A Multi-Source Cooperative Black Start Control Method Based on Hierarchical Startup

By employing a multi-source collaborative black start control method, utilizing multi-source availability quantification modeling and hierarchical start logic, the fault tolerance and stability issues in microgrid black start are resolved, achieving highly reliable black start under extreme operating conditions and a smooth start process adaptable to different operating conditions.

CN123073751APending Publication Date: 2026-11-10YALONG RIVER HYDROPOWER DEV CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611346702.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-01
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

In existing microgrid black start schemes, the single fixed power supply start logic has problems such as poor fault tolerance, low start success rate under extreme conditions, poor dynamic stability during the start process, and weak adaptability to operating conditions. In particular, it is easy to cause black start failure in extreme cases such as energy storage power loss and photovoltaic power output failure.

Method used

A multi-source collaborative black start control method based on hierarchical startup is adopted. Through multi-source availability quantification modeling, adaptive startup mode matching and four-level hierarchical startup, combined with global current limiting and voltage regulation, multi-source collaborative black start is achieved to adapt to different operating conditions.

Benefits of technology

It achieves highly reliable, oscillation-free black start under extreme operating conditions, improves the system's self-healing capability and stability, avoids the startup shock and voltage and current fluctuations in traditional solutions, and has full-coverage fault tolerance capability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure QLYQS_1
    Figure QLYQS_1
  • Figure QLYQS_2
    Figure QLYQS_2
  • Figure QLYQS_3
    Figure QLYQS_3
Patent Text Reader

Abstract

This invention belongs to the field of black-start control technology for DC microgrids, and discloses a multi-source coordinated black-start control method based on hierarchical startup. First, it detects the availability and operational margin of three types of power sources: photovoltaic, energy storage, and mains power in real time. Then, it adaptively classifies the startup mode into battery-led, photovoltaic-assisted, and mains-coordinated startup modes using quantitative criteria. Following a hierarchical timing logic of bus pre-charging, rated voltage establishment, smooth multi-source grid connection, and gradual load connection, the system completes the black start. Adaptive current limiting and voltage stabilization closed-loop control are used throughout the process to suppress startup shocks. This invention overcomes the limitations of traditional single-source black-start methods, which suffer from poor fault tolerance and limited adaptability to specific operating conditions. It can comprehensively cover harsh scenarios such as extreme power outages, weak power sources, and power shortages, significantly improving the success rate of microgrid black-start and the dynamic stability of the startup process. It requires no hardware modification, has strong algorithm portability, and possesses extremely high engineering application value and universality.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of microgrid power failure restart black start control technology, specifically to a multi-source collaborative black start control method based on hierarchical start. Background Technology

[0002] Three-input DC microgrids combining photovoltaic, energy storage, and grid connection are widely used in distributed renewable energy power supply, outdoor monitoring base stations, emergency backup power supplies, and building energy-saving power supply scenarios. Microgrids are highly susceptible to grid-wide power outages under extreme weather conditions, grid faults, overload protection, and line disturbances. After a complete power outage, the system needs to rely on black-start technology to autonomously restart and restore power. Black-start capability is a core technology for ensuring the reliability and self-healing ability of microgrid power supply.

[0003] Currently, most microgrid black-start solutions in the industry adopt a single fixed power supply startup logic, with the vast majority of scenarios defaulting to energy storage batteries as the sole black-start power source. This results in significant limitations in operating conditions and startup defects.

[0004] First, the startup fault tolerance is extremely poor. When the energy storage is in a low SOC state and cannot provide startup power, or when the photovoltaic has no output at night or in the case of cloudy and rainy conditions, the failure of a single startup power source will directly lead to black start failure and long-term system paralysis.

[0005] Second, the dynamic stability of the startup process is poor. Traditional single-source startup has no graded buffering mechanism. It directly builds up voltage at full power and directly connects to the load, which is very easy to generate large inrush current, bus voltage overshoot and oscillation. In severe cases, it triggers overcurrent and overvoltage protection, causing repeated startup failures.

[0006] Third, the adaptability to operating conditions is weak. Existing technologies cannot adaptively switch the starting entity according to the real-time availability of photovoltaic, energy storage and grid power, and cannot adapt to complex power outage conditions under multi-source combinations. The self-healing capability in extreme scenarios is seriously insufficient.

