set, an energy storage unit power instruction sequence is generated.

2. The photovoltaic energy storage system coordinated control method according to claim 1, characterized in that: The steps for obtaining the coordinated voltage regulation partition set are: Obtaining the topological connection relationship of all nodes in the power grid and the electrical parameters of the corresponding lines, calculating the self-admittance and mutual-admittance of the lines one by one based on the topological connection relationship and the electrical parameters of the lines, and constructing a power grid admittance matrix based on the self-admittance and mutual-admittance of the lines to generate a power grid admittance matrix; Based on the grid admittance matrix, inject unit reactive power into each photovoltaic energy storage system node in turn, calculate and record the voltage change amplitude of all other nodes in the grid when the node injects unit reactive power, and obtain the voltage change value between each node; Based on the voltage variation values ​​between the nodes, the voltage variation between the nodes of the photovoltaic energy storage system is counted node by node, a screening threshold for the node voltage variation is set, and the nodes whose voltage variation reaches or exceeds the screening threshold are classified into the same group to form a coordinated voltage regulation partition set.

3. The photovoltaic energy storage system coordinated control method according to claim 1, characterized in that: The steps for obtaining the aggregate adjustable power range within the partition are: Based on the coordinated voltage regulation zone set, each photovoltaic energy storage system in each zone is visited one by one, and the current photovoltaic output fluctuation data and the state of charge value of the corresponding energy storage unit of each photovoltaic energy storage system are collected in real time to form a set of real-time photovoltaic output fluctuation data and energy storage unit state of charge value; Based on the real-time photovoltaic output fluctuation data and the energy storage unit state of charge value set, the maximum and minimum active power values, as well as the maximum and minimum reactive power values, that can be adjusted for each photovoltaic energy storage system in its current state are calculated one by one to form the adjustable active and reactive power limits of each photovoltaic energy storage system; Based on the adjustable active and reactive power limits of each photovoltaic energy storage system, the sum of the adjustable active power limits and the sum of the adjustable reactive power limits of all photovoltaic energy storage systems in each partition are aggregated and calculated to form an aggregated adjustable power range within the partition.

4. The photovoltaic energy storage system coordinated control method according to claim 1, characterized in that: The steps for obtaining the partition coordinated response weight set are: Based on the aggregated adjustable power range within the partition, the maximum active power, minimum active power, maximum reactive power, and minimum reactive power of each photovoltaic energy storage system in the partition are sequentially read, and the voltage change amplitude generated by the node when injecting unit reactive power to all other nodes in the partition is simultaneously retrieved, and the power regulation boundary and node voltage sensitivity value set of the photovoltaic energy storage system in the partition are integrated; Calculating a coordinated response weight of the photovoltaic energy storage system based on a power regulation boundary and a node voltage sensitivity value set of the photovoltaic energy storage system in the partition; Based on the coordinated response weights of the photovoltaic energy storage systems, the coordinated response weight results of all photovoltaic energy storage systems in the partition are summarized and sorted item by item according to the partition to which they belong, forming a partition coordinated response weight set.

5. The photovoltaic energy storage system coordinated control method according to claim 1, characterized in that: The steps for obtaining the power transmission distribution factor set are: The complete power transfer distribution factor matrix and the identifier of the target congested line in the control state are retrieved from the power grid dispatching center. The column vectors of the grid connection points of each photovoltaic energy storage system partition relative to the target congested line are screened to generate a set of power transfer distribution factors.

6. The photovoltaic energy storage system coordinated control method according to claim 1, characterized in that: The steps for obtaining the maximum collaborative flow diversion potential are: Based on the power transfer distribution factor set, read the current active power value, maximum adjustable active power value, and minimum adjustable active power value of all members in each photovoltaic energy storage system partition, and sum the upward active power difference and downward active power difference within the same partition to form a partition directional adjustable active power pair; Based on the partitioned directional adjustable active power pairs, the maximum collaborative power flow diversion potential of the target line is calculated.

7. The photovoltaic energy storage system coordinated control method according to claim 1, characterized in that: The steps for obtaining the power instruction sequence of the energy storage unit are: The total amount of power flow grooming tasks currently required to be completed on the target line is obtained from the power grid dispatching center. The maximum coordinated power flow grooming potential is called, and the contribution value provided by each photovoltaic energy storage system partition in the calculation of the maximum coordinated power flow grooming potential is extracted one by one. The ratio of the partition's contribution value to the maximum coordinated power flow grooming potential is used as the partition's contribution ratio. The power grooming instructions allocated to each photovoltaic energy storage system partition are determined based on the contribution ratio, and the power grooming instructions for the partition are generated. Based on the partitioned power grooming instructions, the partitioned coordinated response weight set is called and the coordinated response weights of all photovoltaic energy storage systems in each partition are extracted one by one. The coordinated response weight of each photovoltaic energy storage system is divided by the sum of the coordinated response weights of all photovoltaic energy storage systems in the partition in turn to complete the normalization of the weights within the partition. The partitioned power grooming instructions are then subdivided into each photovoltaic energy storage system in the partition according to the normalized weight values ​​to generate an initial power instruction for each photovoltaic energy storage system.

8. The photovoltaic energy storage system coordinated control method according to claim 7, characterized in that: The step of acquiring the energy storage unit power instruction sequence further includes: based on the initial power instruction of each photovoltaic energy storage system, retrieving the real-time state of charge of the current energy storage unit of the corresponding photovoltaic energy storage system and the real-time adjustable active power range of the inverter one by one, comparing whether the initial power instruction falls within the real-time adjustable active power range of the inverter; if the initial power instruction exceeds the range, using the boundary value of the real-time adjustable active power range of the inverter as the corrected power instruction of the photovoltaic energy storage system; after all corrections are completed, issuing the determined corrected power instruction to the inverter of the corresponding photovoltaic energy storage system to generate the energy storage unit power instruction sequence.

Images