Energy storage black start system and control method thereof

Abstract

The present application relates to the technical field of energy storage, and particularly relates to an energy storage black start system and a control method thereof, the system comprising a physical layer, a coordination control layer and a network communication layer; the physical layer comprises at least one black start power module, at least one load control module and at least one grid synchronization module, the coordination control layer comprises a centralized system management unit and a local autonomous controller, and the network communication layer is a star-shaped heterogeneous redundant network based on a time-sensitive network, and is used for realizing data interaction between the coordination control layer and the physical layer. The present application avoids the whole black start failure caused by single point failure and the start interruption caused by the mismatch between the preset plan and the actual situation, improves the success rate of black start, guarantees continuous high-quality power supply to sensitive loads, enhances the robustness under incomplete or delayed communication, ensures the completion of cross-layer instruction transmission and state synchronization within milliseconds, and prolongs the service life of equipment.

Description

Technical Field

[0001] This invention relates to the field of energy storage technology, specifically to an energy storage black start system and its control method. Background Technology

[0002] The safe and stable operation of the power grid is of paramount importance. However, in extreme situations such as extreme natural disasters, severe equipment failures, cyberattacks, or cascading failures, the power grid may experience widespread blackouts or even complete grid collapse, entering a state of total blackout. In this state, the entire power system loses voltage support and frequency reference, and conventional generating units cannot start on their own due to the loss of power supply. Therefore, how to quickly, reliably, and orderly restore power supply from a state of total blackout—that is, to complete a black start—is the last line of defense in the power system's security defense system and a key indicator for measuring the resilience and recovery capability of the power grid.

[0003] Traditional black start technology primarily relies on specific types of generator sets with self-starting capabilities, such as hydroelectric generator sets (especially pumped storage units), gas turbines, and some diesel generator sets. These units typically serve as starting power sources, initially restoring power to an isolated island grid, providing auxiliary power to other large thermal or nuclear power units. Once these units are up and connected to the grid, the restoration scope is gradually expanded, ultimately achieving the reconstruction of the entire main grid and restoring power to all loads. Although this model has been widely used in practice, it still has some problems: (1) Limited start-up speed and flexibility: The start-up and load-bearing process of traditional black start power sources (such as hydropower units) is relatively slow, and their availability is highly dependent on external conditions such as reservoir water level and natural gas supply, making it impossible to respond quickly at any time. At the same time, their geographical location is fixed, making it difficult to flexibly respond to the recovery needs of different areas, especially load centers far away from hydropower stations.

[0004] (2) Large load connection impact: In the early stage of recovery, the island power grid with no load or light load has small capacity and weak inertia. When a large capacity load (especially impact loads such as motors) is connected, it is very easy to cause drastic fluctuations in the grid frequency and voltage. In severe cases, it may cause the already recovered "island" power grid to become unstable again, causing repeated recovery processes or even interruptions. Existing solutions usually adopt a conservative load connection strategy of step-by-step and trial-and-error, which seriously affects the recovery speed.

[0005] The grid connection process is complex: After the local power grid is restored to a certain scale, it needs to be synchronously connected with the restored main grid or other islands that are in the process of restoration. Due to possible deviations in frequency, phase, and voltage between the two systems, traditional grid connection operations place extremely high demands on dispatcher experience and equipment performance. The operation process is complex and risky, and can easily generate large grid connection inrush currents, threatening equipment safety and system stability.

[0006] (3) Insufficient coordination and control capabilities: The traditional black start process relies heavily on the manual decision-making and instruction transmission of the scheduler. The coordination and synchronization between the recovery links are poor, making it difficult to achieve dynamic optimization of the recovery strategy and global optimization of the recovery path of the entire network. The entire recovery process takes a long time.

[0007] In recent years, energy storage technologies, represented by electrochemical energy storage (such as lithium-ion batteries), have made groundbreaking progress. Energy storage systems have outstanding advantages such as high power density, fast response speed (millisecond level), flexible configuration, and strong controllability, providing a new technical path for black start. Using an energy storage system as an initial black start power source can quickly establish voltage and frequency references, theoretically significantly accelerating the recovery process.

