A control method and device for a photovoltaic-storage AC / DC microgrid based on a configurable group of controllable objects.
By establishing a standardized virtual topology model and a hierarchical coordination control strategy, the software fragmentation and data compatibility issues of the microgrid coordinator were resolved, enabling adaptive control across devices and topologies, and improving the stability and engineering applicability of the microgrid.
Patent Information
- Application Number
- CN202611547404.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-10-08
- Publication Date
- 2026-11-10
AI Technical Summary
In existing technologies, microgrid coordinator controllers need to be customized for different topologies and devices, resulting in software version fragmentation, strong coupling between control strategies and physical devices making it difficult to achieve multi-scenario adaptation, poor data compatibility under heterogeneous communication networks, and high system integration complexity.
A control method for photovoltaic-storage AC/DC microgrids based on configurable controllable object groups is adopted. A standardized virtual topology model is established, and the physical devices in the microgrid are mapped to controllable object groups in the topology model. Real-time operating parameters are collected through heterogeneous communication networks to perform discrete information modeling, and a hierarchical coordinated control strategy is executed based on aggregated state variables to achieve a unified control logic architecture and multi-protocol adaptation.
It enables seamless application of the same software version in different scenarios, reduces development and maintenance costs, improves the stability of microgrid operation and battery life, and simplifies system integration complexity.
Smart Images

Figure CN123074013A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of microgrid control technology, and more specifically to a control method and equipment for photovoltaic-storage AC / DC microgrids based on configurable controllable object groups. Background Technology
[0002] Photovoltaic-storage AC / DC microgrids, composed of photovoltaics, energy storage, and AC / DC loads, have become an important technical means to improve energy utilization efficiency and enhance power supply reliability. These microgrids achieve power conversion through power electronic converters (PCS / DCDC), enabling them to operate in grid-connected mode to exchange power with the main grid, or to operate independently in islanded mode to supply loads. Application scenarios cover industrial and commercial parks, data centers, islands, and remote areas.
[0003] As the control core of a microgrid system, the microgrid coordinator undertakes the core functions of energy management, power scheduling, and safe and stable control. In existing engineering practices, the microgrid coordinator typically adopts a "monitoring + logic control" architecture. That is, the upper-level monitoring unit collects data from each device through the communication bus, calculates the power commands for each converter according to preset logic strategies (such as peak-valley arbitrage and planned curve tracking), and then sends them down to the lower level for execution. However, due to the extremely complex topology of photovoltaic-storage AC / DC microgrids (such as DC coupling, AC coupling, AC / DC hybrid, etc.), and the significant differences in communication protocols (such as Modbus, CAN, I04, and proprietary protocols) and control logic of PCS, DC-DC converters, BMS, and other devices from different manufacturers in the market, the development of the microgrid coordinator faces enormous adaptation challenges.
[0004] Currently, for different types of microgrid projects, technicians typically need to perform customized software development based on the specific topology and equipment models on-site. For example, when a microgrid includes grid-connected energy storage (GFM) and grid-linked energy storage (GFL), separate adaptation codes for voltage source control and current source control need to be written; when a photovoltaic DC-DC-MPPT is connected to the DC side, DC bus voltage stabilization logic needs to be developed separately. This customized development model based on topology type leads to severe software fragmentation, resulting in long development cycles, high labor costs, and difficult maintenance. Once on-site equipment is replaced or the topology is fine-tuned, it is often necessary to modify the code and debug on-site, making it difficult to guarantee system stability.
[0005] Furthermore, existing microgrid control strategies often lack standardized information model support. The coordinating controller directly interacts with the underlying measurement points of physical devices (such as voltage, current, and SOC), without abstracting or aggregating the devices. This results in a strong coupling between the control algorithm and the physical topology, making it impossible to achieve unified power allocation and SOC balancing at a higher level. For example, in photovoltaic power-limited control, without a unified aggregated power-limited percentage model, the controller needs to calculate the power-limited value for each individual device, making it difficult to meet the rapid response requirements of large-scale clusters. Similarly, in energy storage SOC balancing, without a capacity-weighted aggregated power model, uneven power distribution can easily occur, accelerating battery aging.
[0006] In summary, existing technologies have the following shortcomings: 1) They lack a unified, standardized topology modeling method, requiring customized development of microgrid coordinators for different manufacturers' equipment and topologies, resulting in low software reusability; 2) Control strategies are strongly bound to physical devices, failing to decouple into a two-layer architecture of "aggregate decision-distribution execution," making it difficult to achieve adaptive tailoring and adaptation for multiple scenarios; 3) They do not systematically address data compatibility issues under heterogeneous communication networks, leading to high system integration complexity and high operation and maintenance costs. Therefore, there is an urgent need for a control method and device for optical-storage AC / DC microgrids based on microgrid coordinators that can achieve standardized modeling, decoupled control, and multi-protocol adaptation. Summary of the Invention
[0007] This invention aims to solve the technical problems in the prior art, such as the need for customized development of microgrid coordinator for different topologies and devices, resulting in software version fragmentation, strong coupling between control strategies and physical devices making it difficult to achieve multi-scenario adaptation, and poor data compatibility under heterogeneous communication networks leading to high system integration complexity. It provides a control method and device for optical-storage AC / DC microgrids based on configurable control object groups.
[0008] To achieve the above objectives, the present invention adopts the following technical solution:
[0009] A control method for a photovoltaic-storage AC / DC microgrid based on a configurable group of controllable objects, the method comprising:
[0010] A standardized virtual topology model of a photovoltaic-storage AC / DC microgrid is established in the microgrid coordinator controller. Typical physical devices in the microgrid are mapped to controllable object groups in the topology model according to device type. The controllable object groups include photovoltaic control device object groups on the AC side and DC side, grid-type energy storage device object groups, and grid-connected energy storage device object groups.
[0011] The microgrid coordinator collects real-time operating parameters of the physical single devices corresponding to each controllable object group through a heterogeneous communication network for discrete information modeling, and obtains the aggregated state variables of each controllable object group based on the discrete information.
[0012] The microgrid coordinator executes a hierarchical coordination control strategy based on the aggregated state variables, including: an aggregated control layer: calculating group-level control instructions for each controllable object group based on the aggregated state variables and preset control objectives, including power allocation instructions or voltage / frequency reference instructions; and an allocation execution layer: calculating control instructions for each device based on the group-level control instructions and device operating status data, and issuing them to the corresponding devices for execution.
[0013] Each of the controllable object groups is configured with participation parameters. When the photovoltaic-storage AC / DC microgrid is missing physical equipment corresponding to a certain type of controllable object group, the participation parameters corresponding to the controllable object group with missing physical equipment are set to default invalid values, so that the corresponding controllable object group is automatically removed in the calculation of aggregated state variables, thereby realizing the logical pruning of the controllable object group.
[0014] In some embodiments of the present invention, obtaining the aggregated state variables of each controllable object group includes:
[0015] For a group of photovoltaic control equipment, the peak power, current output power, and maximum allowable output power of each individual device in the controllable group are collected; and aggregated into the total peak power, total output power, and power limit percentage of the corresponding controllable group.
[0016] For DC-side grid-type energy storage equipment groups, the charging and discharging power limits, current output power, output voltage, current charge level, and total battery capacity of each individual device in the controllable group are collected and aggregated into the total charging and discharging power limits, total output power, output voltage, SOC, and total capacity of the corresponding controllable group. For AC-side grid-type energy storage equipment groups, the output frequency is additionally collected and aggregated into the output frequency of the corresponding controllable group.
[0017] For grid-connected energy storage device groups, the charging and discharging power limits, current output power, current charge level, and total battery capacity of each individual device in the controllable group are collected and aggregated into the total charging and discharging power limits, total output power, SOC, and total capacity of the corresponding controllable group.
[0018] In some embodiments of the present invention, for a group of photovoltaic control devices, the participating configuration parameter is the peak power of each individual device;
[0019] For the grid-connected energy storage device group, the participating configuration parameters are the total battery capacity, maximum discharge power, and maximum charging power of each individual device;
[0020] For the grid-type energy storage device group, the participating configuration parameters are the total battery capacity, maximum discharge power, and maximum charging power of each individual device;
[0021] The default value of the configuration parameters is set to 0.
[0022] In some embodiments of the present invention, the aggregation control layer performs DC-side control decisions, including:
[0023] Monitor the aggregated state variables of the DC-side grid-type energy storage device group. When any of the following conditions are met, such as aggregated SOC being higher than the first preset SOC threshold, aggregated charging power being higher than the first preset charging power threshold, or output voltage being higher than the preset voltage threshold, issue a power limiting command to the DC-side photovoltaic control device group. Reduce the power limiting percentage of the photovoltaic control device group according to the preset decreasing slope. When the triggering condition is eliminated, restore the maximum power point tracking control mode and increase the power limiting percentage according to the preset increasing slope until the maximum power point is reached.
