A distributed new energy power grid-connected stable regulation system and self-adaptive control method

Abstract

The application discloses a kind of distributed new energy power grid-connected stable regulation systems and adaptive control methods, belong to new energy grid-connected control technical field.The application constructs hierarchical distributed collaborative grid-connected stable regulation architecture, fusion grid state real-time perception, impedance parameter online identification, multi-source power adaptive distribution, virtual inertia damping dynamic regulation, grid disturbance closed-loop suppression and fault-tolerant collaborative regulation core module.Accurate working condition acquisition and local primary regulation of each distributed new energy power monomer are realized by bottom layer terminal unit, regional multi-source power coupling decoupling and collaborative scheduling are completed by middle layer cluster coordination unit, and global scheduling unit realizes grid-connected state overall optimization and stable threshold adaptive update.The application solves the problem of poor system stability, control lag and insufficient collaboration caused by high penetration rate distributed new energy grid connection, greatly improves the adaptability of distributed new energy grid connection, disturbance rejection capability and grid operation stability.

Description

Technical Field

[0001] This invention relates to the field of new energy grid connection control technology, and in particular to a distributed new energy power grid connection stability regulation system and adaptive control method. Background Technology

[0002] Distributed renewable energy generation technology, with its advantages of being clean and low-carbon, being consumed locally, and being flexibly deployed, has become a core component of the energy supply in new power systems. However, the dispersed grid connection of a large number of distributed renewable energy power units poses significant challenges to the voltage stability, frequency stability, power stability, and power quality of traditional distribution networks.

[0003] Existing distributed renewable energy grid-connected control technologies are mainly divided into two categories: centralized control and fixed parameter droop control. Among them, centralized control relies on the upper-level dispatch center to uniformly issue control commands, and all distributed units passively execute the dispatch strategy. This has problems such as high communication latency, high risk of single point of failure, and poor scalability. When the scale of distributed renewable energy access expands and the number of grid connection points increases, it is very easy to experience command scheduling lag and asynchronous adjustment of multiple units, which can lead to regional power oscillations and cannot be adapted to large-scale distributed energy cluster grid connection scenarios.

[0004] Fixed-parameter droop control is currently the most commonly used local control strategy for distributed grid connection. It achieves autonomous power allocation by preset fixed active-frequency and reactive-voltage droop coefficients. However, this technology has significant drawbacks: First, the fixed droop parameters cannot adapt to complex operating conditions such as dynamic changes in grid impedance, random load fluctuations, and sudden changes in renewable energy output. Under strong grid conditions, the adjustment accuracy is insufficient, and under weak grid conditions, overshoot, oscillations, or even system instability are likely to occur. Second, independent control of multiple distributed renewable energy units lacks a clustered collaborative adjustment mechanism. Each unit responds independently to grid disturbances, easily leading to reactive power contention and active power adjustment conflicts, exacerbating grid power fluctuations. Third, it lacks online grid status perception and parameter adaptive iteration capabilities, making it unable to accurately identify grid strength and disturbance types, resulting in extremely poor suppression of transient power surges, voltage drops, frequency shifts, and other disturbances. Fourth, the lack of a hierarchical fault-tolerant control mechanism means that a single renewable energy unit failure or communication interruption can easily trigger regional grid connection system control failure, reducing overall grid stability.

[0005] Existing improved adaptive control technologies only fine-tune parameters for single voltage or frequency indicators, failing to achieve comprehensive coupled perception of grid status, power coupling, disturbance characteristics, and operating conditions. This single control dimension cannot solve core problems such as multi-source coupled disturbances, instability during operating condition switching, and imbalance in multi-unit coordination. Furthermore, existing technologies lack differentiated control strategies for various operating conditions, including strong grids, weak grids, sudden load changes, fluctuations in renewable energy output, equipment failures, and communication anomalies. This results in poor adaptability and makes it difficult to meet the stable regulation requirements of highly penetrated distributed renewable energy clusters connected to the grid.