[0008] However, existing research and practice on applying energy storage to black start mainly focus on the simple connection and start-up verification of a single energy storage unit or power plant, which has obvious limitations and has not yet formed a mature and efficient system-level solution. (1) Weak system coordination: There is a lack of coordinated control strategies for multiple distributed energy storage power stations, different types of energy storage (such as power type and energy type), and other possible distributed power sources (such as photovoltaic and fuel cells). Each power unit often operates independently, making it difficult to form a stable and sufficiently large initial startup network, which limits the scope and capability of recovery.

[0009] (2) Coarse load restoration strategy: During the load restoration phase, the traditional hard access mode is still used, which fails to make full use of the energy storage system’s fast and accurate power control capabilities to achieve intelligent load sensing and flexible access. For example, it is unable to automatically decide the optimal access sequence and power ramp rate based on load characteristics and real-time grid status, or to mitigate the impact through load-side management (such as interruptible loads).

[0010] (3) Lack of grid synchronization mechanism: There is a lack of a coordination and control mechanism specifically designed for smooth and automatic grid synchronization between black-start systems with energy storage as the core and the main grid. Existing methods often directly apply the grid connection process of traditional power sources, failing to give full play to the unique advantages of energy storage as a "flexible" power source in terms of phase tracking, power smoothing, and active support, making it difficult to achieve shock-free grid connection.

[0011] (4) Lack of adaptive control throughout the entire process: The entire black start process is divided into several relatively independent stages such as start-up, network construction, load recovery, and grid connection, lacking an integrated, adaptive global controller. The system cannot dynamically adjust the control strategy according to real-time operating conditions (such as energy storage SOC status, network topology, and load demand), and cannot intelligently cope with various uncertainties and disturbances that occur during the recovery process. The robustness and intelligence level of the recovery process are insufficient.

[0012] No solutions have yet been proposed for the relevant technical issues. Summary of the Invention

[0013] To address the problems in related technologies, this invention proposes an integrated energy system for zero-carbon industrial parks to overcome the aforementioned technical issues in existing technologies. The purpose of this invention is to avoid black-start failures caused by single-point faults and start-up interruptions due to mismatches between preset plans and actual conditions, significantly improving reliability under extreme operating conditions and increasing the success rate of black starts. It fundamentally reduces inrush current during closing. The adaptive phase-locked loop dynamically optimizes bandwidth during adjustment, balancing speed and anti-interference capabilities, ensuring continuous high-quality power supply to sensitive loads, enhancing robustness under incomplete or delayed communication, ensuring cross-layer command transmission and status synchronization within milliseconds, and extending equipment lifespan.

[0014] To achieve the above objectives, the present invention provides the following technical solution: an energy storage black start system, comprising a physical layer, a coordination control layer, and a network communication layer; The physical layer includes at least one black-start power supply module, at least one load control module, and at least one grid-connected synchronization module. The black-start power supply module is composed of several energy storage converters connected in parallel, and each of the energy storage converters has a battery management subunit and a power conversion subunit. The load control module is composed of several intelligent load switching units, and each of the intelligent load switching units has a load characteristic identification subunit and a tiered switching subunit, used to connect loads sequentially according to a preset strategy. The grid-connected synchronization module has a grid status monitoring subunit and a flexible synchronization control subunit, used to achieve smooth and shock-free grid connection between the black-start islanded system and the main grid. The coordination and control layer includes a centralized system management unit and local autonomous controllers. There are three local autonomous controllers, which are respectively located inside the black-start power module, the load control module, and the grid-connected synchronization module. The centralized system management unit is used to send macroscopic instruction sequences to the local autonomous controllers according to the black-start plan, receive global status information, and execute distributed coordination decisions based on a multi-agent consensus algorithm. The network communication layer is a star-shaped heterogeneous redundant network based on time-sensitive networking, used to realize data interaction between the coordination and control layer and the physical layer.