[0024] A voltage reference value is issued to the DC-side grid-type energy storage device group, which automatically maintains output voltage stability and completes power balance based on the voltage reference value;
[0025] Based on the DC bus power balance relationship, and combining the aggregated SOC of the DC-side grid-type energy storage device group and the grid-connected energy storage device group, the total power setting value of the DC-side grid-connected energy storage device group is calculated.
[0026] In some embodiments of the present invention, the aggregation control layer performs power limiting decisions for a group of AC-side photovoltaic control devices, including:
[0027] When the microgrid is in grid-connected operation mode, the aggregate charging power, aggregate SOC, and grid-connected power of the AC-side energy storage device group are monitored. If any of the following conditions are met: the grid-connected SOC is higher than the second preset SOC threshold, the grid-connected charging power is higher than the second preset charging power threshold, or the active power at the grid-connected point is less than the anti-reverse current reserve value, the power limiting mode of the AC-side photovoltaic control device group is triggered, and the power limiting percentage is reduced by a preset decreasing slope. When the triggering condition is eliminated, the maximum power point tracking control mode is restored, and the power limiting percentage is increased by a preset increasing slope until the maximum power point is reached. The grid-connected SOC is calculated using a capacity-weighted average method based on the aggregate SOC and total aggregate battery capacity of each AC-side energy storage device group. The grid-connected charging power is the sum of the aggregate charging power of each AC-side energy storage device group.
[0028] When the microgrid is in off-grid operation mode, the aggregate SOC, aggregate charging power, and output frequency of the AC-side grid-type energy storage device group are monitored. When any of the following conditions are met, such as aggregate SOC being higher than the second preset SOC threshold, aggregate charging power being higher than the third preset charging power threshold, or output frequency being higher than the preset frequency threshold, the power limiting mode of the AC-side photovoltaic control device group is triggered, and the power limiting percentage is reduced by a preset slope.
[0029] In some embodiments of the present invention, the aggregation control layer performs power allocation decisions for a group of AC-side energy storage devices, including:
[0030] When the microgrid is in grid-connected operation mode, all AC-side energy storage devices operate in current source mode.
[0031] Based on the aggregated SOC and total battery capacity of the AC-side energy storage device object group, and in accordance with the SOC balancing principle, the aggregated active power command of the entire station is distributed to each AC-side energy storage device object group to obtain the active power setting value of each group.
[0032] Based on the maximum reactive power ratio of the AC-side energy storage device object group, the total station aggregated reactive power command is distributed to each AC-side energy storage device object group to obtain the reactive power setting value of each group; the total station aggregated active / reactive power command is the sum of the active / reactive power output power setting values of each AC-side energy storage device object group.
[0033] When the microgrid is in off-grid operation mode, the grid-type energy storage device group operates as a voltage source;
[0034] The reference frequency setting value and reference voltage setting value are issued to the grid-type energy storage device object group;
[0035] Based on the aggregated SOC and total battery capacity of the AC-side energy storage device group, the active power setting value of the grid-connected energy storage device group is calculated according to the SOC balancing principle.
[0036] Based on the maximum reactive power ratio of each AC-side energy storage device group under the current active power control conditions, the reactive power setting value of the grid-connected energy storage device group is calculated.
[0037] In some embodiments of the present invention, the allocation execution layer performs DC-side allocation decisions, including:
[0038] For the photovoltaic control equipment object group, based on the power limit percentage determined by the aggregated control layer, the maximum power limit setting value of each individual device is calculated and issued;
[0039] For the grid-type energy storage device group, the voltage offset is calculated based on the deviation between the SOC of each individual device and the average SOC, combined with a preset proportional coefficient; the voltage offset is superimposed on the reference voltage to generate the voltage setting value of each individual device and then distributed.
[0040] For grid-connected energy storage equipment groups, the total power setting value is proportionally allocated to each individual device based on the SOC and total battery capacity of each device, and the charging and discharging power is limited.
[0041] In some embodiments of the present invention, the allocation execution layer performs communication-side allocation decisions, including:
[0042] For the group of photovoltaic control equipment on the AC side, based on the power limit percentage determined by the aggregated control layer and combined with the peak power of each individual device, the maximum power limit setting value of each individual device is calculated and issued.
[0043] For AC-connected grid-type energy storage equipment groups, in grid-connected mode, based on the group-level active power setting value, according to the SOC of each individual device and the total battery capacity, active power is evenly distributed based on SOC to generate active power instructions for each individual device; at the same time, reactive power instructions for each individual device are generated according to the maximum reactive power ratio under the current active power control conditions.
[0044] In off-grid mode, based on the deviation between the SOC of each individual device and the average SOC, and combined with a preset proportional coefficient, the frequency offset is calculated; the frequency offset is superimposed on the reference frequency to generate the frequency setting value for each individual device and send it down.
[0045] At the same time, reactive power is allocated according to the proportion of the maximum reactive power that each individual device can generate, and the voltage setting value corresponding to each individual device is calculated and issued according to the allocated reactive power using the droop formula.
[0046] For AC-side grid-connected energy storage device groups, based on the group-level active power setting value, the SOC of each individual device and the total battery capacity, active power is allocated according to SOC balancing, and active power commands for each individual device are generated and issued.
[0047] At the same time, based on the group-level reactive power setting value, reactive power is allocated according to the proportion of the maximum reactive power that each individual device can issue, and reactive power commands for each individual device are generated and issued.
[0048] In some embodiments of the present invention, the heterogeneous communication network includes:
[0049] The first communication network used to acquire electrical quantities and protection signals of equipment, and to issue emergency power cut-off commands or fast trip commands;
[0050] A second communication network used to acquire basic electrical parameters and status variables of the equipment, as well as to issue charging and discharging power settings or start / stop commands;
[0051] A third communication network used to acquire equipment operating data and measurement values, as well as to issue voltage reference values, frequency reference values, or power limit percentage settings;
[0052] The first, second, and third communication networks all use standard industrial protocols.
[0053] The present invention further provides a microgrid coordinator controller, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above method.
[0054] The present invention has the following beneficial effects:
[0055] (1) A standardized topology model of controllable object group was constructed, and a general control algorithm was developed for the standardized topology model. The ever-changing physical microgrid was abstracted into a unified logical architecture, and the adaptive tailoring of the topology structure was achieved through parameter settings. There is no need to adjust the code for the specific project equipment model and topology structure, so that the same software version can be seamlessly applied to different scenarios. This fundamentally solves the problem of software version fragmentation, reduces development and maintenance costs, and greatly improves the engineering applicability and market competitiveness of the product.
[0056] (2) By using information aggregation modeling, the complex measurement points of the underlying devices are aggregated into high-level cluster state variables (such as total SOC and total power), enabling control decisions to be made from a global perspective; then, the aggregated instructions are precisely decomposed to each individual machine for execution through the allocation execution layer. This architecture not only simplifies the control logic, but also realizes advanced control functions such as SOC balancing and power limiting, thereby improving the stability of microgrid operation and battery life.
[0057] (3) Except for the set values, all operating parameters are acquired in real time through a heterogeneous communication network composed of standard industrial protocols. Control commands are also issued through these protocols, enabling the microgrid coordinator to be compatible with equipment from different manufacturers. Access can be completed by simply configuring the corresponding communication protocol parameters, which greatly reduces the complexity of system integration. Attached Figure Description
[0058] Figure 1 This is a schematic diagram of a standardized topology for a photovoltaic-storage AC / DC microgrid established in one embodiment of the present invention.
[0059] Figure 2 This is a schematic diagram of the microgrid topology of an energy storage power station according to a certain embodiment of the present invention.
[0060] Figure 3 This is a schematic diagram of the microgrid topology of a photovoltaic-storage AC-coupled power station according to a certain embodiment of the present invention.
[0061] Figure 4 This is a schematic diagram of the microgrid topology of a photovoltaic-storage DC-DC coupled power station according to a certain embodiment of the present invention.
[0062] Figure 5 This is a schematic diagram of the AC / DC coupled microgrid topology of DC-side photovoltaic according to a certain embodiment of the present invention.
[0063] Figure 6This is a schematic diagram of the AC / DC coupled microgrid topology of the AC-side photovoltaic system according to a certain embodiment of the present invention.
[0064] Figure 7 This is a schematic diagram of a microgrid topology according to one embodiment of the present invention.
