Alternating current and direct current hybrid microgrid distributed cooperative control system and method based on multiple controllers

Abstract

The invention discloses an AC / DC hybrid microgrid distributed cooperative control system and method based on multiple controllers, and belongs to the technical field of power system control. The system comprises a device layer, a control layer and a network communication layer. And the control layer is provided with a plurality of equipment-level controllers, so that the control layer can bypass an upper-layer energy management system when detecting a system power fluctuation or fault event, directly exchanges local operation data through a transverse communication link, and autonomously negotiates and issues a control instruction to a local power converter based on a preset distributed cooperative strategy. The method correspondingly comprises the steps of data perception, distributed collaborative decision making, local execution and the like. Through event triggering and peer-to-peer communication mechanisms, millisecond-level fast power balance and stable control of the system are achieved, the problems that centralized control is slow in response, single-point failure risks exist and heterogeneous equipment collaboration is difficult are effectively solved, and the reliability, the dynamic response speed and the control flexibility of the system are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of power system control, and particularly relates to a multi-controller-based distributed collaborative control system and method for AC / DC hybrid microgrids. BACKGROUND

[0002] As a power supply system capable of efficiently accommodating distributed renewable energy such as wind power and photovoltaic power, the AC / DC hybrid microgrid has important application value in modern power systems. By organically integrating the advantages of AC and DC power supply forms, the system effectively improves the energy utilization efficiency and power supply reliability, and is particularly suitable for local energy interconnection scenarios such as campuses, industrial parks and remote areas, helping to achieve on-site consumption and optimal allocation of energy. The existing system usually adopts a centralized energy management system (EMS) for global optimization scheduling, and realizes economic optimization in steady-state operation. At the same time, the fixed grid-following or grid-forming control mode is generally adopted for various converters in the system to maintain the basic stability of voltage and frequency, with the characteristics of clear structure and mature control logic.

[0003] However, the existing control system still has significant deficiencies in practical application. On the one hand, the optimization calculation and instruction issuing period of the centralized energy management system is relatively long, usually on the order of seconds or even minutes, which makes it difficult to respond to rapid dynamic processes such as power shortage and load mutation, and there is obvious control lag. On the other hand, this architecture has a single point of failure risk. Once the central controller fails, the system as a whole is likely to lose control, seriously affecting the operation reliability. In addition, the dynamic response characteristics of photovoltaic, energy storage and wind power devices connected to the system are significantly different, and there is a lack of effective and rapid coordination mechanism between their local controllers, making it difficult to achieve power balance and dynamic support in multiple time scales, which limits the robustness of the system in transient processes. The above problems restrict the further promotion and advanced application of AC / DC hybrid microgrids. SUMMARY

[0004] To solve the above technical problems, the present application proposes a multi-controller-based distributed collaborative control system and method for AC / DC hybrid microgrids to solve the problems existing in the prior art.

[0005] To achieve the above purpose, in a first aspect, the present application provides a multi-controller-based distributed collaborative control system for AC / DC hybrid microgrids, comprising a device layer, a control layer and a network communication layer. The device layer includes wind-solar-storage DC microgrid subsystem devices and AC microgrid subsystem devices. The control layer includes an upper energy management system and multiple device-level controllers. The network communication layer includes a communication network. The multiple device-level controllers are configured to: when detecting system power fluctuations or fault events, bypass the upper-level energy management system, directly exchange local operating data through the horizontal communication link established by the communication network, and autonomously negotiate and issue control commands to the local power converter based on a preset distributed cooperative strategy, so as to achieve rapid power balance and stable control at the system level.

[0006] Preferably, the system power fluctuation or fault event includes DC bus voltage deviation exceeding a preset threshold.

[0007] Preferably, the communication network includes a CAN bus and an industrial Ethernet, and the lateral communication link is a point-to-point communication based on TCP / IP or CAN protocol.

[0008] Preferably, the device-level controller uses DSP+FPGA as the core of the control hardware.

[0009] Preferably, the pre-defined distributed coordination strategy includes: When the device-level controller detects that the DC bus voltage deviation exceeds a preset threshold, it is determined to be a power imbalance event; The controller that detects the event acts as the master coordinator and broadcasts the power demand signal to other controllers via the lateral communication link; After receiving the signal, the battery energy storage unit controller and the supercapacitor energy storage unit controller calculate their respective power compensation amounts in a distributed manner based on their own state and power capabilities.