[0015] Preferably, the output terminals of several energy storage converters are connected in parallel through a common connection point and are equipped with adaptive current sharing and circulating current suppression circuits. The local autonomous controllers inside several energy storage converters have pre-stored virtual synchronous generator control algorithms and droop control algorithms, and can seamlessly switch between voltage source mode and current source mode according to the instructions of the centralized system management unit.

[0016] Preferably, the load characteristic identification subunit identifies the equivalent impedance characteristics, start-up impulse characteristics, and power factor of the load to be connected online using a micro-current injection method before switching, and uploads the identification results to the centralized system management unit for dynamic optimization of the load switching sequence.

[0017] Preferably, the flexible synchronous control subunit adopts a control strategy based on a pre-synchronous phase sliding window and adaptive phase-locked loop bandwidth adjustment. After detecting the recovery of the main grid, it first controls the voltage and frequency of the black-start islanded system to stabilize within a preset allowable deviation, and then gradually approaches the main grid phase with a controlled phase difference change rate to achieve quasi-synchronous grid connection.

[0018] To achieve the above objectives, the present invention also provides the following technical solution: A control method for an energy storage black start system includes the following steps: S1, System Self-Test and Initialization: After power-on, the centralized system management unit broadcasts a self-test command through the network communication layer. The black-start power supply module, load control module, and grid-connected synchronization module complete local self-tests and report their status. The centralized system management unit dynamically generates or selects an appropriate black-start contingency plan based on the number and capacity of available black-start power supply modules, load control modules, and grid-connected synchronization modules. S2, Black Start Power Module Constructs Initial Islanded Power Grid: The centralized system management unit instructs the black start power module to start, and the designated main energy storage converter in the black start power module operates in VSG mode to establish the initial voltage and frequency. Other slave energy storage converters are automatically connected in parallel according to the adaptive current sharing algorithm to form a stable initial islanded bus. S3, Modular and Hierarchical Flexible Load Access: The centralized system management unit sends a hierarchical switching instruction sequence to the load control module according to the pre-plan and the load characteristics obtained in real time; After receiving the instruction, several intelligent load switching unit modules, under the control of their local autonomous controllers, perform soft switching operations at the voltage zero crossing point or current zero crossing point, and monitor the bus voltage fluctuation at the moment of switching. If the fluctuation exceeds the limit, the module will immediately lock out and report. S4, System Expansion and Dynamic Adjustment: As the load increases, the centralized system management unit dynamically instructs new black-start power modules or newly added energy storage converters to start and pre-synchronize, connecting them to the initial island bus to expand the system capacity; the black-start power modules dynamically share the load according to the VSG algorithm or droop algorithm; S5, Grid Connection Synchronization and Mode Switching: The grid connection synchronization module continuously monitors the status of the main grid; once the main grid returns to stability, the centralized system management unit instructs the grid connection synchronization module to start the flexible synchronization control process; when the grid connection conditions are met, the grid connection point switch is closed, and then the centralized system management unit issues an instruction to switch the control mode of the black start power supply module from voltage source mode to current source mode or power control mode, completing the seamless transition from black start island operation to grid connection operation.

[0019] Preferably, step S3 further includes, for the identified load, the corresponding intelligent load switching unit applying to the centralized system management unit for temporary power support; the centralized system management unit coordinates the black-start power module to temporarily increase reactive power output or briefly increase virtual inertia at the moment of switching to smooth out voltage drops.

[0020] Preferred options also include: S6: System Restructuring and Contingency Plan Learning: After the black start task is completed, the centralized system management unit records the key data of the entire startup process, including response time, load switching impact, and grid synchronization process parameters. It then uses this data to adaptively optimize and update the parameters of the black start contingency plan, forming an optimized contingency plan library for specific site configurations.