[0065] Figure 8 The graph shows the change curves of the current total output power of the microgrid GFM_PCS and the power limit percentage setting value of DCAC_MPPT during the coordinated control process of a certain embodiment of the present invention.
[0066] Figure 9 The graph shows the change curves of the current SOC and maximum output power settings of the microgrid GFM_DCDC and DCCDC_MPPTx during the coordinated control process in a certain embodiment of the present invention.
[0067] Figure 10 This is a curve showing the SOC change of the PCS on the AC side of the microgrid during the coordinated control process according to a certain embodiment of the present invention.
[0068] Figure 11 This is a curve showing the SOC change of the DC-side DC-DC converter of the microgrid during the coordinated control process according to a certain embodiment of the present invention. Detailed Implementation
[0069] The technical solution of the present invention will be further described below with reference to the accompanying drawings and specific embodiments.
[0070] Example 1
[0071] This embodiment provides a control method for a photovoltaic-storage AC / DC microgrid based on a microgrid coordinator controller, which specifically includes the following steps:
[0072] S100 establishes a standardized topology for photovoltaic-storage AC / DC microgrids.
[0073] Due to the complexity and diversity of microgrid topologies, microgrid coordinator controllers typically require customized control strategies for different microgrid sites, resulting in a large workload for software customization, multiple maintenance versions, and high management costs. To standardize the control strategies for AC / DC microgrids, standardized modeling of AC / DC microgrids is first performed.
[0074] In addition, the operating logic of actuators such as DC / DC converters, PCS, or MPPTs in microgrids varies from manufacturer to manufacturer. To ensure the universality of this standardized control method, typical actuators need to be mapped as controllable object groups when modeling AC / DC microgrids.
[0075] Based on the above requirements, a standardized virtual topology model of the photovoltaic-storage AC / DC microgrid is first established.
[0076] The physical devices in the photovoltaic-storage AC / DC microgrid are mapped into six groups of controllable objects, including: the DCAC_MPPT group of photovoltaics composed of MPPT controllers on the AC side, the GFM_PCS group composed of grid-connected energy storage converters (GFM-PCS), the GFL_PCS group composed of grid-connected energy storage converters (GFL-PCS), the DCCDC_MPPT group of photovoltaics composed of MPPT controllers on the DC side, the GFM_DCDC group composed of grid-connected DC / DC converters, and the GFL_DCDC group composed of grid-connected DC / DC converters.
[0077] Among them, DCAC_MPPT group, GFM_PCS group and GFL_PCS group are connected to the AC bus of the microgrid, DCC_MPPT group, GFM_DCDC group and GFL_DCDC group are connected to the DC bus of the microgrid, and the AC bus of the microgrid is connected to the transformer via the grid connection point switch (PCC_Breaker), and after voltage transformation by the transformer, it is connected to the main grid.
[0078] The microgrid coordinator connects to the individual physical devices mapped in each controllable object group via a heterogeneous communication network. This network is used to acquire parameters for discrete information modeling and to issue control commands. The heterogeneous communication network includes:
[0079] Fiber optic GOOSE communication is used to acquire high-speed electrical quantities and protection signals at the millisecond level, as well as to issue emergency power cut-off commands or fast trip commands.
[0080] The RS485 serial link is used to obtain the basic electrical parameters and status of the device based on the Modbus RTU protocol, as well as to issue charging and discharging power settings or start / stop commands.
[0081] Industrial Ethernet is used to acquire equipment operating data and measurement values based on Modbus TCP protocol or IEC 60870-5-104 protocol, as well as to issue voltage reference values, frequency reference values or power limit percentage settings.
[0082] The command issuance interval can be set from milliseconds to seconds and can be adjusted according to the communication delay, i.e. the response time of the execution device. By appropriately increasing the command issuance interval according to the sum of the communication delay and the response time, or by reading the output feedback, it can be ensured that the previous control command has been executed before the second command is issued.
[0083] The microgrid coordinator is compatible with the above communication methods to facilitate access to devices from different manufacturers.
[0084] The standardized virtual topology for photovoltaic-energy storage AC / DC conversion constructed based on the above process is as follows: Figure 1 As shown.
[0085] S200 DC-side Control Object Information Modeling
[0086] S210 DCDC_MPPT Information Modeling
[0087] S211 DCDC_MPPTx Discrete Information Modeling (DCDC_MPPTx refers to each independent MPPT controller in the DCCDC_MPPT group, where x is the sequence number. The following text uses a similar approach to describe the independent devices in the group)
[0088] First, discrete information modeling is performed on DCDC_MPPTx, as shown in Table 1:
[0089] Table 1 Discrete Information Modeling of DCDC_MPPTx
[0090]
[0091] In this application, the parameters in the discrete information modeling are data obtained directly by the microgrid coordinator through communication with the physical device, or values preset according to the control strategy. Parameters with the _Set suffix are preset control parameters for the device.
[0092] S212 DCDC_MPPT Aggregated Information Modeling
[0093] After completing the discrete information modeling of DCDC_MPPT, the aggregated information modeling of DCDC_MPPT is performed, as shown in Table 2:
[0094] Table 2 DCDC_MPPT Aggregated Information Modeling
[0095]
[0096] Assuming there are a+1 DCCDC_MPPT units, the DCCDC_MPPT_Curtailment_Pct_Now is:
[0097]
[0098] S220 GFM_DCDC Information Modeling
[0099] S221 GFM_DCDCx Discrete Information Modeling
[0100] First, discrete information modeling is performed on GFM_DCDCx, as shown in Table 3:
[0101] Table 3 Discrete Information Modeling of GFM_DCDCx
[0102]
[0103] S222 GFM_DCDC Aggregated Information Modeling
[0104] After completing the discrete information modeling of GFM_DCDC, the aggregated information modeling of GFM_DCDC is performed, as shown in Table 4:
[0105] Table 4 GFM_DCDC Aggregated Information Modeling
[0106]
[0107] Assuming there are a total of b+1 GFM_DCDC units, the GFM_DCDC_SOC is:
[0108]
[0109] The result of obtaining GFM_DCDC_Vout is:
[0110]
[0111] S230 GFL_DCDC Information Modeling
[0112] S231 GFL_DCDCx Discrete Information Modeling
[0113] First, discrete information modeling is performed on GFL_DCDCx, as shown in Table 5:
[0114] Table 5 Discrete Information Modeling of GFL_DCDCx
[0115]
[0116] S232 GFL_DCDC Aggregated Information Modeling
[0117] After completing the discrete information modeling of GFL_DCDC, the aggregated information modeling of GFL_DCDC is performed, as shown in Table 6:
[0118] Table 6 GFL_DCDC Aggregated Information Modeling
[0119]
[0120] Assuming there are a total of b+1 GFL_DCDC units, the GFL_DCDC_SOC is:
[0121]
[0122] S300 AC-side control object information modeling
[0123] S310 DCAC_MPPT Information Modeling
[0124] S311 DCAC_MPPTx Discrete Information Modeling
[0125] First, discrete information modeling is performed on DCAC_MPPTx, as shown in Table 7:
[0126] Table 7 Discrete Information Modeling of DCAC_MPPTx
[0127]
[0128] S312 DCAC_MPPT Aggregated Information Modeling
[0129] After completing the discrete information modeling of DCAC_MPPT, the aggregated information modeling of DCAC_MPPT is performed, as shown in Table 8:
[0130] Table 8 DCAC_MPPT Aggregated Information Modeling
[0131]
[0132] Assuming there are a+1 DCAC_MPPTs in total, the DCAC_MPPT_Curtailment_Pct_Now is:
[0133]
[0134] S320 GFM_PCS Information Modeling
[0135] S321 GFM_PCSx Discrete Information Modeling
[0136] First, discrete information modeling is performed on GFM_PCSx, as shown in Table 9:
[0137] Table 9 GFM_PCSx Discrete Information Modeling
[0138]
[0139] S322 GFM_PCS Aggregated Information Modeling
[0140] After completing the discrete information modeling of GFM_PCS, the aggregated information modeling of GFM_PCS is performed, as shown in Table 10:
[0141] Table 10 GFM_PCS Aggregated Information Modeling
[0142]
[0143] Assuming there are a total of e+1 GFM_PCS, the GFM_PCS_SOC is:
[0144]
[0145] S330 GFL_PCS Information Modeling
[0146] S331 GFL_PCSx Discrete Information Modeling
[0147] First, discrete information modeling is performed on GFL_PCSx, as shown in Table 11:
[0148] Table 11 Discrete Information Modeling of GFL_PCSx
[0149]
[0150] S332 GFL_PCS Aggregated Information Modeling
[0151] After completing the discrete information modeling of GFL_PCS, the aggregated information modeling of GFL_PCS is performed, as shown in Table 12:
[0152] Table 12 GFL_PCS Aggregated Information Modeling
[0153]
[0154] Assuming there are a total of f+1 GFL_PCS, the GFM_PCS_SOC is:
[0155]
[0156] S400 DC bus standard control strategy
[0157] S410 Aggregation Control Strategy
[0158] S411 DCDC_MPPT Aggregation Control Strategy
[0159] Since GFM_DCDC is the cornerstone of DC bus voltage stability on the DC side, there are three conditions for DCCDC_MPPT to perform photovoltaic power limiting:
[0160] (1) When the SOC of GFM_DCDC is too high, photovoltaic power limiting should be implemented. Otherwise, when GFM_SOC is fully charged, it will lose its photovoltaic absorption capacity, causing the DC bus voltage to rise rapidly and then collapse.