[0006] In summary, existing distributed renewable energy grid-connected control technologies suffer from technical defects such as poor adaptability to operating conditions, insufficient multi-source coordination, weak disturbance suppression capability, poor fault tolerance, and low dynamic control accuracy. They cannot meet the grid-connected stable operation requirements of large-scale, highly decentralized, and multi-type distributed renewable energy power under the new power system. Therefore, this invention proposes a layered distributed coordination, multi-parameter adaptive iteration, multi-condition precise adaptation, strong anti-disturbance, and high fault tolerance grid-connected stable regulation system and adaptive control method. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a distributed new energy power grid-connected stability regulation system and adaptive control method. This invention solves the technical problems in existing technologies, such as poor adaptability of distributed new energy grid-connected operating conditions, conflicts in multi-unit coordinated regulation, lag in grid disturbance suppression, high risk of instability in weak grids, fixed control parameters, and insufficient fault tolerance. It achieves hierarchical coordinated grid-connected regulation of multiple types of distributed new energy power units, real-time adaptive iteration of control parameters, differentiated and precise control under multiple operating conditions, and comprehensive suppression of grid-connected disturbances, thereby significantly improving the grid-connected stability, power quality, and fault-tolerant operation capability of distributed new energy.

[0008] A distributed new energy power grid-connected stability regulation system includes a three-layer distributed collaborative architecture consisting of a top-down global scheduling layer, a cluster coordination layer, and a terminal execution layer. The global scheduling layer is deployed in the regional power grid dispatch center and includes a network-wide status monitoring unit, a stability threshold update unit, a global optimization scheduling unit, and a data storage and analysis unit. It is used to realize comprehensive perception of the network-wide grid connection status, adaptive update of stability thresholds, global power optimization scheduling, and operation data storage and analysis. The cluster coordination layer is deployed at the gateway of the distributed new energy cluster in each area and corresponds to multiple terminal execution layer units, including regional status perception unit, multi-source power decoupling unit, collaborative power allocation unit, regional disturbance identification unit, and communication fault tolerance unit, which are used to realize refined perception of regional operating conditions, multi-source power coupling and decoupling, regional power collaborative allocation, disturbance identification and communication fault tolerance control. The terminal execution layer consists of several distributed new energy grid-connected regulation terminals. Each terminal is matched one-to-one with a distributed new energy power unit, including a real-time operating condition acquisition unit, a grid impedance online identification unit, an adaptive parameter adjustment unit, a local disturbance suppression unit, a grid-connected execution unit, and a fault protection unit. It is used to realize the acquisition of operating conditions of individual equipment, grid impedance identification, adaptive adjustment of control parameters, local disturbance suppression, and fault safety protection. The system is equipped with an adaptive control module, a disturbance closed-loop suppression module, a multi-condition switching module, and a fault-tolerant collaborative module, which are used to realize real-time iteration of control parameters, hierarchical disturbance suppression, multi-condition strategy switching, and fault-tolerant collaborative operation.

[0009] Furthermore, the network-wide status monitoring unit collects real-time network topology information, grid impedance parameters, bus voltage, system frequency, network-wide load data, total output of distributed new energy clusters in each region, and grid-connected power quality parameters; the stability threshold update unit adaptively and iteratively updates the core control thresholds for voltage, frequency, power fluctuation, and harmonic limits based on the penetration rate of new energy, load fluctuation amplitude, and grid impedance characteristics.

[0010] Furthermore, the multi-source power decoupling unit constructs a power coupling mathematical model for multiple types of new energy power units in the area, and eliminates cross-coupling interference of active and reactive power regulation through a feedforward decoupling algorithm; the collaborative power allocation unit combines global scheduling instructions with the regulation capabilities of each terminal unit, and uses a particle swarm optimization algorithm to dynamically allocate the active and reactive power regulation power of each unit.

[0011] Furthermore, the online grid impedance identification unit adopts a high-frequency small-signal injection method to identify the equivalent grid impedance at the grid connection point in real time, and combines the system short-circuit ratio to complete the classification of strong grid, medium-strong grid, and weak grid operating conditions; the adaptive parameter adjustment unit dynamically iterates the droop control coefficient, virtual inertia coefficient, damping coefficient, and power limiting parameter.