[0021] Compared with the prior art, the beneficial effects of the present invention are: (1) This invention is an energy storage black start system and its control method. By setting up a physical layer, a coordination control layer and a network communication layer, functional isolation and redundancy backup are achieved. When the communication link fails, the system can maintain degraded operation through the local autonomous controller and the heterogeneous redundant network, avoiding the failure of the entire black start caused by a single point of failure, and greatly improving the reliability of the system under extreme conditions. After power-on, the system automatically performs a full device self-test and dynamically generates or matches the optimal start plan based on the available resources in real time, avoiding the start interruption caused by the mismatch between the preset plan and the actual situation, and improving the black start success rate by more than 30%. (2) This invention is an energy storage black start system and its control method. By using the micro current injection method to identify the load impedance characteristics, impulse characteristics and power factor online before switching, it provides an accurate data basis for intelligent switching. Compared with the traditional time sequence switching or rough estimation based on rated power, this method reduces the load impulse current by more than 60%. Soft switching operation is implemented at the voltage or current zero crossing point, which fundamentally reduces the inrush current. At the same time, according to the identified impulse characteristics, the power module can be coordinated to provide temporary reactive power support or increase virtual inertia at the moment of switching, dynamically compensate for voltage drop, and ensure that the bus voltage fluctuation rate is controlled within 5%, effectively protecting the stability of the fragile island power grid. (3) This invention is an energy storage black start system and its control method. By adopting a synchronization method based on phase sliding window and adaptive phase-locked loop, the islanded system can approach the main grid phase at a controlled and slow rate. The adaptive phase-locked loop dynamically optimizes the bandwidth during the adjustment process, taking into account both speed and anti-interference, and reduces the current surge at the moment of grid connection to less than 10% of the rated current. By designing the energy storage converter to support dual modes of voltage source and current source, and executing a smooth mode switching algorithm after grid connection, the power oscillation caused by the sudden change of control mode is avoided, realizing the conversion from independent power supply to grid-connected operation, and ensuring continuous high-quality power supply to sensitive loads. (4) This invention is an energy storage black start system and its control method. After the black start task is executed, the key performance data of the whole process is automatically recorded, and the pre-plan parameters are optimized offline or online using machine learning algorithms. A customized optimal start strategy is formed for the load characteristics and network structure of a specific site, and the subsequent start time can be shortened by 15%-20%. The coordination control layer adopts a distributed decision algorithm based on multiple agents. Each local controller achieves consistency of control objectives through information exchange. This not only enhances the robustness of the system under incomplete or delayed communication, but also achieves better power allocation and frequency regulation, thereby improving the overall dynamic performance of the system. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the system framework of the present invention. Detailed Implementation

[0023] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0024] Example Please see Figure 1 This invention proposes a technical solution for an energy storage black start system and its control method: an energy storage black start system, comprising a physical layer, a coordination control layer, and a network communication layer; The physical layer includes at least one black-start power supply module, at least one load control module, and at least one grid-connected synchronization module. The black-start power supply module consists of several energy storage converters connected in parallel, specifically used to jointly provide the voltage and power support required for black starting. Each energy storage converter internally includes a battery management subunit and a power conversion subunit. The load control module consists of several intelligent load switching units, specifically used to connect loads sequentially according to a strategy to prevent impact. Each intelligent load switching unit internally includes a load characteristic identification subunit and a tiered switching subunit, used to connect loads sequentially according to a preset strategy. The grid-connected synchronization module internally includes an electrical... The grid status monitoring subunit and the flexible synchronization control subunit are used to achieve smooth and shock-free grid connection between the black-start islanded system and the main grid. Specifically, the grid status monitoring subunit continuously monitors the voltage, frequency, and phase of the main grid. When the main grid recovers and reaches the grid-connection state, the flexible synchronization control subunit is activated. Its control strategy first adjusts the voltage and frequency of the islanded system to within the allowable deviation range. Then, through a phase sliding window algorithm, it adjusts the phase of the islanded system with a controlled and slow phase difference change rate to gradually approach the phase of the main grid. At the same time, it adaptively adjusts the bandwidth of the phase-locked loop to balance dynamic response and disturbance rejection. Finally, when the phase difference is close to zero, it triggers the grid connection switch to close. The coordination and control layer includes a centralized system management unit and local autonomous controllers. There are three local autonomous controllers, which are respectively located inside the black-start power module, the load control module, and the grid-connected synchronization module. The centralized system management unit is used to send macro-level instruction sequences to the local autonomous controllers according to the black-start plan, receive global status information, and execute distributed coordination decisions based on a multi-agent consensus algorithm. Specifically, it realizes a combination of centralized management and distributed autonomy. The network communication layer is a star-shaped heterogeneous redundant network based on time-sensitive networking, used to realize data interaction between the coordination control layer and the physical layer.