[0007] Existing improvement technologies are mostly limited to voltage regulation optimization of a single power source, failing to innovate from the core logic of multi-source collaboration, mode adaptation, and tiered start-up. Therefore, they cannot fundamentally solve the industry pain points of low black-start success rate, large fluctuations, and poor fault tolerance under extreme operating conditions. Thus, there is an urgent need for a multi-source collaborative black-start control scheme that can adapt to multiple source states, provide tiered smooth start-up, and fully cover extreme power failure conditions. Summary of the Invention

[0008] This invention aims to overcome the shortcomings of existing single-power-source black-start strategies, such as poor fault tolerance, low success rate of starting under extreme conditions, severe voltage and current fluctuations during the startup process, and limited adaptability to different operating conditions. It provides a multi-source collaborative black-start control method based on hierarchical startup. Through quantitative modeling of multi-source availability, adaptive startup mode matching, four-level hierarchical startup, and global current limiting and voltage regulation, it achieves microgrid black-start with high reliability and oscillation-free operation in all scenarios, significantly improving the system's self-healing capability and stability under extreme conditions.

[0009] To achieve the above objectives, the present invention adopts the following technical solution:

[0010] A multi-source coordinated black-start control method based on hierarchical start-up is applied to a three-input DC microgrid system integrating photovoltaic, mains power, and energy storage, comprising:

[0011] S1. Power loss determination and multi-source status full-domain perception: Real-time acquisition of bus residual voltage, energy storage SOC, energy storage output power, real-time photovoltaic output, light intensity, and mains power operation status, accurately determining the system's complete power loss status, and completing the black start pre-state screening.

[0012] S2. Multi-source availability quantitative modeling: Construct a three-dimensional availability calculation formula for energy storage, photovoltaic and grid power, quantitatively evaluate the current black start support capability of each power source, abandon the traditional qualitative judgment, and realize the quantitative judgment of the optimal start-up power source.

[0013] S3, Adaptive Startup Mode Matching: Automatically switches between three startup modes—battery-dominated, photovoltaic-assisted, and grid-assisted—based on the maximum availability coefficient, adapting to various extreme operating conditions such as power depletion, low light, and grid anomalies.

[0014] S4, Four-level graded black start: Starts up step by step according to the sequence of pre-charging, voltage building up, multi-source grid connection, and load graded connection, with step-by-step buffering and voltage stabilization to completely eliminate the start-up shock;

[0015] S5, Global Hierarchical Current Limiting and Voltage Regulation: Differentiated current limiting thresholds and voltage feedforward compensation are matched for each startup stage to achieve a smooth, oscillation-free, and highly reliable black start throughout the entire process.

[0016] Preferably, the multi-source black-start availability quantification evaluation model in step S2 includes three types of quantification coefficients: energy storage availability, photovoltaic availability, and grid availability. The calculation formula is as follows:

[0017] ;

[0018] ;

[0019] ;

[0020] in: These are the black start-up availability coefficients for energy storage, photovoltaic, and mains power, respectively. The minimum and maximum charge thresholds for energy storage startup; For energy storage, it can output power. Rated output power; For real-time light intensity, Rated light intensity; This is the mains power availability switching coefficient, where 1 is normal and 0 is abnormal. Let be the weight correction coefficient, and satisfy... .

[0021] Preferably, three black start modes are adaptively divided based on availability coefficients, and the mode determination rules are as follows:

[0022] when and At that time, it is determined to be a battery-dominated black start mode, with energy storage as the main starting power source and photovoltaic and mains power as auxiliary backup power sources.

[0023] when and When the system is in a photovoltaic-assisted black start mode, the photovoltaic system is used as the main starting power source, with energy storage buffering and voltage stabilization, and the mains power supply as a backup.

[0024] when and At that time, it was determined to be a black start mode with grid power coordination, using grid power as the main starting power source to avoid problems such as energy storage power loss and photovoltaic weak light start failure.