[0012] A distributed new energy power grid-connected adaptive control method, applied to a grid-connected stability regulation system, includes the following steps: S1. Real-time perception of the entire network status and adaptive update of stability thresholds: Collect data on the entire network topology, system frequency, bus voltage, load, new energy penetration rate, and power quality, and adaptively iteratively update the core control thresholds for grid connection stability. S2. Online identification of grid impedance and classification of operating conditions: The grid impedance is identified online by high-frequency small signal injection method. The grid is classified into strong and weak operating conditions by combining the short-circuit ratio. The operating conditions are classified into steady state, small disturbance, large disturbance, and fault abnormal operation by combining the output and load fluctuations. S3. Multi-source power coupling decoupling and regional power collaborative allocation: Construct a multi-source power coupling model, eliminate power coupling interference through a feedforward decoupling algorithm, and dynamically allocate the power regulation of each distributed new energy unit based on global scheduling instructions; S4. Hierarchical Adaptive Parameter Iterative Control: Based on the power grid operating conditions and disturbance levels, the control parameters of the global scheduling layer, cluster coordination layer, and terminal execution layer are optimized respectively to achieve dynamic iteration of inertia, damping, and droop coefficient. S5. Multi-condition differentiated adaptive stability control: Dynamically switches exclusive control strategies for strong grid steady state, weak grid, load / output fluctuation, and fault abnormal conditions to achieve precise regulation across all operating conditions. S6. Disturbance graded closed-loop suppression: Grid connection disturbances are divided into three levels: steady-state small amplitude, transient moderate, and sudden severe. Layered closed-loop suppression is implemented to quickly suppress grid connection disturbances. S7. Fault and Communication Anomaly Tolerance Coordination: Monitors communication and equipment operating status, switches to local autonomous control mode when communication is abnormal, and compensates for power gaps by other units when a single unit fails, so as to achieve fault-tolerant and stable operation of the system.

[0013] Furthermore, in step S2, a short-circuit ratio greater than 3 is determined to be a strong power grid condition, a short-circuit ratio of 1.5 to 3 is determined to be a medium-strong power grid condition, and a short-circuit ratio less than 1.5 is determined to be a weak power grid condition; the output and load change rate is less than 5% for a steady-state condition, 5% to 15% for a small disturbance condition, and greater than 15% for a large disturbance condition.

[0014] Furthermore, in step S4, the virtual inertia is reduced and the power response speed is improved under strong grid conditions; the virtual inertia and damping coefficient are increased and the droop coefficient is optimized under weak grid conditions to suppress voltage frequency oscillations; and the parameters are rapidly iterated under disturbance conditions to enhance disturbance suppression capabilities.

[0015] Furthermore, in step S5, high-precision power point tracking control is adopted for the steady-state operation of the strong power grid; inertia damping adaptive control + reactive power and voltage coordinated support strategy is adopted for the weak power grid operation; fuzzy PID adaptive adjustment is adopted for the small fluctuation operation; model prediction and advance control is adopted for the large disturbance operation; and current limiting, voltage support and islanding fault-tolerant control are adopted for the fault operation.

[0016] Furthermore, in step S6, steady-state minor disturbances are mitigated through local reactive power compensation and harmonic correction at the terminal; transient moderate disturbances are smoothed through cluster-level collaborative power regulation; and sudden severe disturbances are rapidly restored to grid steady state through a multi-level linkage approach of global coordination and terminal parameter iteration.

[0017] The beneficial effects of this invention are as follows: 1. This invention adopts a three-layer distributed collaborative architecture, which overcomes the shortcomings of traditional centralized control, such as high communication latency, high risk of single point of failure, and poor scalability. At the same time, it solves the problems of poor coordination and regulation conflicts of traditional local independent control of multiple units, and realizes all-round grid-connected control with overall network planning, regional coordination, and precise terminal control. It is suitable for large-scale, highly decentralized, and multi-type distributed new energy cluster grid-connected scenarios.

[0018] 2. This invention enables online identification of grid impedance, automatic classification of operating conditions, and full-dimensional adaptive iteration of control parameters. It eliminates the technical drawbacks of traditional fixed parameter control and can dynamically switch control strategies according to multiple operating conditions such as strong grid, weak grid, load fluctuation, output disturbance, and fault anomaly. This significantly improves the system's adaptability to operating conditions and the accuracy of dynamic adjustment, fundamentally solving the industry problem of instability and oscillation of new energy grid connection under weak grid conditions.

[0019] 3. This invention sets up a multi-source power decoupling and collaborative allocation mechanism, which effectively eliminates the coupling interference of active and reactive power regulation of multiple types of distributed new energy units, solves the problems of multi-unit regulation competition and power oscillation, and improves the stability of multi-source grid connection and the capacity for new energy consumption.