[0161] (2) When the charging power of GFM_DCDC is too high, photovoltaic power limiting should be implemented. Otherwise, when the charging power of GFM_DCDC reaches the maximum charging power, it will be unable to absorb the excess photovoltaic power, causing the DC bus voltage to rise rapidly and then collapse.
[0162] (3) Photovoltaic power limiting should be started after the DC bus voltage of GFM_DCDC is too high. This is because a high DC bus voltage indicates that the photovoltaic absorption capacity of GFM_DCDC has reached the critical point. In order to avoid voltage collapse caused by further increase in DC bus voltage, power limiting should be started.
[0163] Based on the defined control objectives and preset values, as shown in Table 13:
[0164] Table 13 Set values preset according to the control target
[0165]
[0166] The three conditions for triggering photovoltaic power curtailment are related by an "OR" condition; if any one of them occurs, the power curtailment mode will be entered. The triggering conditions are as follows:
[0167]
[0168] After triggering, the photovoltaic power curtailment is downwards according to the slope PARA_PV_Curt_Slope_down. It controls the interval time:
[0169]
[0170] The system will return to maximum power point control mode only when all three conditions are no longer met, and will increase the slope PARA_PV_Curt_Slope_up upwards until the maximum power point is reached.
[0171]
[0172] S412 GFM_DCDC Aggregation Control Strategy
[0173] Since the control target of the DC-side GFM_DCDC is the DC bus voltage, the power balance output is automatically achieved based on the output power of DCDC_MPPT, the output power of GFL_DCDC, and the input power of PCS (GFM / GFL).
[0174]
[0175] Therefore, the aggregated GFM_DCDC only needs to set the reference DC bus output voltage setpoint, and the output power is automatically balanced by its voltage closed-loop control.
[0176] S413 GFL_DCDC Aggregation Control Strategy
[0177] Since the power of GFM_DCDC is automatically obtained through power balancing, the sum of the output power of GFM_DCDC and GFL_DCDC is the total output power required by the DCCDC. GFL_DCDC mainly acts as a current source to share the power control burden of GFM_DCDC, and it needs to control the aggregated SOC of GFM_DCDC and GFL_DCDC to achieve power balance. The setpoint PARA_DCDC_SOC_MIN is defined as the lower limit of the DCCDC's SOC discharge; discharge is prohibited below this value. Therefore, the control power of GFL_DCDC is:
[0178] During discharge, the power distribution is as follows:
[0179]
[0180] During charging, the power allocation is as follows:
[0181]
[0182] In addition, GFL_DCDC_Pset needs to be limited according to the upper and lower limits of the charge and discharge power. Definition:
[0183]
[0184]
[0185] The amplitude limiting formula is obtained as follows:
[0186]
[0187] S420 allocation control strategy
[0188] S421 DCDC_MPPTx Allocation Control Strategy
[0189] The maximum power limit setting value, DCCDC_MPPTx_P_MAX_Set, for each MPPT is calculated based on the power limit percentage DCCDC_MPPT_Curtailment_Pct_Set and the peak power of DCCDC_MPPTx using the aggregation control strategy.
[0190] S422 GFM_DCDCx Allocation Control Strategy
[0191] GFM_DCDCx typically operates in parallel in droop mode, where V is the actual output voltage, Vref is the output reference voltage, I is the output current, and k_DCDC is the droop factor. The droop formula is:
[0192]
[0193] To achieve SOC equilibrium, the above formula will be used. Add a SOC-related one above deviation :
[0194]
[0195] When in a discharge state, if the SOC is low, the reference voltage needs to be reduced, thereby reducing the discharge power ratio.
[0196] When charging, if the SOC is low, the reference voltage needs to be reduced to increase the charging power ratio.
[0197] Therefore, the reference voltage offset method is unaffected by the charge / discharge conversion direction and can achieve a smooth transition. Definition The average SOC of GFM_DCDCx SOC deviation set With voltage deviation proportionality coefficient, design For: Definition:
[0198]
[0199]
[0200] In addition, settings are required. Upper and lower limits ( ).
[0201]
[0202] For GFM_DCDC This is the reference voltage value for GFM_DCDC, i.e., GFM_DCDC_Vset, which is set to the rated voltage on the DC side of the PCS. The resulting voltage setting value for GFM_DCDCx, used to control SOC equalization, is:
[0203]
[0204] S423 GFL_DCDCx Allocation Control Strategy
[0205] In the allocation control strategy, the aggregated control value is based on SOC balancing. The power is allocated to each GFL_DCDCx device. Assuming there are a total of c+1 GFL_DCDC devices, the GFL_DCDCx_Pset power allocation method based on SOC equalization is as follows, where x = 0~c:
[0206] when When ≥0:
[0207]
[0208] when <0 o'clock:
[0209]
[0210] In addition, GFL_DCDCx_Pset needs to be limited according to the upper and lower limits of the charge and discharge power. For GFL_DCDCx, the definition is:
[0211] Maximum discharge power
[0212] Maximum charging power
[0213] The amplitude limiting formula is obtained as follows:
[0214]
[0215] S500 AC bus standard control strategy
[0216] S510 Aggregation Control Strategy
[0217] S511 DCAC_MPPT Aggregation Control Strategy
[0218] (1) Grid-connected anti-reverse current power limiting
[0219] If there is no anti-reverse current control requirement during grid-connected operation, then photovoltaic power generation can proceed at maximum power. When there is an anti-reverse current requirement during grid-connected operation, the excess photovoltaic power needs to be absorbed by GFM_PCS and GFL_PCS. When the absorption capacity of the PCS approaches the critical value, photovoltaic power limiting is required to prevent power reverse current. Preset values are defined as shown in Table 14.
[0220] Table 14. Preset values based on control objectives
[0221]
[0222] Since both GFM_PCS and GFL_PCS operate as current sources during grid connection, PCS_SOC is defined as:
[0223]
[0224] Define the total output power of PCS, PCS_Pout, as:
[0225]
[0226] Define the total maximum charging power of PCS as PCS_Chr_P_MAX as:
[0227]
[0228] PCC_P represents the active power at the grid connection point. The condition for triggering photovoltaic power curtailment is as follows:
[0229]
[0230] When grid connection is required to prevent backflow, if any one of the above three conditions is met, photovoltaic power curtailment will be triggered. After triggering, the photovoltaic power curtailment will be limited downwards according to the slope PARA_PV_Curt_Slope_down. The system will revert to maximum power point tracking (MPPT) mode only when all three conditions are no longer met, increasing the slope PARA_PV_Curt_Slope_up until the maximum power point is reached. The specific control method is the same as in S411.
[0231] (2) Off-grid photovoltaic power curtailment
[0232] When operating off-grid, excess photovoltaic (PV) power needs to be absorbed by GFM_PCS and GFL_PCS. When the absorption capacity of the PCS approaches the critical value, PV power curtailment is required to prevent power reverse flow. Furthermore, if the PCS cannot absorb the PV active power in time, the frequency will rise. Therefore, a higher frequency power curtailment value, PARA_PCS_F_HIGH, is defined. PV power curtailment is also required when the output frequency of the grid-connected PCS, GFM_PCS_Fout, exceeds this value. Since power balancing is actively achieved by the grid-connected PCS (GFM_PCS) during off-grid operation, the off-grid absorption capacity is primarily judged by the operating status of the grid-connected PCS. PV power curtailment is also required when the charging power or SOC of the grid-connected PCS is too high. Therefore, all conditions triggering PV power curtailment are:
[0233]
[0234] When the grid is off-grid, if any one of the above three conditions is met, photovoltaic power curtailment will be triggered. After triggering, the photovoltaic power curtailment will be limited downward according to the slope PARA_PV_Curt_Slope_down. The system will revert to maximum power point tracking (MPPT) mode only when all three conditions are no longer met, increasing the slope PARA_PV_Curt_Slope_up upwards until the maximum power point is reached. The specific control method is the same as in S411.