[0025] Preferably, the four-level hierarchical step-by-step black start timing logic in step S3 is as follows:

[0026] First stage: Bus pre-charging stage, the main starting power supply is selected to pre-charge the DC bus in a low current limiting mode to eliminate residual voltage difference and capacitive load impact, and raise the bus voltage to 50% to 60% of the rated voltage;

[0027] Second stage: Voltage establishment stage, increase the output power of the main starting power supply, gradually stabilize the bus voltage to the rated voltage range, start voltage closed-loop regulation, and suppress voltage overshoot;

[0028] Level 3: Multi-source coordinated grid connection stage. Based on the availability of each power source, auxiliary power sources are connected sequentially. A power ramp-up strategy is adopted to achieve multi-source grid connection without impact and complete the multi-source power coordinated matching.

[0029] Level 4: Load grading phase, loads are gradually applied in stages from light load to medium load to heavy load to avoid startup failure caused by instantaneous high power surges and to complete a full black start recovery of the system.

[0030] Preferably, the global current limiting and voltage regulation in step S4 includes tiered adaptive current limiting control, with the current limiting threshold increasing progressively at each startup stage:

[0031] The first current limiting threshold is used during the pre-charging stage to limit the inrush current; the second current limiting threshold is used during the voltage establishment stage, and the second current limiting threshold is greater than the first current limiting threshold; the rated current limiting threshold is used during the multi-source grid connection and load connection stage to improve the startup speed while ensuring startup safety.

[0032] The present invention also provides a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, it implements the steps of the multi-source cooperative black-start control method based on hierarchical startup described above.

[0033] Compared with the prior art, the present invention has the following outstanding innovations and technical advantages:

[0034] 1. This invention uses a multi-source availability quantification model to adaptively select the optimal power supply, completely solving the problem of startup failure under extreme conditions such as energy storage power loss and photovoltaic weak light. It achieves full coverage of black start scenarios and its fault tolerance capability far exceeds that of traditional fixed strategies.

[0035] 2. This invention adopts a four-level timing-based startup logic, combined with a step-by-step current limiting and voltage regulation mechanism, to eliminate the current surges and voltage oscillations caused by traditional direct voltage build-up and direct load-bearing, resulting in a smooth startup process without overshoot.

[0036] 3. This invention uses formulas to quantitatively calculate the starting capability of each power supply, rather than relying on traditional empirical qualitative judgment. It has high control precision, strong adaptability, and extremely high patent innovation and protection barriers.

[0037] 4. This invention does not require modification of the original microgrid topology and power devices. Functional upgrades can be achieved simply by iterating the main control program. It has strong engineering applicability and wide applicability. Detailed Implementation

[0038] As one embodiment of the present invention, a multi-source coordinated black-start control method based on hierarchical start-up is first provided, applied to a three-input DC microgrid system of photovoltaic, mains power, and energy storage, for black-start recovery after a complete power outage. The method includes the following steps:

[0039] S1. System power failure determination and multi-source status perception: Detect the residual voltage of the system's DC bus to determine the system's complete power failure state; collect data in real time on the energy storage charge status, energy storage output power, real-time photovoltaic output, light intensity, and grid availability to construct a multi-source black start availability evaluation dataset.

[0040] S2. Quantitative evaluation of availability of multi-source black start: Construct a multi-source black start availability calculation model, solve the black start availability coefficients of energy storage, photovoltaic and grid power respectively, and adaptively match the system black start mode according to the size of the availability coefficients;

[0041] S3, graded black start mode switching control: According to the matched start mode, the system black start recovery is completed step by step according to the four-level time sequence of bus pre-charging stage, rated voltage establishment stage, multi-source collaborative grid connection stage, and load graded connection stage.

[0042] S4. Full-domain current limiting and voltage regulation during startup: Differentiated current limiting thresholds and voltage regulation closed-loop parameters are matched at each startup stage to suppress startup inrush current and bus voltage fluctuations, achieving smooth black startup without oscillation.

[0043] As a preferred implementation, the multi-source black-start availability quantification evaluation model in step S2 includes three types of quantification coefficients: energy storage availability, photovoltaic availability, and grid availability. The calculation formula is as follows:

[0044] Abat=ω1⋅SOC−SOCminSOCmax−SOCmin⋅Pbat_outPbat_rated;

[0045] Apv=ω2⋅PpvrealPpvmax⋅GGrated;

[0046] Agrid=ω3⋅δgrid⋅(1−|Ugrid−Ugrid_rated|Ugrid_rated);

[0047] Where: Abat, Apv, and Agrid are the black-start availability coefficients for energy storage, photovoltaic, and mains power, respectively; SOCmin and SOCmax are the minimum and maximum charge thresholds for energy storage startup; Pbat_out is the output power of energy storage, and Pbat_rated is the rated output power; G is the real-time illuminance, and Grated is the rated illuminance; δgrid is the mains power availability switching coefficient, which is 1 for normal mains power and 0 for abnormal mains power; ω1, ω2, and ω3 are weight correction coefficients, and satisfy ω1+ω2+ω3=1.