[0020] 4. This invention adopts a hierarchical closed-loop disturbance suppression technology to achieve layered, rapid, and precise suppression of grid-connected disturbances of different levels, effectively improving the stability of grid-connected voltage, frequency, and power, reducing harmonic distortion rate, and comprehensively improving the grid-connected power quality; at the same time, it introduces virtual inertia and damping adaptive control to significantly improve the system's ability to resist transient disturbances.

[0021] 5. This invention has complete fault-tolerant control capabilities for communication anomalies, equipment failures, and power grid failures. It can achieve uninterrupted and stable operation of the system under local faults, greatly improving the reliability and fault tolerance of the grid-connected system and adapting to complex power grid operation scenarios.

[0022] 6. This invention has a high degree of automation, can achieve adaptive control under all working conditions without manual intervention, has a fast response speed, high adjustment accuracy, strong engineering adaptability, is compatible with various existing distributed new energy grid-connected equipment, has low transformation cost, and has extremely strong engineering promotion value and market application prospects. Attached Figure Description

[0023] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart illustrating the overall control process of the three-layer architecture of the system of this invention. Figure 2 This is a flowchart of the multi-condition disturbance hierarchical suppression and power coordinated regulation of the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0026] The first aspect of this invention provides a distributed new energy power grid-connected stability regulation system, which adopts a three-layer distributed collaborative architecture, including a global scheduling layer, a cluster coordination layer, and a terminal execution layer from top to bottom. Each layer is modularly deployed, operates independently, and works in coordination to achieve overall optimization of the grid-connected status, regional collaborative regulation, and precise execution at the terminal. The system is equipped with a status perception module, a parameter identification module, an adaptive control module, a power decoupling module, a disturbance suppression module, and a fault-tolerant control module to comprehensively support grid-connected stability adaptive regulation.

[0027] 1. Global Scheduling Layer The global dispatch layer is deployed in the regional power grid dispatch center and is the top-level control unit of the system. It mainly includes the network-wide status monitoring unit, stability threshold update unit, global optimization dispatch unit, and data storage and analysis unit.

[0028] The network-wide status monitoring unit is used to collect real-time information on the entire distribution network topology, grid impedance parameters, bus voltage, system frequency, network-wide load data, total output of distributed renewable energy clusters in each region, and grid-connected power quality parameters, achieving comprehensive real-time perception of the network's grid-connected operation status. The stability threshold update unit is used to adaptively and iteratively update core control thresholds such as system voltage stability threshold, frequency stability threshold, power fluctuation allowable threshold, and harmonic limits based on the network's operating conditions, renewable energy penetration rate, and load fluctuation amplitude, adapting to different operating scenarios. The global optimization scheduling unit is used to coordinate the operating status of renewable energy clusters in each region, generate global power optimization scheduling instructions, and issue them to the coordination layers of each cluster, achieving dynamic matching between the network's renewable energy output and grid load. The data storage and analysis unit is used to store historical operating data, control parameters, disturbance data, and adjustment records, providing data support for iterative optimization of control parameters, operating condition prediction, and fault tracing.

[0029] 2. Cluster Coordination Layer The cluster coordination layer is deployed at the gateway of the distributed new energy cluster in each area, corresponding to multiple terminal execution layer units. It is the core of regional collaborative control and mainly includes regional status perception unit, multi-source power decoupling unit, collaborative power allocation unit, regional disturbance identification unit, and communication fault tolerance unit.

[0030] The regional status sensing unit is used to accurately collect the output power, grid connection point voltage and current, operating temperature, and equipment conditions of each distributed new energy power unit within the area. It also collects data on load fluctuations and local grid impedance fluctuations to achieve refined regional status sensing. The multi-source power decoupling unit is used to construct a power coupling decoupling model based on the power coupling characteristics of various types of new energy power units such as photovoltaic, wind power, and energy storage within the area. This eliminates mutual interference between active and reactive power regulation of multiple units and resolves multi-source regulation conflicts. The collaborative power allocation unit receives data from the global scheduling layer. The system optimizes scheduling instructions, combining the operating margin, output characteristics, and regulation capabilities of each terminal unit in the area, and dynamically allocates active and reactive power regulation instructions to each terminal unit to achieve optimal regional power coordination and regulation. The regional disturbance identification unit is used to identify the types and levels of disturbances such as voltage drops, frequency deviations, power oscillations, sudden load changes, and sudden changes in new energy output in the area in real time, and triggers the corresponding level of disturbance suppression strategy. The communication fault tolerance unit is used to monitor the communication status between levels and units, and automatically switches to the local autonomous control mode when there is communication delay or communication interruption to ensure that regional control does not fail.