[0025] Furthermore, the output terminals of several energy storage converters are connected in parallel through a common connection point and are equipped with adaptive current sharing and circulating current suppression circuits. The local autonomous controllers inside several energy storage converters are pre-stored with virtual synchronous generator control algorithms and droop control algorithms, and can seamlessly switch between voltage source mode and current source mode according to the instructions of the centralized system management unit.

[0026] In this embodiment, during startup, the main energy storage converter establishes voltage in VSG mode, and the slave energy storage converters are synchronously connected in parallel through an adaptive current sharing algorithm to jointly support the bus.

[0027] Furthermore, before switching, the load characteristic identification subunit identifies the equivalent impedance characteristics, startup impulse characteristics, and power factor of the load to be connected online using the micro-current injection method, and uploads the identification results to the centralized system management unit for dynamic optimization of the load switching sequence.

[0028] In this embodiment, the intelligent load switching unit is connected in series between the feeder and the load. Before switching, the load characteristic identification subunit inside the intelligent load switching unit injects a small probe current into the load and analyzes the response to identify the equivalent impedance, peak starting current, power factor and other characteristics of the load online, and reports the results. After receiving the switching command, the hierarchical switching subunit selects to close the switch at the voltage zero crossing point to achieve soft switching. If the switching causes the voltage fluctuation to exceed the limit, it immediately disconnects and reports.

[0029] Furthermore, the flexible synchronous control subunit adopts a control strategy based on a pre-synchronous phase sliding window and adaptive phase-locked loop bandwidth adjustment. After detecting the recovery of the main grid, it first controls the voltage and frequency of the black-start islanded system to stabilize within the preset allowable deviation, and then gradually approaches the main grid phase with a controlled phase difference change rate to achieve quasi-synchronous grid connection.

[0030] A control method for an energy storage black start system includes the following steps: S1, System Self-Test and Initialization: After power-on, the centralized system management unit broadcasts a self-test command through the network communication layer. The black-start power supply module, load control module, and grid-connected synchronization module complete local self-tests and report their status. The centralized system management unit dynamically generates or selects an appropriate black-start contingency plan based on the number and capacity of available black-start power supply modules, load control modules, and grid-connected synchronization modules. S2, Black Start Power Module Constructs Initial Islanded Power Grid: The centralized system management unit instructs the black start power module to start, and the designated main energy storage converter in the black start power module operates in VSG mode to establish the initial voltage and frequency. Other slave energy storage converters are automatically connected in parallel according to the adaptive current sharing algorithm to form a stable initial islanded bus. S3, Modular and Hierarchical Flexible Load Access: The centralized system management unit sends a hierarchical switching instruction sequence to the load control module based on the pre-plan and the load characteristics obtained in real time; After receiving the instruction, several intelligent load switching unit modules, under the control of their local autonomous controllers, perform soft switching operations at the voltage zero crossing point or current zero crossing point, and monitor the bus voltage fluctuation at the moment of switching. If the fluctuation exceeds the limit, it will be immediately blocked and reported. S4, System Expansion and Dynamic Adjustment: As the load increases, the centralized system management unit dynamically instructs new black-start power modules or newly added energy storage converters to start and pre-synchronize, connecting them to the initial island bus to expand the system capacity; the black-start power modules dynamically share the load according to the VSG algorithm or droop algorithm; S5, Grid Connection Synchronization and Mode Switching: The grid connection synchronization module continuously monitors the status of the main grid; once the main grid returns to stability, the centralized system management unit instructs the grid connection synchronization module to start the flexible synchronization control process; when the grid connection conditions are met, the grid connection point switch is closed, and then the centralized system management unit issues an instruction to switch the control mode of the black start power supply module from voltage source mode to current source mode or power control mode, completing the seamless transition from black start island operation to grid connection operation.