[0048] In a further preferred embodiment, three black start modes are adaptively divided based on the availability coefficient, and the mode determination rule is as follows:

[0049] when and At that time, it is determined to be a battery-dominated black start mode, with energy storage as the main starting power source and photovoltaic and mains power as auxiliary backup power sources.

[0050] when and When the system is in a photovoltaic-assisted black start mode, the photovoltaic system is used as the main starting power source, with energy storage buffering and voltage stabilization, and the mains power supply as a backup.

[0051] when and At that time, it was determined to be a black start mode with grid power coordination, using grid power as the main starting power source to avoid problems such as energy storage power loss and photovoltaic weak light start failure.

[0052] In another preferred embodiment, the four-level hierarchical step-by-step black start timing logic in step S3 is specifically as follows:

[0053] First stage: Bus pre-charging stage, the main starting power supply is selected to pre-charge the DC bus in a low current limiting mode to eliminate residual voltage difference and capacitive load impact, and raise the bus voltage to 50% to 60% of the rated voltage;

[0054] Second stage: Voltage establishment stage, increase the output power of the main starting power supply, gradually stabilize the bus voltage to the rated voltage range, start voltage closed-loop regulation, and suppress voltage overshoot;

[0055] Level 3: Multi-source coordinated grid connection stage. Based on the availability of each power source, auxiliary power sources are connected sequentially. A power ramp-up strategy is adopted to achieve multi-source grid connection without impact and complete the multi-source power coordinated matching.

[0056] Level 4: Load grading phase, loads are gradually applied in stages from light load to medium load to heavy load to avoid startup failure caused by instantaneous high power surges and to complete a full black start recovery of the system.

[0057] More preferably, the global current limiting and voltage regulation in step S4 includes tiered adaptive current limiting control, with the current limiting threshold increasing progressively at each startup stage:

[0058] The first current limiting threshold is used during the pre-charging stage to limit the inrush current; the second current limiting threshold is used during the voltage establishment stage, and the second current limiting threshold is greater than the first current limiting threshold; the rated current limiting threshold is used during the multi-source grid connection and load connection stage to improve the startup speed while ensuring startup safety.

[0059] The present invention also relates to a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the multi-source cooperative black-start control method based on hierarchical startup as described in any of the preceding claims.

[0060] The present invention will now be described with reference to specific embodiments.

[0061] Example 1

[0062] This embodiment uses a 48V DC microgrid as the application object, with a system rated voltage of... Set weight coefficients The system experienced a complete power outage, with the residual voltage on the busbars falling below 5V, indicating a total power failure.

[0063] Real-time status parameters: Energy storage SOC = 72%, capable of normal high-power discharge; sufficient sunlight, photovoltaic power generation at full capacity; mains power is available normally. Substituting these parameters into the availability formula, we can calculate that the energy storage availability coefficient is at its maximum, indicating that the system is matched to a battery-dominated black-start mode.

[0064] Tiered startup process:

[0065] 1. Pre-charging stage: The energy storage pre-charges the bus in a low-current limiting mode, raising the bus voltage to 28V and eliminating capacitive load impact.

[0066] 2. Voltage setup phase: Gradually increase energy storage output, and use closed-loop voltage regulation to stabilize the bus voltage within the rated range of 48V, with voltage fluctuations controlled within ±0.3V;

[0067] 3. Multi-source grid connection stage: Photovoltaic and grid power are connected to the grid smoothly in sequence, and a power ramp-up strategy is adopted to achieve coordinated power distribution from multiple sources;

[0068] 4. Load grading: Light load, medium load, and heavy load are applied in sequence without impact or oscillation, and the system successfully completes black start recovery.

[0069] Example 2

[0070] This embodiment is an extreme operating condition verification scenario. The system is also a 48V three-input microgrid, simulating common engineering faults: energy storage power loss, no photovoltaic output at night, and power outage of the entire grid.