[0031] 3. Terminal Execution Layer The terminal execution layer consists of several distributed new energy grid-connected regulation terminals. Each terminal corresponds to a distributed new energy power unit (photovoltaic inverter, wind power converter, energy storage converter, etc.) for performing precise local regulation. It mainly includes a real-time operating condition acquisition unit, a grid impedance online identification unit, an adaptive parameter adjustment unit, a local disturbance suppression unit, a grid-connected execution unit, and a fault protection unit.

[0032] The real-time operating condition acquisition unit is used to collect high-frequency data on the output active power, reactive power, instantaneous voltage at the grid connection point, instantaneous current, grid connection frequency, and harmonics of individual new energy power units. The sampling frequency is no less than 10kHz to ensure the real-time performance and accuracy of the operating condition data acquisition. The online grid impedance identification unit uses a high-frequency small-signal injection method to achieve real-time online identification of distribution network line impedance and equivalent grid impedance, accurately determining the strength of the grid and providing a basis for adaptive parameter adjustment. The adaptive parameter adjustment unit dynamically adjusts parameters based on grid status, disturbance level, and operating conditions. Iterative droop control coefficient, virtual inertia, damping coefficient, and power limiting parameters replace traditional fixed parameter control; the local disturbance suppression unit is used to perform voltage closed-loop regulation, frequency damping regulation, and power smoothing regulation for individual grid-connected disturbances, quickly suppressing transient power surges and power quality disturbances; the grid-connected execution unit is used to accurately execute active and reactive power regulation commands to achieve dynamic adaptation of grid-connected power; the fault protection unit is used to monitor equipment overvoltage, overcurrent, overload, islanding, and other fault states, triggering fault-tolerant regulation and safety protection to ensure the safety of equipment and the power grid.

[0033] 4. Core Functional Modules of the System This system is based on a three-layer architecture and integrates an adaptive control module, a disturbance closed-loop suppression module, a multi-condition switching module, and a fault-tolerant coordination module to achieve stable regulation of the system under all operating conditions. The adaptive control module enables real-time adaptive iterative optimization of control parameters; the disturbance closed-loop suppression module enables graded suppression of steady-state disturbances, transient disturbances, and sudden disturbances; the multi-condition switching module enables adaptive switching of strategies for strong power grid, weak power grid, load fluctuations, sudden output changes, and fault conditions; and the fault-tolerant coordination module enables coordinated fault-tolerant operation of the system in the event of communication anomalies, equipment failures, or single-unit failures.

[0034] The second aspect of this invention provides a distributed new energy power grid-connected stability adaptive control method, applied to the above-mentioned grid-connected stability regulation system, including the following core steps: full network state perception and threshold adaptive update, online identification of grid impedance and operating condition determination, multi-source power decoupling and collaborative allocation, hierarchical adaptive parameter iterative regulation, multi-operating condition differentiated stability control, disturbance hierarchical closed-loop suppression, and fault-tolerant collaborative regulation.

[0035] Step 1: Real-time perception of the entire network status and adaptive updating of stable thresholds The global scheduling layer's network status monitoring unit collects real-time data on the entire network's topology, system frequency, bus voltage, total network load, total renewable energy penetration rate, and grid-connected power quality. Combining renewable energy penetration rate ranges, load fluctuation coefficients, and grid impedance characteristics, a big data fitting algorithm is used to adaptively and iteratively update core stability parameters such as voltage stability range, frequency stability range, power fluctuation threshold, harmonic control threshold, and power regulation response threshold. This enables the regulation thresholds to dynamically adapt to the grid's operating conditions, avoiding over- or under-regulation caused by fixed thresholds.