[0031] Furthermore, step S3 also includes the intelligent load switching unit corresponding to the identified load requesting temporary power support from the centralized system management unit; the centralized system management unit coordinates the black-start power module to temporarily increase reactive power output or briefly increase virtual inertia at the moment of switching to smooth out voltage drops.

[0032] Furthermore, it also includes: S6: System Restructuring and Contingency Plan Learning: After the black start task is completed, the centralized system management unit records the key data of the entire startup process, including response time, load switching impact, and grid synchronization process parameters. It then uses this data to adaptively optimize and update the parameters of the black start contingency plan, forming an optimized contingency plan library for specific site configurations.

[0033] Working principle of the invention: 1. The working principle of the physical layer: Several energy storage converters are electrically connected in parallel through a common connection point. Each energy storage converter can operate as both a voltage source and a current source. An adaptive current sharing circuit monitors the current in each branch in real time and ensures that each energy storage converter shares the load proportionally according to its capacity by adjusting the power distribution coefficient. At the same time, a circulating current suppression circuit detects and eliminates circulating currents caused by parameter differences.

[0034] Voltage source mode (black start phase): The main energy storage converter operates in virtual synchronous generator mode, simulating the rotor motion equation of a synchronous generator through the VSG algorithm to establish stable voltage and frequency support. VSG control provides inertia and damping characteristics to the system through virtual inertia elements (J·dω / dt) and damping elements (D·Δω), enhancing the stability of the islanded power grid.

[0035] Current source mode (grid-connected stage): After receiving the mode switching command, the energy storage converter control algorithm switches from the voltage outer loop to the current inner loop priority, and outputs the controlled current according to the phase tracking of the main grid voltage, so as to realize a smooth transition from independent power supply to controlled grid connection.

[0036] Working principle of load control module: Before switching, the load characteristic identification subunit injects a micro-current signal (typically 1%-5% of the rated current) with controllable amplitude and adjustable frequency into the load port. By measuring the terminal voltage response, the equivalent impedance of the load at multiple frequency points is calculated using impedance frequency domain analysis. Combined with time-domain starting current analysis, the RLC equivalent model and starting characteristic curve of the load are established.

[0037] The hierarchical switching subunit generates the optimal switching sequence based on the identification results and preset priorities, accurately captures the zero-crossing point of the grid voltage, and closes the power electronic switch (such as a solid-state relay) at this moment, so that the load is connected when the instantaneous voltage value is zero, theoretically achieving zero inrush current. The bus voltage is sampled at the millisecond level and calculated using a sliding window RMS. Once a voltage drop exceeding the threshold (such as 10%) is detected, protective disconnection is immediately triggered.

[0038] The power grid condition monitoring subunit adopts a multi-stage phase-locked loop design, extracting the positive-sequence component of the main grid through αβ / dq transformation and filtering out harmonics and asymmetric components. Recovery criteria are set as follows: voltage within ±10% of nominal value for 5 consecutive cycles, frequency within 49.5-50.5Hz, and phase fluctuation rate less than 1° / s.