[0235] S512 GFM_PCS Aggregation Control Strategy
[0236] (1) Grid connection control strategy
[0237] During grid connection, GFM_PCS and GFL_PCS operate as current sources, with the total active power command (PCS_Pset) and total reactive power command (PCS_Qset) for the entire station. The total station commands need to be allocated to the aggregated grid-connected active power command GFM_PCS_Pset and the aggregated grid-connected active power command GFL_PCS_Pset. The active power commands are allocated according to the SOC of the grid-connected and grid-connected energy storage and the total battery capacity, resulting in GFM_PCS_Pset as follows:
[0238] when When ≥0:
[0239]
[0240] when <0 o'clock:
[0241]
[0242] Reactive power is allocated according to the proportion of maximum reactive power under current active power control conditions, ensuring that high active power output results in low reactive power output. GFM_PCS_Q_MAX is defined as the maximum reactive power output of the network PCS, and GFM_PCS_S_MAX is the maximum apparent power of the network PCS.
[0243] The GFM_PCS_Q_MAX value is obtained as follows:
[0244]
[0245] Similarly, GFL_PCS_Q_MAX represents the maximum reactive power output of the grid-connected PCS, and GFL_PCS_S_MAX represents the maximum apparent power of the grid-connected PCS.
[0246] The GFL_PCS_Q_MAX value is obtained as follows:
[0247]
[0248] The final allocation formula for GFM_PCS_Qset is as follows:
[0249]
[0250] In addition, GFM_PCS_Pset needs to be limited according to the upper and lower limits of the charging and discharging power. For GFM_PCS, the definition is:
[0251] Maximum discharge power
[0252] Maximum charging power
[0253] The amplitude limiting formula is obtained as follows:
[0254]
[0255] In addition, GFM_PCS_Qset needs to be limited according to the maximum reactive power. Definition:
[0256]
[0257] The amplitude limiting formula is obtained as follows:
[0258]
[0259] (2) Off-grid control strategy
[0260] Since the control target of the AC side GFM_PCS is the AC voltage and frequency value, it will automatically adjust the active and reactive power output based on the load and distributed power sources to achieve frequency and voltage stability according to the frequency setting value GFM_PCS_Fset and the voltage setting value GFM_PCS_Vset. Therefore, the energy storage coordinator only needs to set the frequency setting value GFM_PCS_Fset and the voltage setting value GFM_PCS_Vset according to the microgrid requirements.
[0261] S513 GFL_PCS Aggregation Control Strategy
[0262] (1) Grid connection control strategy
[0263] When connected to the grid, GFL_PCS and GFM_PCS operate as current sources, with the aggregated active power command for the entire station being PCS_Pset and the aggregated reactive power command being PCS_Qset. The entire station's commands need to be allocated to the aggregated grid-connected active power command GFM_PCS_Pset and the aggregated grid-connected active power command GFL_PCS_Pset. The active power commands are allocated according to the SOC of the grid-connected energy storage and the total battery capacity, resulting in GFL_PCS_Pset as follows:
[0264] when When ≥0:
[0265]
[0266] when <0 o'clock:
[0267]
[0268] The reactive power is allocated according to the proportion of the maximum reactive power under the current active power control conditions, so that the active power output is high and the reactive power output is low.
[0269] The GFL_PCS_Qset allocation formula is obtained as follows:
[0270]
[0271] (2) Off-grid control strategy
[0272] Since the power of GFM_PCS is automatically obtained from power balancing when off-grid, the sum of the active power outputs of GFM_PCS and GFL_PCS is the total active power output required by the PCS, and the sum of the reactive power outputs is the total reactive power output required by the PCS. GFL_PCS mainly acts as a current source to share the power control burden of GFM_PCS, and it needs to control the aggregated SOC of GFM_PCS and GFL_PCS to achieve power balance. The setpoint PARA_PCS_SOC_MIN is defined as the lower limit of PCS SOC discharge; discharge is prohibited below this value. Therefore, the control power of GFL_PCS is:
[0273] During discharge, the power distribution should be as follows:
[0274]
[0275] During charging, the power allocation should be as follows:
[0276]
[0277] The reactive power is allocated according to the proportion of the maximum reactive power under the current active power control conditions, so that the active power output is high and the reactive power output is low.
[0278] The GFL_PCS_Qset allocation formula is obtained as follows:
[0279]
[0280] (3) Active and reactive power limiting strategies
[0281] In addition, GFL_PCS_Pset needs to be limited according to the upper and lower limits of the charge and discharge power. For GFL_PCS, the definition is:
[0282] Maximum discharge power
[0283] Maximum charging power
[0284] The amplitude limiting formula is obtained as follows:
[0285]
[0286] In addition, GFL_PCS_Qset must be limited according to the maximum reactive power. Definition:
[0287]
[0288] The amplitude limiting formula is obtained as follows:
[0289]
[0290] S520 allocation control strategy
[0291] S521 DCAC_MPPTx Allocation Control Strategy
[0292] The maximum power limit setting value DCAC_MPPTx_P_MAX_Set for each MPPT is calculated based on the power limit percentage DCAC_MPPT_Curtailment_Pct_Set and the maximum power of DCAC_MPPT using the aggregation control strategy.
[0293]
[0294] S522 GFM_PCSx Allocation Control Strategy
[0295] (1) Grid connection control strategy
[0296] When connected to the grid, the GFM_PCS operates as a current source. The power allocation goal is primarily to achieve SOC equalization. A setpoint, PARA_PCS_SOC_MIN, is defined as the lower limit for SOC discharge of both the GFM_PCS and GFL_PCS; discharge is prohibited below this value. Assuming a total of e+1 GFM_PCS units, the power allocation method for GFM_PCSx_Pset based on SOC equalization is as follows, where x = 0~e:
[0297] when When ≥0:
[0298]
[0299] when <0 o'clock:
[0300]
[0301] Reactive power is allocated according to the proportion of maximum reactive power under the current active power control conditions, ensuring that higher active power output results in lower reactive power output. GFM_PCSx_Q_MAX is defined as the maximum output reactive power, and GFM_PCSx_S_MAX as the maximum apparent power, resulting in GFM_PCSx_Q_MAX as follows:
[0302]
[0303] The allocation formula for GFM_PCSx_Qset is obtained as follows, where x = 0 ~ e:
[0304]
[0305] In addition, GFM_PCSx_Pset needs to be limited according to the upper and lower limits of the charging and discharging power. Definition:
[0306] Maximum discharge power
[0307] Maximum charging power
[0308] The amplitude limiting formula is obtained as follows:
[0309]
[0310] In addition, GFM_PCSx_Qset needs to be limited according to the maximum reactive power. Definition:
[0311]
[0312] The amplitude limiting formula is obtained as follows:
[0313]
[0314] (2) Off-grid control strategy
[0315] GFM_PCSx typically operates in parallel in droop mode. F is the actual output frequency, Fref is the output reference frequency, P is the output active power, and k_F is the frequency droop coefficient. The frequency droop formula is:
[0316]
[0317] To achieve SOC equilibrium, the above formula will be used. Add a SOC-related one above deviation :
[0318]
[0319] When in a discharge state, if the SOC is low, the reference frequency needs to be reduced, thereby reducing the discharge power ratio.
[0320] When charging, if the SOC is low, the reference frequency needs to be reduced to increase the charging power ratio.
[0321] Therefore, the reference frequency offset method is unaffected by the charge / discharge conversion direction and can achieve a smooth transition. Definition The average SOC of GFM_PCSx SOC deviation set Frequency deviation proportionality coefficient, design for:
[0322]
[0323]
[0324] In addition, settings are required. Upper and lower limits ( ).