[0010] Prior to this, the pre-set distributed coordination strategy also includes: when a power deficit event occurs, the supercapacitor energy storage unit controller prioritizes power compensation, and the battery energy storage unit controller provides subsequent power support based on its own SOC state.

[0011] Preferably, the device-level controller includes a third device-level controller configured to control the grid-connected converter to operate in either grid-following mode or grid-connected mode; in grid-connected mode, a virtual synchronous machine control algorithm is used to autonomously establish and stabilize the voltage and frequency of the AC bus.

[0012] Preferably, the virtual synchronous machine control algorithm is based on the virtual synchronous machine rotor motion equation, which includes: In the formula, For virtual mechanical power, Electromagnetic power, For virtual electrical angular velocity, For the preset angular velocity, Here, θ is the VSG virtual damping coefficient, and θ is the rotor angle. Let t be the virtual moment of inertia, and t be time.

[0013] Preferably, the upper-level energy management system employs a central controller based on an ARM Cortex-A8 processor.

[0014] Secondly, the present invention also provides a distributed cooperative control method for AC / DC hybrid microgrids based on multiple controllers, for implementing the system described in the first aspect, the method comprising the following steps: During system initialization, each device-level controller powers on, completes self-test, establishes communication connections with the upper-level energy management system and with each other, and initializes local control parameters. Each device-level controller collects real-time voltage, current, power, and SOC status information of the units under its jurisdiction; Each device-level controller can determine whether a local event requiring collaborative processing has occurred based on the collected data. If an event occurs, any device-level controller will act as the initiator and send a coordination request to other relevant controllers directly via point-to-point communication, including the event type, power deficit, and timestamp information. Based on the received information and preset rules, the relevant controllers independently calculate the control instructions they should undertake and immediately execute the local control actions. Each device-level controller reports its action results to the upper-level energy management system; the upper-level energy management system performs global optimization calculations based on the new system state and issues new operating setpoints.

[0015] Compared with the prior art, the present invention has the following advantages and technical effects: This invention utilizes the technical feature that allows the device-level controller to bypass the upper-level energy management system and directly exchange local operating data and autonomously negotiate via a horizontal communication link when detecting system power fluctuations or fault events. This enables the system to immediately initiate distributed collaborative decision-making upon event triggering, avoiding the second- or minute-level delays associated with centralized optimization scheduling. This feature reduces the system's response time to transient processes such as power deficits and voltage surges to the millisecond level, effectively solving the power imbalance and stability problems caused by control lag in existing technologies.

[0016] This invention utilizes the technical feature of "multiple device-level controllers autonomously negotiating and issuing control commands through a horizontal communication link." This eliminates the system's dependence on a single central controller, and the distributed architecture ensures that even in the event of a local controller or communication failure, other controllers can still maintain system operation through horizontal cooperation. This feature effectively overcomes the single-point-of-failure risk of existing centralized architectures, enhancing the system's survivability and overall reliability under fault conditions.

[0017] This invention, through its technical feature of "autonomous negotiation based on a pre-set distributed collaborative strategy," enables devices with vastly different dynamic characteristics, such as photovoltaics, battery energy storage, supercapacitor energy storage, and wind power, to calculate and execute power compensation commands in real time based on their local conditions (e.g., SOC, power capacity). This feature, through an event-driven peer-to-peer communication mechanism, achieves refined power allocation across multiple time scales, solving the problems of system oscillation and insufficient power support caused by large differences in device response characteristics and a lack of rapid coordination in existing technologies. Attached Figure Description

[0018] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a diagram illustrating the overall architecture of the distributed collaborative control system according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the system control structure according to an embodiment of the present invention; Figure 3 This is a circuit topology diagram of the distributed collaborative control system according to an embodiment of the present invention; Figure 4 This is the virtual synchronous machine control algorithm for grid-connected inverters in an embodiment of the present invention. Detailed Implementation