[0039] When the main grid meets the conditions, the phase sliding window algorithm is activated. This algorithm limits the phase difference (θ_grid - θ_island) to a time-varying allowable window, with the window width decreasing exponentially from an initial ±30° to ±2°. The phase-locked loop bandwidth is dynamically adjusted based on the phase difference—high bandwidth (e.g., 100Hz) is used for fast tracking with large deviations, while switching to low bandwidth (e.g., 10Hz) enhances noise immunity with small deviations. Phase adjustment uses a controlled ramp function: dθ / dt = K·tanh(Δθ), ensuring continuous phase change rate without overshoot. A grid connection command is issued when |ΔV| < 2%, |Δf| < 0.1Hz, |Δθ| < 5°, and dθ / dt < 1° / s. After the switch is closed, the grid synchronization module continues to monitor power fluctuations at the grid connection point and makes fine adjustments to ensure no impact.

[0040] 2. Working principle of the coordination and control layer: The centralized system management unit incorporates multiple standard black-start contingency plans (such as minimum system startup, critical load priority, and maximum range recovery). During the initialization phase, contingency plan matching and parameter adaptation are performed based on available resources (energy storage capacity, load list, and communication status), generating a time-stamped instruction sequence. Each local autonomous controller is treated as an agent, employing distributed decision-making based on a consensus algorithm. Each agent maintains a local state vector (such as voltage, frequency, and power), exchanges information with neighboring nodes through a communication network, and iteratively calculates to achieve a globally consistent control objective (such as frequency unification and optimal power allocation). Real-time solution of constrained optimization problems is performed: the objective function is to minimize recovery time, maximize power supply to critical loads, and minimize switching operations; constraints include equipment capacity limitations, voltage fluctuation limits, and grid synchronization conditions.

[0041] The local autonomous controller on the black-start power supply module performs two-layer control: the upper layer receives centralized commands (mode switching, power setting), and the lower layer implements local closed-loop control (VSG algorithm, droop control). It features plug-and-play capability; newly connected energy storage converters automatically register via a broadcast discovery protocol to obtain system parameters. The local autonomous controller on the load control module maintains a local load database and executes a closed-loop process of identification-evaluation-application-execution-feedback. The local autonomous controller on the grid synchronization module runs an independent state machine and grid connection logic, enabling basic grid synchronization operations based on local measurements even when centralized commands are lost, achieving fault-degraded operation.

[0042] 3. Working principle of network communication layer An IEEE 802.1Qbv time-aware shaper is used to divide communication traffic into periodically scheduled traffic (such as control commands and synchronization signals) and best-effort traffic (such as status monitoring and event logging). Fixed time slots are allocated to critical control messages to ensure end-to-end latency is less than 1ms and jitter is less than 10μs.

[0043] The core switch and each device node form a star topology, with dual-link redundancy and dual protocol stacks (such as Profinet+EtherCAT) deployed. A fault detection mechanism continuously monitors link quality, switching to a backup link within 50ms when the primary link fails, ensuring uninterrupted control.

[0044] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An energy storage black start system, characterized in that, It includes the physical layer, coordination and control layer, and network communication layer; The physical layer includes at least one black-start power supply module, at least one load control module, and at least one grid-connected synchronization module. The black-start power supply module is composed of several energy storage converters connected in parallel, and each of the energy storage converters has a battery management subunit and a power conversion subunit. The load control module is composed of several intelligent load switching units, and each of the intelligent load switching units has a load characteristic identification subunit and a tiered switching subunit, used to connect loads sequentially according to a preset strategy. The grid-connected synchronization module has a grid status monitoring subunit and a flexible synchronization control subunit, used to achieve smooth and shock-free grid connection between the black-start islanded system and the main grid. The coordination and control layer includes a centralized system management unit and local autonomous controllers. There are three local autonomous controllers, which are respectively located inside the black-start power supply module, the load control module, and the grid-connected synchronization module. The centralized system management unit is used to send a sequence of macroscopic instructions to the local autonomous controller according to the black start plan, receive global status information, and execute distributed coordination decisions based on a multi-agent consensus algorithm. The network communication layer is a star-shaped heterogeneous redundant network based on time-sensitive networking, used to realize data interaction between the coordination and control layer and the physical layer.