[0325]
[0326] For GFM_PCS, This is the reference frequency value of GFM_PCS (i.e., GFM_PCS_Fset), which is set as the reference output frequency of GFM_PCS. The resulting frequency setting value for GFM_PCSx, which controls SOC equalization, is:
[0327]
[0328] The voltage control of GFM_PCSx is also achieved through droop control, where V is the actual output voltage, Vref is the output reference voltage, Q is the output reactive power, and k_V is the voltage droop coefficient. The voltage droop formula is:
[0329]
[0330] Reactive power is allocated according to the proportion of maximum reactive power under the current active power control conditions, ensuring that higher active power output results in lower reactive power output. GFM_PCSx_Q_MAX is defined as the maximum output reactive power, and GFM_PCSx_S_MAX as the maximum apparent power, resulting in GFM_PCSx_Q_MAX as follows:
[0331]
[0332] The expected allocation formula for GFM_PCSx_Qset is obtained as follows, where x = 0 ~ e:
[0333]
[0334] Substituting into the voltage droop formula, substituting GFM_PCSx_Qset into Q, and substituting GFM_PCSx_Vout into V, we obtain the reference voltage setting value GFM_PCSx_Vset as:
[0335]
[0336] S523 GFL_PCSx Allocation Control Strategy
[0337] Whether grid-connected or off-grid, GFL_PCS operates as a current source. The power allocation goal is primarily to achieve SOC equalization. A setpoint, PARA_PCS_SOC_MIN, is defined as the lower limit for SOC discharge of GFM_PCS and GFL_PCS; discharge is prohibited below this value. Assuming a total of f+1 GFL_PCS, the power allocation method for GFL_PCSi_Pset based on SOC equalization is as follows, where x = 0~f:
[0338] when When ≥0:
[0339]
[0340] when <0 o'clock:
[0341]
[0342] Reactive power is allocated according to the proportion of maximum reactive power under the current active power control conditions, ensuring that higher active power output results in lower reactive power output. GFL_PCSx_Q_MAX is defined as the maximum output reactive power, and GFL_PCSx_S_MAX as the maximum apparent power, resulting in GFL_PCSx_Q_MAX as follows:
[0343]
[0344] The allocation formula for GFL_PCSx_Qset is obtained as follows, where x = 0 ~ f:
[0345]
[0346] In addition, GFL_PCSx_Pset needs to be limited according to the upper and lower limits of the charging and discharging power. For GFL_PCS, the definition is:
[0347] Maximum discharge power
[0348] Maximum charging power
[0349] The amplitude limiting formula is obtained as follows:
[0350]
[0351] In addition, GFL_PCSx_Qset must be limited according to the maximum reactive power. Definition:
[0352]
[0353] The amplitude limiting formula is obtained as follows:
[0354]
[0355] Example 2
[0356] This embodiment illustrates the adaptive capability of the standardized photovoltaic-storage AC / DC microgrid model constructed by the method of the present invention for model pruning.
[0357] The established standardized photovoltaic-storage AC / DC microgrid model is customizable, and the control strategies for each part are decoupled, eliminating the need for redevelopment. When devices are connected, the discrete parameters acquire collected values; when devices are not connected, the discrete parameters use user-defined default values. Model customization can be achieved simply by setting the default values of the corresponding parameters to 0. Specific examples are as follows:
[0358] I. DC-side model trimming
[0359] (1) DCDC_MPPT clipping
[0360] DC-side DC-MPPT can be trimmed. If there is no DC-side DC-MPPT, simply set the default value of DC-MPPTx_Peak_P to 0. The aggregated DC-MPPT_Peak_P will naturally also be 0. When the photovoltaic peak power is 0, the issued power limit value is fixed at 0.
[0361] Furthermore, after removing DCDC_MPPT, the remaining GFM_DCDC and GFL_DCDC can still operate according to the original strategy. GFM_DCDC continues to stabilize the DC bus voltage and allocate power according to SOC, while GFL_DCDC allocates power according to SOC and shares the output power of GFM_DCDC.
[0362] (2) GFL_DCDC clipping
[0363] The DC-side GFL_DCDC can be trimmed. If there is no GFL_DCDC on the DC side, simply set the default values of GFL_DCDCx_C, GFL_DCDCx_Dischr_P_MAX, and GFL_DCDCx_Chr_P_MAX to 0. Then, the SOC equalization will exclude GFL_DCDC as 0, and the control value will also be limited to 0.
[0364] Furthermore, after removing GFL_DCDC, the remaining GFM_DCDC and DCDC_MPPT can still operate according to their original strategies. GFM_DCDC continues to stabilize the DC bus voltage and allocates power according to SOC. DCDC_MPPT typically discharges at maximum power, and then limits power when the power-limiting condition is reached.
[0365] (3) GFM_DCDC clipping
[0366] DC-side DC-DC converters are typically not pruned because they are required to support the DC bus voltage in normal scenarios. Even if both DC-DC_MPPT and GFL_DC are pruned when GFM_DC is present, the original strategy remains unaffected. After pruning GFL_DC, GFL_DCx_C becomes 0. When the original strategy performs balancing between GFM_DC and GFL_DC, GFL_DC is naturally excluded, and only internal balancing of GFM_DC is performed. Furthermore, in this pruning scenario, the PCS DC side is directly connected to the DC-DC converter and battery, essentially functioning as a traditional AC PCS. Therefore, this standard AC / DC microgrid model can be transformed into a traditional AC microgrid model through pruning.
[0367] In scenarios involving residential energy storage or low-voltage, low-power systems, the GFM_DCDC model can also be pruned. In this case, no control is required on the DC side (default parameters are all configured to zero, and the corresponding control points are not associated with communication data), and the control strategy on the AC side remains unaffected after pruning the GFM_DCDC model. However, such scenarios typically do not require the use of a microgrid coordinator controller; therefore, this section only describes a special case to illustrate the universality of the model pruning scenario in this invention.
[0368] II. Communication Side Model Cutting
[0369] (1) DCAC_MPPT clipping
[0370] The AC-side DCAC_MPPT can be trimmed. If there is no DCAC_MPPT on the AC side, simply set the default value of DCAC_MPPTx_Peak_P to 0. The aggregated DCAC_MPPT_Peak_P will naturally also be 0. When the photovoltaic peak power is 0, the issued power limit value is fixed at 0.
[0371] Furthermore, after removing DCAC_MPPT, the remaining GFM_PCS and GFL_PCS can still operate according to their original strategies. When grid-connected, both GFM_PCS and GFL_PCS operate as current sources, and SOC balancing control remains effective. When off-grid, GFM_PCS continues to stabilize the AC side frequency and voltage, and allocates power according to SOC; GFL_PCS also allocates power according to SOC, sharing the output power of GFM_PCS.
[0372] (2) GFL_PCS clipping
[0373] The AC-side GFL_PCS can be trimmed. If there is no GFL_PCS on the AC side, simply set the default values of GFL_PCSx_C, GFL_PCSx_Dischr_P_MAX, and GFL_PCSx_Chr_P_MAX to 0. Then, the SOC equalizer will exclude GFL_PCS as 0, and the control value will also be limited to 0.
[0374] Furthermore, after removing GFL_PCS, the remaining GFM_PCS and DCAC_MPPT can still operate according to their original strategies. When connected to the grid, GFM_PCS operates as a current source, and SOC balancing control remains effective. When disconnected from the grid, it continues to stabilize the AC side frequency and voltage, distributing power according to SOC. DCAC_MPPT typically discharges at maximum power, and then limits power once the power-limiting condition is reached.
[0375] (3) GFM_PCS clipping
[0376] If there is only a need for grid-connected operation, since both GFM_PCS and GFL_PCS operate as current sources during grid connection (at least one of GFM_PCS and GFL_PCS must be retained during grid connection), GFM_PCS can be trimmed. By setting the default values of GFM_PCSx_C, GFM_PCSx_Dischr_P_MAX, and GFM_PCSx_Chr_P_MAX to 0, GFM_PCS will naturally be excluded as 0 in the SOC balancing process, and the control value will also be limited to 0.
[0377] If there is a need for off-grid operation, the AC-side network GFM_PCS cannot be removed because it is required to support AC-side frequency and voltage. Furthermore, in this case, even if both DCAC_MPPT and GFL_PCS are removed, it will not affect the original strategy operation. After GFL_PCS is removed, GFL_PCSx_C becomes 0. When the original strategy performs GFM_PCS and GFL_PCS balancing, GFL_PCS is naturally excluded, and only the internal balancing of GFM_PCS is performed.
[0378] III. Trimming and Adapting Typical Topologies
[0379] (1) Energy storage power station
[0380] Configuration reduction: On the DC side, reduce DCCDC_MPPT and GFL_DCDC, and configure one GFM_DCDC per PCS; on the AC side, reduce DCAC_MPPT. Retain GFM_DCDC, GFM_PCS, and GFL_PCS.
[0381] Topology Characteristics: The system contains no photovoltaic power source and relies entirely on the grid for charging and discharging. The DC bus voltage is maintained stable by the GFM_DCDC group through voltage droop control. In grid-connected operation, GFM_PCS and GFL_PCS operate as current sources, performing SOC balancing dispatch; in off-grid operation, the GFM_PCS group serves as the sole voltage and frequency support source, constructing a local microgrid to supply power to critical loads. For example... Figure 2 .