[0019] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0021] Example 1 This embodiment provides a distributed collaborative control system for AC / DC hybrid microgrids based on multiple controllers, aiming to achieve rapid power balance, fault alarm, and economically optimized operation under complex operating conditions. The system adopts a "centralized coordination, distributed execution" architecture, integrating upper-layer slow-time-scale scheduling with lower-layer millisecond-level control, possessing good reliability, scalability, and flexibility. The system specifically includes a device layer, a control layer, and a network communication layer. The equipment layer includes wind-solar-storage DC microgrid subsystem equipment and AC microgrid subsystem equipment; Specifically, the equipment layer encompasses photovoltaic (PV) power generation units, battery energy storage units, supercapacitor energy storage units, direct-drive wind power units, and AC microgrid interface units. The PV power generation units employ interleaved parallel Boost converters based on SiC devices, effectively improving conversion efficiency and reducing switching losses, and achieving MPPT control through incremental conductance. The battery energy storage units are equipped with bidirectional Buck-Boost converters, supporting constant current / constant voltage charging and droop control, and can automatically adjust charging and discharging power according to the DC bus voltage. The supercapacitor units use a similar topology, specifically designed to suppress power surges caused by sudden load changes and renewable energy fluctuations. The direct-drive wind power units include permanent magnet synchronous generators and full-power back-to-back converters, achieving maximum wind energy tracking and grid-connected control. The AC microgrid interface units include two-level and three-level ANPC converters, capable of operating in VF control or PQ control modes.

[0022] The control layer includes an upper-level energy management system and multiple device-level controllers; Furthermore, the upper-level energy management system employs a central controller based on an ARM Cortex-A8 processor.

[0023] Furthermore, the device-level controller uses DSP+FPGA as the core of the control hardware.

[0024] Specifically, the control layer consists of distributed device-level local controllers and an upper-level coordination controller. The local controllers employ a high-performance DSP+FPGA architecture, enabling the implementation of various complex control algorithms, including proportional-resonant control, sliding-mode variable-structure control, and model predictive control. These controllers acquire unit operating data in real time and, upon detecting events such as power fluctuations, voltage or frequency exceeding limits, conduct distributed negotiations with other controllers through a horizontal communication network to quickly reach collaborative control decisions, such as power allocation. The upper-level coordination controller uses a multi-core ARM processor and runs a real-time operating system, enabling system-level monitoring, economical operation strategy generation, and scheduling command issuance. The coordination controller communicates with the local controllers via a communication protocol, without interfering with the millisecond-level real-time control of the underlying devices.

[0025] The network communication layer includes a communication network; Furthermore, the communication network includes a CAN bus and an industrial Ethernet, and the lateral communication link is a point-to-point communication based on TCP / IP or CAN protocol.

[0026] Specifically, the network communication layer employs a communication network comprised of a CAN bus and a gigabit industrial Ethernet. The CAN bus uses the CANopen protocol to ensure real-time transmission of critical control commands, with a communication cycle of up to 1ms. The industrial Ethernet supports the IEEE 1588 precision time protocol, achieving clock synchronization accuracy better than 1μs between controllers. The network architecture supports vertical communication between device-level controllers and the coordinating controller, as well as horizontal peer-to-peer (P2P) communication between device-level controllers. Horizontal communication uses a publish-subscribe mechanism, supporting rapid information broadcasting and negotiation triggered by events, significantly improving the system's responsiveness to transient events.

[0027] The multiple device-level controllers are configured to: when detecting system power fluctuations or fault events, bypass the upper-level energy management system, directly exchange local operating data through the horizontal communication link established by the communication network, and autonomously negotiate and issue control commands to the local power converter based on a preset distributed cooperative strategy, so as to achieve rapid power balance and stable control at the system level.

[0028] Furthermore, the system power fluctuation or fault event includes DC bus voltage deviation exceeding a preset threshold.

[0029] Furthermore, the pre-defined distributed collaboration strategy includes: When the device-level controller detects that the DC bus voltage deviation exceeds a preset threshold, it is determined to be a power imbalance event; The controller that detects the event acts as the master coordinator and broadcasts the power demand signal to other controllers via the lateral communication link; After receiving the signal, the battery energy storage unit controller and the supercapacitor energy storage unit controller calculate their respective power compensation amounts in a distributed manner based on their own state and power capabilities.

[0030] Furthermore, the pre-set distributed coordination strategy also includes: when a power deficit event occurs, the supercapacitor energy storage unit controller prioritizes power compensation, and the battery energy storage unit controller provides subsequent power support based on its own SOC state.

[0031] Furthermore, the device-level controller includes a third device-level controller, which is configured to control the grid-connected converter to operate in grid-following mode or grid-connected mode; in grid-connected mode, a virtual synchronous machine control algorithm is used to autonomously establish and stabilize the voltage and frequency of the AC bus.

[0032] Figure 1 This is the overall architecture diagram of the distributed collaborative control system in this embodiment, as shown below. Figure 1The control layer, as shown, makes global or local control strategy decisions through the EMS (Energy Management System) and local controllers; the communication layer uses technologies such as CAN bus to realize data interaction between the transmission of control commands and the transmission of device status; the device layer consists of converters, power generation equipment, and energy storage units, and is responsible for performing physical actions such as power conversion and energy storage. These three layers form a command transmission link of "control → communication → device" and a status feedback link of "device → communication → control," together constituting an interactive closed loop.