[0071] Real-time status: Energy storage SOC=18%, below the reliable start-up threshold, unable to take the lead in starting; Nighttime light intensity is 0, photovoltaic availability is 0; Mains power is normal and available.

[0072] Calculated using an availability quantification model, the grid availability coefficient is much greater than that of photovoltaics and energy storage. The system adaptively switches to a grid-coordinated black start mode to avoid start-up failures caused by energy storage power loss or photovoltaic failure.

[0073] The entire startup process employs a phased approach to voltage building and current limiting within the mains power supply. First, the busbar is pre-charged, then the rated voltage is established. Subsequently, low-power activation of the energy storage in standby mode is performed, multi-source grid connection is completed, and finally, the load is connected in stages. Under these extreme conditions, traditional single-energy storage startup schemes would inevitably fail. This invention can achieve a stable, oscillating-free black start, verifying the strong fault tolerance and engineering practicality of this scheme.

[0074] Example 3

[0075] This embodiment simulates the most common combined power outage condition in engineering sites: continuous rainy weather leads to a significant decrease in photovoltaic output, while the energy storage is in a moderately depleted state due to prolonged shallow discharge and charging. Neither single energy storage nor single photovoltaic startup can stably build up voltage, representing a typical scenario where traditional black-start strategies are prone to startup failure. This embodiment still uses a 48V DC bus three-input microgrid system with a rated voltage... Using fixed weighting coefficients .

[0076] Operating parameters: The entire system loses power, and the residual voltage on the bus is 3.6V, indicating a complete power loss; the real-time SOC of the energy storage is 35%, indicating a moderate power shortage. It can discharge slightly but cannot independently support the entire black start voltage build-up process; the sunlight intensity is significantly reduced during cloudy or rainy weather, only 35% of the rated sunlight. The photovoltaic system can output a small amount of power but cannot start the system independently; there are slight voltage fluctuations in the mains power, and the grid quality is generally poor, making it unsuitable as the main starting power source.

[0077] Substitute the real-time parameters into the multi-source available quantification model formula:

[0078] ;

[0079] ;

[0080] ;

[0081] The quantitative calculation results show that the photovoltaic availability coefficient is the best, followed by the energy storage availability coefficient, and the grid power availability coefficient is the lowest. The system adaptively matches the photovoltaic-assisted black start mode.

[0082] The specific process of tiered startup:

[0083] 1. Bus pre-charging stage: Using low-output photovoltaic as the main starting power source, the first-stage low-current limiting mode is adopted to slowly pre-charge the capacitive load of the bus, avoid the impact voltage drop under weak source conditions, and steadily raise the bus voltage to 26V to complete the early stage of voltage stabilization and buffering.

[0084] 2. Voltage setup stage: Slightly increase the output power of photovoltaics, and at the same time activate the energy storage buffer voltage stabilization auxiliary. Utilize the fast response characteristics of energy storage to offset the small power fluctuations of photovoltaics. With the voltage feedforward correction mechanism, the bus voltage is accurately stabilized in the 48V rated range, solving the problem of unstable voltage build-up of weak photovoltaics.

[0085] 3. Multi-source coordinated grid connection stage: The power ramp-up strategy is adopted to first stabilize the main output of photovoltaic power and dynamically shave and fill the valley of energy storage. After the bus voltage is completely stable, it is flexibly connected to the grid for bottom-line voltage stabilization to achieve coordinated power matching of the three sources.

[0086] 4. Load grading and commissioning phase: Strictly follow the light load and medium load grading to commission the load, postpone the commissioning of heavy load, and wait for the power of multiple sources to be dynamically balanced and the voltage to be stable before commissioning all loads, so as to completely avoid the risk of startup collapse under the combined weak source conditions.

[0087] Under these combined operating conditions, traditional single-energy storage startup will fail due to energy storage depletion, leading to voltage build-up failure. Single-photovoltaic startup will cause voltage oscillation and collapse due to insufficient output and large fluctuations, rendering the fixed-priority startup strategy completely ineffective. However, this invention, through a quantitative model to adaptively match the photovoltaic-assisted startup mode, combined with multi-source collaborative buffering and graded current limiting and voltage stabilization mechanisms, can smoothly complete the entire black start process. It is perfectly adapted to the combined harsh operating conditions of weak photovoltaic power generation and energy storage depletion, demonstrating extremely strong engineering adaptability.