[0382] (2) Photovoltaic-storage AC-coupled power station
[0383] Trimming configuration: Trim DCDC_MPPT and GFL_DCDC on the DC side; retain GFM_DCDC, DCAC_MPPT, GFM_PCS and GFL_PCS.
[0384] Topology: Photovoltaic modules are directly connected to the AC bus via DCAC_MPPT. Only the grid-connected energy storage GFM_DCDC is retained on the DC side as DC bus voltage support. When photovoltaic output is excessive, surplus power is fed into the grid via the AC bus or converted to DC and stored in batteries via GFM_DCDC; when photovoltaic output is insufficient, the energy released by GFM_DCDC is supplied to AC loads via GFM_PCS / GFL_PCS inverters, achieving decoupled operation of photovoltaic and energy storage. Figure 3 .
[0385] (3) Photovoltaic-storage DC-DC coupled power station
[0386] Trimming configuration: Trimming GFL_DCDC on the DC side; Trimming DCAC_MPPT on the AC side. Retaining DCDC_MPPT, GFM_DCDC, GFM_PCS, and GFL_PCS.
[0387] Topology: Photovoltaic modules are connected to the DC bus via DC-DC-MPPT. The energy storage system is connected to the DC bus via GFM-DCDC, achieving direct collection and buffering of DC-side photovoltaic and energy storage power. Photovoltaic power can be directly supplied to DC loads or inverted to AC power via GFM-PCS, reducing losses from multi-stage power electronic conversion and improving system efficiency. Figure 4 .
[0388] (4) DC-side photovoltaic AC-DC coupled microgrid
[0389] Trimming configuration: Trimming DCAC_MPPT on the AC side. Retaining DCDC_MPPT, GFM_DCDC, GFM_PCS, GFL_DCDC, and GFL_PCS.
[0390] Topology Characteristics: Photovoltaic modules are connected to the DC bus via DC-DC_MPPT. GFL_DCDC is introduced as DC-side grid-connected energy storage to share the DC bus voltage regulation power of GFM_DCDC. The microgrid coordinator achieves SOC balance between GFM_DCDC and GFL_DCDC through aggregation control strategies, reducing the power pressure on grid-connected equipment. On the AC side, GFM_PCS grid-connected energy storage and GFL_PCS grid-connected energy storage are retained to support SOC balance scheduling during grid-connected and off-grid operation. Figure 5 .
[0391] (5) AC-DC coupled microgrid of AC-side photovoltaic power
[0392] Trimming configuration: Trim DCDC_MPPT on the DC side. Retain DCAC_MPPT, GFM_DCDC, GFM_PCS, GFL_DCDC, and GFL_PCS.
[0393] Topology Characteristics: Photovoltaic modules are directly inverted into AC power via DCAC_MPPT and connected to the AC bus. On the DC side, GFM_DCDC is retained as grid-connected energy storage to support DC voltage. GFL_DCDC is retained as DC-side grid-connected energy storage as needed, sharing the DC bus voltage regulation power of GFM_DCDC. On the AC side, GFM_PCS grid-connected energy storage and GFL_PCS grid-connected energy storage are retained to support SOC balanced dispatch during grid-connected and off-grid operation. Figure 6 .
[0394] (6) Other
[0395] In addition to the typical scenarios mentioned above, this method can adapt to any combination of pruning scenarios because each algorithm module is decoupled based on aggregate modeling. Since each algorithm module undergoes aggregate modeling to achieve decoupling of hierarchical control parameters, the microgrid coordinator only needs to read the configuration parameters, substitute them into the formula, and automatically remove components that are zero after pruning. There is no need to redevelop the underlying code, achieving full coverage of the entire series of optical storage microgrid products with a single software version.
[0396] Example 3
[0397] This embodiment demonstrates the control effect of the method of the present invention based on the constructed test environment.
[0398] The test environment was set up with 2 DCAC_MPPTs, 1 GFM_PCS, and 1 GFL_PCS on the AC side; each PCS was configured with 2 DCCDC_MPPTs, 2 GFM_DCDCs, and 2 GFL_DCDCs on the DC side, operating in an off-grid state. The topology is as follows. Figure 7 .
[0399] I. Photovoltaic power limiting control
[0400] (1) The charging power of the AC-side PCS is too high and the photovoltaic power is limited.
[0401] With PARA_PCS_Chr_P_HIGH_Pct set to 75% and GFM_PCS_Chr_P_MAX to 200000W, the resulting GFM_PCS_Pout triggering photovoltaic power curtailment is -150000W. PARA_PV_Curt_Slope_down and PARA_PV_Curt_Slope_up are both set to 8000W / s, and DCAC_MPPT_Peak_P is 400000W. Therefore, the percentage change rate of power curtailment is 8000 / 400000*100%=2% / s. The results are as follows... Figure 8 As shown.
[0402] Around the 7th second, after the load decreased, GFM_PCS_Pout < -150000W, and the photovoltaic power limiting percentage DCAC_MPPT_Curtailment_Pct_Set began to decrease, allowing GFM_PCS_Pout to recover to around -150000W by limiting the photovoltaic power. Around the 15th and 28th seconds, the load decreased further twice, and the photovoltaic power limiting value also decreased further, maintaining GFM_PCS_Pout at around -150000W. When the load increased after 42 seconds, causing GFM_PCS_Pout > -150000W, the photovoltaic power limiting percentage DCAC_MPPT_Curtailment_Pct_Set began to rise, fully utilizing the energy generated by photovoltaic power generation.
[0403] (2) The SOC of DC-side DC-DC converter is too high, resulting in photovoltaic power limitation.
[0404] With PARA_DCDC_SOC_HIGH set to 90%, PARA_PV_Curt_Slope_down and PARA_PV_Curt_Slope_up both set to 38000W / s, DCDC_MPPT0_Peak_P set to 200000W, DCDC_MPPT0_Peak_P set to 180000W, and the aggregated DCDC_MPPT_Peak_P set to 380000W, the percentage change rate of the power limit is 38000 / 380000*100%=10% / s. The results are as follows... Figure 9 As shown.
[0405] Around the 29th second, GFM_DCDC_SOC reaches 90%, and the two DCD_MPPT maximum output power settings, DCD_MPPT0_P_MAX_Set and DCD_MPPT1_P_MAX_Set, both begin to decrease. Once GFM_DCDC_SOC is below 90%, it starts to rise again, maintaining GFM_DCDC_SOC at around 90%. Around the 59th second, as the load increases, GFM_DCDC_SOC continues to decrease, while DCD_MPPT0_P_MAX_Set and DCD_MPPT1_P_MAX_Set both rise to their maximum power values and remain unchanged.
[0406] II. SOC Equalization Control
[0407] The initial SOC of the DC-side DC-DC converter of GFM_PCS is set to 76%, 87%, 77%, and 80% respectively.
[0408] The initial SOC of the DC-side DC-DC converter of GFL_PCS is set to 66%, 77%, 67%, and 70% respectively.
[0409] All DC-DC converters have the same total battery capacity, therefore the initial GFM_PCS_SOC=80% and GFL_PCS_SOC=70%. The following shows the combined SOC balancing of the AC side GFM_PCS and GFL_PCS under SOC balancing control, and the discrete SOC balancing of the DC side DC-DC converters under GFM_PCS. The results are as follows... Figure 10 , Figure 11 As shown.
[0410] (1) SOC equalization of the PCS on the AC side
[0411] Initially, GFM_PCS_SOC and GFL_PCS_SOC differ by about 10%, but after SOC equalization, they gradually become equal during discharge.
[0412] (2) SOC equalization of DC-side DC-DC converters
[0413] The initial states GFM_DCDC0_SOC, GFM_DCDC1_SOC, GFL_DCDC0_SOC, and GFL_DCDC1_SOC are all different. After SOC equalization, the SOC of each device gradually becomes consistent during the discharge period.