2. The energy storage black start system according to claim 1, characterized in that, The output terminals of several energy storage converters are connected in parallel through a common connection point and are equipped with adaptive current sharing and circulating current suppression circuits. The local autonomous controllers inside several energy storage converters have pre-stored virtual synchronous generator control algorithms and droop control algorithms, and can seamlessly switch between voltage source mode and current source mode according to the instructions of the centralized system management unit.

3. The energy storage black start system according to claim 1, characterized in that, Before switching, the load characteristic identification subunit identifies the equivalent impedance characteristics, startup impulse characteristics, and power factor of the load to be connected online using the microcurrent injection method, and uploads the identification results to the centralized system management unit for dynamic optimization of the load switching sequence.

4. The energy storage black start system according to claim 1, characterized in that, The flexible synchronous control subunit adopts a control strategy based on a pre-synchronous phase sliding window and adaptive phase-locked loop bandwidth adjustment. After detecting the recovery of the main grid, it first controls the voltage and frequency of the black-start islanded system to stabilize within the preset allowable deviation, and then gradually approaches the main grid phase with a controlled phase difference change rate to achieve quasi-synchronous grid connection.

5. A control method for an energy storage black start system as described in any one of claims 1 to 4, characterized in that, Includes the following steps: S1, System Self-Test and Initialization: After power-on, the centralized system management unit broadcasts a self-test command through the network communication layer. The black-start power supply module, load control module, and grid-connected synchronization module complete local self-tests and report their status. The centralized system management unit dynamically generates or selects an appropriate black-start contingency plan based on the number and capacity of available black-start power supply modules, load control modules, and grid-connected synchronization modules. S2, Black Start Power Module Constructs Initial Islanded Power Grid: The centralized system management unit instructs the black start power module to start, and the designated main energy storage converter in the black start power module operates in VSG mode to establish the initial voltage and frequency. Other slave energy storage converters are automatically connected in parallel according to the adaptive current sharing algorithm to form a stable initial islanded bus. S3, Modular and Hierarchical Flexible Load Access: The centralized system management unit sends a hierarchical switching instruction sequence to the load control module according to the pre-plan and the load characteristics obtained in real time; After receiving the instruction, several intelligent load switching unit modules, under the control of their local autonomous controllers, perform soft switching operations at the voltage zero crossing point or current zero crossing point, and monitor the bus voltage fluctuation at the moment of switching. If the fluctuation exceeds the limit, the module will immediately lock out and report. S4, System Expansion and Dynamic Adjustment: As the load increases, the centralized system management unit dynamically instructs new black-start power modules or newly added energy storage converters to start and pre-synchronize, connecting them to the initial island bus to expand the system capacity; the black-start power modules dynamically share the load according to the VSG algorithm or droop algorithm; S5, Grid Connection Synchronization and Mode Switching: The grid connection synchronization module continuously monitors the status of the main grid; once the main grid returns to stability, the centralized system management unit instructs the grid connection synchronization module to start the flexible synchronization control process; when the grid connection conditions are met, the grid connection point switch is closed, and then the centralized system management unit issues an instruction to switch the control mode of the black start power supply module from voltage source mode to current source mode or power control mode, completing the seamless transition from black start island operation to grid connection operation.

6. The control method for an energy storage black start system according to claim 5, characterized in that, Step S3 also includes the intelligent load switching unit corresponding to the identified load requesting temporary power support from the centralized system management unit; the centralized system management unit coordinates the black-start power module to temporarily increase reactive power output or briefly increase virtual inertia at the moment of switching to smooth out voltage drops.

7. The control method for an energy storage black start system according to claim 5, characterized in that, Also includes: S6: System Restructuring and Contingency Plan Learning: After the black start task is completed, the centralized system management unit records the key data of the entire startup process, including response time, load switching impact, and grid synchronization process parameters. It then uses this data to adaptively optimize and update the parameters of the black start contingency plan, forming an optimized contingency plan library for specific site configurations.

Images