Claims

1. A multi-source cooperative black-start control method based on hierarchical startup, characterized in that, Applied to a three-input DC microgrid system integrating photovoltaic, mains power, and energy storage, and used for black-start recovery after a complete power outage, the method includes the following steps: S1. System power failure determination and multi-source status perception: Detect the residual voltage of the system's DC bus to determine the system's complete power failure state; collect data in real time on the energy storage charge status, energy storage output power, real-time photovoltaic output, light intensity, and grid availability to construct a multi-source black start availability evaluation dataset. S2. Quantitative evaluation of availability of multi-source black start: Construct a multi-source black start availability calculation model, solve the black start availability coefficients of energy storage, photovoltaic and grid power respectively, and adaptively match the system black start mode according to the size of the availability coefficients; S3, graded black start mode switching control: According to the matched start mode, the system black start recovery is completed step by step according to the four-level time sequence of bus pre-charging stage, rated voltage establishment stage, multi-source collaborative grid connection stage, and load graded connection stage. S4. Full-domain current limiting and voltage regulation during startup: Differentiated current limiting thresholds and voltage regulation closed-loop parameters are matched at each startup stage to suppress startup inrush current and bus voltage fluctuations, achieving smooth black startup without oscillation.

2. The multi-source cooperative black-start control method based on hierarchical startup according to claim 1, characterized in that, The multi-source black-start availability quantification evaluation model mentioned in step S2 includes three types of quantification coefficients: energy storage availability, photovoltaic availability, and grid availability. The calculation formula is as follows: ; ; ; in: These are the black start-up availability coefficients for energy storage, photovoltaic, and mains power, respectively. The minimum and maximum charge thresholds for energy storage startup; For energy storage, it can output power. Rated output power; For real-time light intensity, Rated light intensity; This is the mains power availability switching coefficient, where 1 is normal and 0 is abnormal. Let be the weight correction coefficient, and satisfy... .

3. The multi-source cooperative black-start control method based on hierarchical startup according to claim 2, characterized in that, Based on availability coefficients, three black boot modes are adaptively categorized, with the mode determination rule as follows: when and At that time, it is determined to be a battery-dominated black start mode, with energy storage as the main starting power source and photovoltaic and mains power as auxiliary backup power sources. when and When the system is in a photovoltaic-assisted black start mode, the photovoltaic system is used as the main starting power source, with energy storage buffering and voltage stabilization, and the mains power supply as a backup. when and At that time, it was determined to be a black start mode with grid power coordination, using grid power as the main starting power source to avoid problems such as energy storage power loss and photovoltaic weak light start failure.

4. The multi-source cooperative black-start control method based on hierarchical startup according to claim 1, characterized in that, The four-level hierarchical step-by-step black start timing logic in step S3 is as follows: First stage: Bus pre-charging stage, the main starting power supply is selected to pre-charge the DC bus in a low current limiting mode to eliminate residual voltage difference and capacitive load impact, and raise the bus voltage to 50% to 60% of the rated voltage; Second stage: Voltage establishment stage, increase the output power of the main starting power supply, gradually stabilize the bus voltage to the rated voltage range, start voltage closed-loop regulation, and suppress voltage overshoot; Level 3: Multi-source coordinated grid connection stage. Based on the availability of each power source, auxiliary power sources are connected sequentially. A power ramp-up strategy is adopted to achieve multi-source grid connection without impact and complete the multi-source power coordinated matching. Level 4: Load grading phase, loads are gradually applied in stages from light load to medium load to heavy load to avoid startup failure caused by instantaneous high power surges and to complete a full black start recovery of the system.

5. The multi-source cooperative black-start control method based on hierarchical startup according to claim 1, characterized in that, Step S4, the global current limiting and voltage regulation, includes tiered adaptive current limiting control, with the current limiting threshold increasing progressively at each startup stage: The first current limiting threshold is used during the pre-charging stage to limit the inrush current; the second current limiting threshold is used during the voltage establishment stage, and the second current limiting threshold is greater than the first current limiting threshold; the rated current limiting threshold is used during the multi-source grid connection and load connection stage to improve the startup speed while ensuring startup safety.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the multi-source cooperative black-start control method based on hierarchical startup as described in any one of claims 1-5.