Claims
1. A control method for a photovoltaic-storage AC / DC microgrid based on a configurable group of controllable objects, characterized in that, The method includes: A standardized virtual topology model of a photovoltaic-storage AC / DC microgrid is established in the microgrid coordinator controller. Typical physical devices in the microgrid are mapped to controllable object groups in the topology model according to device type. The controllable object groups include photovoltaic control device object groups on the AC side and DC side, grid-type energy storage device object groups, and grid-connected energy storage device object groups. The microgrid coordinator collects real-time operating parameters of the physical single devices corresponding to each controllable object group through a heterogeneous communication network for discrete information modeling, and obtains the aggregated state variables of each controllable object group based on the discrete information. The microgrid coordinator executes a hierarchical coordination control strategy based on the aggregated state variables, including: an aggregated control layer: calculating group-level control instructions for each controllable object group based on the aggregated state variables and preset control objectives, including power allocation instructions or voltage / frequency reference instructions; and an allocation execution layer: calculating control instructions for each device based on the group-level control instructions and device operating status data, and issuing them to the corresponding devices for execution. Each of the controllable object groups is configured with participation parameters. When the photovoltaic-storage AC / DC microgrid is missing physical equipment corresponding to a certain type of controllable object group, the participation parameters corresponding to the controllable object group with missing physical equipment are set to default invalid values, so that the corresponding controllable object group is automatically removed in the calculation of aggregated state variables, thereby realizing the logical pruning of the controllable object group.
2. The method according to claim 1, characterized in that, The acquisition of aggregated state variables for each controllable object group includes: For a group of photovoltaic control equipment, the peak power, current output power, and maximum allowable output power of each individual device in the controllable group are collected; and aggregated into the total peak power, total output power, and power limit percentage of the corresponding controllable group. For DC-side grid-type energy storage equipment groups, the charging and discharging power limits, current output power, output voltage, current charge level, and total battery capacity of each individual device in the controllable group are collected and aggregated into the total charging and discharging power limits, total output power, output voltage, SOC, and total capacity of the corresponding controllable group. For AC-side grid-type energy storage equipment groups, the output frequency is additionally collected and aggregated into the output frequency of the corresponding controllable group. For grid-connected energy storage device groups, the charging and discharging power limits, current output power, current charge level, and total battery capacity of each individual device in the controllable group are collected and aggregated into the total charging and discharging power limits, total output power, SOC, and total capacity of the corresponding controllable group.
3. The method according to claim 1 or 2, characterized in that, For the photovoltaic control equipment group, the participating configuration parameter is the peak power of each individual device; For the grid-connected energy storage device group, the participating configuration parameters are the total battery capacity, maximum discharge power, and maximum charging power of each individual device; For the grid-type energy storage device group, the participating configuration parameters are the total battery capacity, maximum discharge power, and maximum charging power of each individual device; The default value of the configuration parameters is set to 0.
4. The method according to claim 1, characterized in that, The aggregation control layer performs DC-side control decisions, including: Monitor the aggregation state variables of the DC-side grid-type energy storage device group. When any of the following conditions are met: aggregation SOC is higher than the first preset SOC threshold, aggregation charging power is higher than the first preset charging power threshold, or output voltage is higher than the preset voltage threshold, issue a power limiting command to the DC-side photovoltaic control device group, reduce the power limiting percentage of the photovoltaic control device group by a preset decreasing slope, and restore the maximum power point tracking control mode when the triggering condition is eliminated, and increase the power limiting percentage by a preset increasing slope until the maximum power point is reached. A voltage reference value is issued to the DC-side grid-type energy storage device group, which automatically maintains output voltage stability and completes power balance based on the voltage reference value; Based on the DC bus power balance relationship, and combining the aggregated SOC of the DC-side grid-type energy storage device group and the grid-connected energy storage device group, the total power setting value of the DC-side grid-connected energy storage device group is calculated.
5. The method according to claim 1, characterized in that, The aggregation control layer performs power limiting decisions for the AC-side photovoltaic control device object group, including: When the microgrid is in grid-connected operation mode, the aggregate charging power, aggregate SOC, and grid-connected power of the AC-side energy storage device group are monitored. If any of the following conditions are met: the grid-connected SOC is higher than the second preset SOC threshold, the grid-connected charging power is higher than the second preset charging power threshold, or the active power at the grid-connected point is less than the anti-reverse current reserve value, the power limiting mode of the AC-side photovoltaic control device group is triggered, and the power limiting percentage is reduced by a preset decreasing slope. When the triggering condition is eliminated, the maximum power point tracking control mode is restored, and the power limiting percentage is increased by a preset increasing slope until the maximum power point is reached. The grid-connected SOC is calculated using a capacity-weighted average method based on the aggregate SOC and total aggregate battery capacity of each AC-side energy storage device group. The grid-connected charging power is the sum of the aggregate charging power of each AC-side energy storage device group. When the microgrid is in off-grid operation mode, the aggregate SOC, aggregate charging power, and output frequency of the AC-side grid-type energy storage device group are monitored. When any of the following conditions are met, such as aggregate SOC being higher than the second preset SOC threshold, aggregate charging power being higher than the third preset charging power threshold, or output frequency being higher than the preset frequency threshold, the power limiting mode of the AC-side photovoltaic control device group is triggered, and the power limiting percentage is reduced by a preset slope.
6. The method according to claim 1, characterized in that, The aggregation control layer performs power allocation decisions for the AC-side energy storage device object group, including: When the microgrid is in grid-connected operation mode, all AC-side energy storage devices operate in current source mode. Based on the aggregated SOC and total battery capacity of the AC-side energy storage device object group, and in accordance with the SOC balancing principle, the aggregated active power command of the entire station is distributed to each AC-side energy storage device object group to obtain the active power setting value of each group. Based on the maximum reactive power ratio of the AC-side energy storage device object group, the total station aggregated reactive power command is distributed to each AC-side energy storage device object group to obtain the reactive power setting value of each group; the total station aggregated active / reactive power command is the sum of the active / reactive power output power setting values of each AC-side energy storage device object group. When the microgrid is in off-grid operation mode, the grid-type energy storage device group operates as a voltage source; The reference frequency setting value and reference voltage setting value are issued to the grid-type energy storage device object group; Based on the aggregated SOC and total battery capacity of the AC-side energy storage device group, the active power setting value of the grid-connected energy storage device group is calculated according to the SOC balancing principle. Based on the maximum reactive power ratio of each AC-side energy storage device group under the current active power control conditions, the reactive power setting value of the grid-connected energy storage device group is calculated.
7. The method according to claim 1, characterized in that, The allocation execution layer performs DC-side allocation decisions, including: For the photovoltaic control equipment object group, based on the power limit percentage determined by the aggregated control layer, the maximum power limit setting value of each individual device is calculated and issued; For the grid-type energy storage device group, the voltage offset is calculated based on the deviation between the SOC of each individual device and the average SOC, combined with a preset proportional coefficient; the voltage offset is superimposed on the reference voltage to generate the voltage setting value of each individual device and then distributed. For grid-connected energy storage equipment groups, the total power setting value is allocated to each individual device proportionally based on the SOC and total battery capacity of each individual device, and the charging and discharging power is limited.
8. The method according to claim 1, characterized in that, The allocation execution layer performs allocation decisions on the communication side, including: For the group of photovoltaic control equipment on the AC side, based on the power limit percentage determined by the aggregated control layer and combined with the peak power of each individual device, the maximum power limit setting value of each individual device is calculated and issued. For AC-connected grid-type energy storage equipment groups, in grid-connected mode, based on the group-level active power setting value, according to the SOC of each individual device and the total battery capacity, active power is evenly distributed based on SOC to generate active power instructions for each individual device; at the same time, reactive power instructions for each individual device are generated according to the maximum reactive power ratio under the current active power control conditions. In off-grid mode, based on the deviation between the SOC of each individual device and the average SOC, and combined with a preset proportional coefficient, the frequency offset is calculated; the frequency offset is superimposed on the reference frequency to generate the frequency setting value for each individual device and send it down. At the same time, reactive power is allocated according to the proportion of the maximum reactive power that each individual device can generate, and the voltage setting value corresponding to each individual device is calculated and issued according to the allocated reactive power using the droop formula. For AC-side grid-connected energy storage device groups, based on the group-level active power setting value, the SOC of each individual device and the total battery capacity, active power is allocated according to SOC balancing, and active power commands for each individual device are generated and issued. At the same time, based on the group-level reactive power setting value, reactive power is allocated according to the proportion of the maximum reactive power that each individual device can issue, and reactive power commands for each individual device are generated and issued.
9. The method according to claim 1, characterized in that, The heterogeneous communication network includes: The first communication network used to acquire electrical quantities and protection signals of equipment, and to issue emergency power cut-off commands or fast trip commands; A second communication network used to acquire basic electrical parameters and status variables of the equipment, as well as to issue charging and discharging power settings or start / stop commands; A third communication network used to acquire equipment operating data and measurement values, as well as to issue voltage reference values, frequency reference values, or power limit percentage settings; The first, second, and third communication networks all use standard industrial protocols.
10. A microgrid coordinator controller, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1-9.