[0033] Figure 2 A schematic diagram of the system control structure, such as Figure 2 The architecture shown achieves modular expansion through layered decoupling, combining scalability and real-time response capabilities. It adapts to the dynamic control requirements of power systems and is widely used in scenarios such as energy storage clusters and distributed energy systems.

[0034] Figure 3 The circuit topology diagram of a distributed collaborative control system is shown below. Figure 3 As shown, the wind-solar-storage DC microgrid subsystem includes a photovoltaic power generation unit, a battery energy storage unit, a supercapacitor energy storage unit, and a direct-drive wind power unit. The photovoltaic unit uses a photovoltaic simulator as input, connected to the DC bus via an interleaved parallel Boost converter based on SiC MOSFE, and is controlled by a first-level device controller running an MPPT algorithm (such as the perturbation-observation method) and voltage loop PID control. The battery unit's battery pack, connected via a SiC bidirectional Buck-Boost converter, has a rated power of 20kW. The controller employs a power loop control strategy, with charging and discharging switching triggered by changes in the DC bus voltage. The supercapacitor unit has a total capacity of 166F and a rated voltage of 270V, connected via another SiC bidirectional Buck-Boost converter (5kW), primarily used for high-frequency power compensation, with a controller response time of less than 5ms. The direct-drive wind power unit consists of a 5.5kW permanent magnet synchronous generator and an ACDC full-power converter. A second-level device controller implements generator-side FOC control and grid-side PQ control, and can adjust the generator speed via a communication interface to simulate wind speed changes.

[0035] The AC microgrid subsystem comprises a two-level grid-connected inverter and an ANPC three-level grid-connected converter. A third-level device controller is responsible for its control, supporting both grid-following mode (synchronizing with grid voltage and frequency) and grid-connected mode (using the Virtual Synchronous Generator-VSG algorithm). In grid-connected mode, the controller autonomously establishes the AC bus voltage and frequency by adjusting virtual inertia and damping coefficients, suitable for islanded operation. The controller receives drive signals via fiber optic cable and employs sensor-isolated sampling to ensure control accuracy and safety.

[0036] All device-level controllers adopt a DSP+FPGA architecture, providing up to 36 PWM outputs and 32 ADC channels, supporting Simulink automatic code generation and C language programming. The controllers connect to the communication network via onboard CAN interface and Ethernet port, and can use host computer software for online parameter adjustment, real-time waveform observation, and data recording.

[0037] The system's communication network employs a hybrid CAN bus and industrial Ethernet network. Device-level controllers exchange data directly via horizontal communication links, while transmitting commands and status information with the upper-level coordinating controller via vertical links. The network communication cycle is less than 1ms, ensuring real-time event response and coordinated control.

[0038] The upper-level energy management system hardware adopts a central controller based on an ARM Cortex-A8 processor, equipped with 16 serial ports, 4 Ethernet ports, and 2 CAN ports, running an embedded Linux system and self-developed monitoring software. This system realizes functions such as data acquisition, operational status display, and historical data storage.

[0039] To verify the effectiveness of this embodiment, the hardware configuration of the AC / DC hybrid microgrid and network system scheme based on the "device layer-control layer-network communication layer" architecture and the multi-controller-based AC / DC hybrid microgrid distributed collaborative control system and method was used. The experimental platform was built based on the system circuit topology. The device layer included: a 5kW rated power photovoltaic simulator, a 192V / 50Ah lithium battery energy storage unit (with real-time SOC detection function), a 270V / 166F supercapacitor energy storage unit, a 5.5kW direct-drive wind power simulator, a two-level DCAC grid-connected converter, and an ANPC three-level DCAC converter. The AC side adopted a combined inverter circuit to connect the DC and AC buses. The AC voltage was set to 380V, and the DC bus voltage was set to 600V. The bidirectional flow of AC and DC power was achieved through a three-phase bridge topology. The DC side adopted a half-bridge circuit, which could flexibly switch between Boost, Buck, and bidirectional Buck-Boost modes according to power supply and load requirements.

[0040] The control layer is configured with two device-level controllers (associated with "photovoltaic + lithium battery" and "supercapacitor + direct-drive wind power" respectively), a fast controller (associated with grid-connected converter), and an upper-level energy management system (EMS). The decision and control unit realizes distributed collaborative decision-making through a microprocessor. The network communication layer adopts a CAN bus and industrial Ethernet to form a communication network. The sampling and communication unit includes voltage / current sensors, signal conditioning circuits, and internal and external communication ports to ensure real-time transmission of operating data of each unit.

[0041] Before the experiment started, the parameters of each unit of the system were calibrated and the status was adjusted: the output of the photovoltaic simulator was set to 4kW, the AC load power was adjusted to 5kW, the state of charge (SOC) of the lithium battery energy storage unit was precharged to 80% and maintained in floating charge state, the supercapacitor energy storage unit was precharged to 80% and maintained in floating charge state, and the DC bus reference voltage was set to 600V; the grid-connected converter initially operated in grid-following mode, and grid-connected operation was achieved by tracking the external grid voltage and frequency, and the PCC (point of common coupling) grid-connected switch was kept closed; the upper-level EMS established normal communication with each device-level controller, and collected operating data such as photovoltaic output, energy storage SOC, and converter output power in real time to ensure that the system had no alarms and no faults, and met the "stable operating condition" requirements described in this embodiment.

[0042] After the system has been running stably for 30 seconds, at t=5 seconds, the upper-level EMS sends an "off-grid switching command" to the PCC grid-connected switch and grid-connected converter: First, it controls the PCC grid-connected switch to open, cutting off the electrical connection with the external power grid. Then, it commands the grid-connected converter to switch from grid-connected mode to grid-connected mode, activating the built-in VSG control algorithm. Figure 4 For grid-connected inverter virtual synchronous machine control algorithms, such as Figure 4 The voltage and frequency of the AC bus are autonomously established and stabilized by the converter based on the VSG rotor motion equations. The VSG rotor motion equations are as follows: In the formula, For virtual mechanical power, Electromagnetic power, For virtual electrical angular velocity, For the preset angular velocity, Here, θ is the VSG virtual damping coefficient, and θ is the rotor angle. Let t be the virtual moment of inertia, and t be time.

[0043] Within 10 seconds of mode switching, key data acquisition is completed through the sampling and communication unit described in this embodiment: the AC side current sensor is used to collect the AC side output current waveform of the grid-connected converter in real time, the current change curve at the moment of switching is recorded, and the current change is analyzed; the AC bus frequency data uploaded by the grid-connected converter controller is read, and the frequency fluctuation range within 10 seconds after switching is analyzed in combination with the VSG excitation regulation equation. The VSG excitation regulation equation is as follows: In the formula, and These are the reactive voltage droop coefficient and the reactive voltage sag adjustment coefficient, respectively. and These are the voltage setting value and the actual effective voltage value, respectively. and These are the reference value and the actual measured value of reactive power, respectively. The no-load electromotive force of the VSG is The VSG is given an electromotive force; the output power changes of the supercapacitor and lithium battery are recorded in real time through the controllers associated with the "supercapacitor + direct-drive wind power" and the "photovoltaic + lithium battery" to determine the compensation amount and response delay time of the two for the power deficit after off-grid.

[0044] The beneficial effects of this embodiment: Compared to traditional centralized architectures, this approach employs device-level local controllers and high-speed communication networks, eliminating single points of failure. Its modular design supports plug-and-play expansion, and global economic optimization is achieved through an upper-level coordinator, significantly improving system reliability, dynamic performance, and operational economy. This embodiment enables stable, economical, and controllable AC / DC hybrid microgrids, possessing a clear structure, well-defined control strategies, and feasible communication mechanisms, making it suitable for practical engineering applications.

[0045] Example 2 This embodiment provides a distributed cooperative control method for AC / DC hybrid microgrids based on multiple controllers, used to implement the system described in Embodiment 1. The method includes the following steps: During system initialization, each device-level controller powers on, completes self-test, establishes communication connections with the upper-level energy management system and with each other, and initializes local control parameters. Each device-level controller collects real-time voltage, current, power, and SOC status information of the units under its jurisdiction; Each device-level controller can determine whether a local event requiring collaborative processing has occurred based on the collected data. If an event occurs, any device-level controller will act as the initiator and send a coordination request directly to other relevant controllers via point-to-point communication, including the event type, power deficit, and timestamp information; the receiving controllers will reply with their available capacity and response priority information. Based on the received information and preset rules, the relevant controllers independently calculate the control instructions they should undertake and immediately execute the local control actions. Each device-level controller reports its action results to the upper-level energy management system; the upper-level energy management system performs global optimization calculations based on the new system state and issues new operating setpoints.

[0046] Specifically, during system initialization, each controller loads parameter configuration files, establishes communication connections, and completes self-tests. During operation, local controllers collect electrical quantities in real time. When a DC voltage deviation or frequency deviation is detected, a distributed event is triggered. The event initiating controller transmits event information via the CAN bus. The receiving controller calculates the power adjustment amount based on a consensus algorithm or weighted allocation strategy and completes the control decision within 50ms. Each controller independently generates a PWM signal and performs power compensation. Finally, the execution results are uploaded to the coordinating controller, which performs scheduling optimization based on the updated system state to achieve long-term stable and economical system operation.

[0047] This embodiment effectively solves the problems of centralized control response delay and single point of failure by adopting a hierarchical distributed control architecture and a fast horizontal communication mechanism, significantly improving the system's reliability, flexibility, and operating efficiency. Experimental results show that the system can restore DC voltage stability within 100ms, with frequency deviation controlled within ±0.2Hz.

[0048] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A distributed collaborative control system for AC / DC hybrid microgrids based on multiple controllers, characterized in that, It includes the device layer, control layer, and network communication layer; The equipment layer includes wind-solar-storage DC microgrid subsystem equipment and AC microgrid subsystem equipment; The control layer includes an upper-level energy management system and multiple device-level controllers; The network communication layer includes a communication network; The multiple device-level controllers are configured to: when detecting system power fluctuations or fault events, bypass the upper-level energy management system, directly exchange local operating data through the horizontal communication link established by the communication network, and autonomously negotiate and issue control commands to the local power converter based on a preset distributed cooperative strategy, so as to achieve rapid power balance and stable control at the system level.

2. The system according to claim 1, characterized in that, The system power fluctuations or fault events include DC bus voltage deviations exceeding a preset threshold.

3. The system according to claim 1, characterized in that, The communication network includes a CAN bus and an industrial Ethernet, and the horizontal communication link is a point-to-point communication based on TCP / IP or CAN protocol.

4. The system according to claim 1, characterized in that, The device-level controller uses DSP+FPGA as its core control hardware.

5. The system according to claim 1, characterized in that, The pre-defined distributed collaboration strategies include: When the device-level controller detects that the DC bus voltage deviation exceeds a preset threshold, it is determined to be a power imbalance event; The controller that detects the event acts as the master coordinator and broadcasts the power demand signal to other controllers via the lateral communication link; After receiving the signal, the battery energy storage unit controller and the supercapacitor energy storage unit controller calculate their respective power compensation amounts in a distributed manner based on their own state and power capabilities.

6. The system according to claim 5, characterized in that, The pre-defined distributed collaborative strategy also includes: when a power deficit event occurs, the supercapacitor energy storage unit controller prioritizes power compensation, and the battery energy storage unit controller provides subsequent power support based on its own SOC state.

7. The system according to claim 1, characterized in that, The device-level controller includes a third device-level controller, which is configured to control the grid-connected converter to operate in grid-following mode or grid-connected mode; in grid-connected mode, a virtual synchronous machine control algorithm is used to autonomously establish and stabilize the voltage and frequency of the AC bus.

8. The system according to claim 7, characterized in that, The virtual synchronous machine control algorithm is based on the virtual synchronous machine rotor motion equation, which includes: In the formula, For virtual mechanical power, Electromagnetic power, For virtual electrical angular velocity, For the preset angular velocity, Here, θ is the VSG virtual damping coefficient, and θ is the rotor angle. Let t be the virtual moment of inertia, and t be time.

9. The system according to claim 1, characterized in that, The upper-level energy management system uses a central controller based on an ARM Cortex-A8 processor.

10. A distributed cooperative control method for AC / DC hybrid microgrids based on multiple controllers, characterized in that, A method for implementing the system according to any one of claims 1-9 includes the following steps: During system initialization, each device-level controller powers on, completes self-test, establishes communication connections with the upper-level energy management system and with each other, and initializes local control parameters. Each device-level controller collects real-time voltage, current, power, and SOC status information of the units under its jurisdiction; Each device-level controller can determine whether a local event requiring collaborative processing has occurred based on the collected data. If an event occurs, any device-level controller will act as the initiator and send a coordination request to other relevant controllers directly via point-to-point communication, including the event type, power deficit, and timestamp information. Based on the received information and preset rules, the relevant controllers independently calculate the control instructions they should undertake and immediately execute the local control actions. Each device-level controller reports its action results to the upper-level energy management system; the upper-level energy management system performs global optimization calculations based on the new system state and issues new operating setpoints.

Images