[0036] Step 2: Online identification of power grid impedance and determination of power grid operating condition classification The terminal execution layer injects a weak high-frequency detection signal into the grid connection point through the high-frequency small signal injection method, collects the response signal, calculates the equivalent grid resistance and inductance parameters, and completes online real-time identification of grid impedance. Based on the identified grid impedance value and combined with the system short-circuit ratio index, the grid operating conditions are divided into three categories: strong grid operating conditions, medium-strong grid operating conditions, and weak grid operating conditions. At the same time, combined with the fluctuation amplitude of new energy output and the load change rate, the operating conditions are divided into four states: steady-state operation, small disturbance, large disturbance, and fault abnormality, providing a basis for subsequent differentiated control strategy switching.

[0037] Step 3: Decoupling of multi-source power coupling and coordinated regional power allocation The cluster coordination layer constructs a multi-source power coupling mathematical model based on the output characteristics, regulation response speed, and power coupling relationships of multiple types of distributed new energy units within the region. It eliminates cross-coupling interference between active and reactive power through a feedforward decoupling algorithm, solving the power oscillation problem caused by synchronous regulation of multiple units. Combined with the power optimization instructions issued by the global dispatch, the active and reactive power regulation power of each unit is dynamically allocated according to the current output, maximum regulation margin, and equipment operating status of each terminal unit, using a particle swarm optimization algorithm. This achieves the optimal coordinated output of multiple new energy units within the region, balancing the power supply and demand of the power grid.

[0038] Step 4: Hierarchical Adaptive Parameter Iterative Control Based on the results of grid operating condition classification, a three-layer architecture with adaptive parameter optimization is implemented: the global scheduling layer optimizes the power scheduling weight and stability threshold of the entire network; the cluster coordination layer optimizes the regional power allocation coefficient and collaborative adjustment step size; and the terminal execution layer dynamically iterates the droop control coefficient, virtual inertia coefficient, damping coefficient, and power limiting parameters. Under strong grid operating conditions, the virtual inertia is reduced and the adjustment response speed is improved to ensure that power quickly tracks the scheduling command; under weak grid operating conditions, the virtual inertia and damping coefficient are increased and the droop coefficient is optimized to suppress voltage frequency oscillations and improve the grid connection stability of weak grids; under steady-state operating conditions, parameters are finely adjusted to ensure grid connection stability; and under disturbance operating conditions, parameters are rapidly iterated to enhance disturbance suppression capabilities.

[0039] Step 5: Multi-condition differentiated adaptive stability control Specialized control strategies are implemented for different operating conditions: 1) Strong grid steady-state conditions: High-precision power tracking control is adopted, with fixed small-scale optimization parameters to ensure maximum absorption of new energy output while maintaining stable grid-connected power quality; 2) Weak grid conditions: Inertial damping adaptive control + reactive power and voltage coordinated support strategy is adopted to improve grid voltage support capability and suppress low-frequency oscillations and voltage drift; 3) Small load / output fluctuation conditions: Fuzzy PID adaptive adjustment is adopted, with small-scale iterative control parameters to smooth power fluctuations; 4) Large load / output disturbance conditions: Model predictive control is adopted to predict power change trends in advance, achieve proactive adjustment, and suppress transient power surges; 5) Fault and abnormal conditions: Fault current limiting, voltage support, and islanding fault-tolerant control are triggered to ensure that the system does not disconnect from the grid or become unstable.

[0040] Step 6: Disturbance hierarchical closed-loop suppression control Based on the disturbance identification results, grid-connected disturbances are divided into three levels: steady-state minor disturbances, transient moderate disturbances, and sudden severe disturbances. A graded closed-loop suppression is implemented: steady-state minor harmonic disturbances and voltage deviation disturbances are accurately corrected by the local reactive power compensation and harmonic suppression modules at the terminal; transient power fluctuations and minor frequency offset disturbances are quickly smoothed by cluster-level coordinated power regulation; and sudden voltage drops, sudden power changes, and severe frequency oscillation disturbances are suppressed through global-level overall scheduling and rapid parameter iteration at the terminal, achieving multi-level linkage disturbance suppression and quickly restoring the grid to steady-state operation.

[0041] Step 7: Fault and Communication Anomaly Tolerance Coordination Adjustment The system monitors the hierarchical communication status and terminal equipment operation status in real time: when hierarchical communication delays or data packet loss occur, the cluster layer and terminal layer automatically switch to local autonomous control mode, breaking away from global command dependence and ensuring regional grid connection stability; when a single distributed new energy terminal fails to operate or its regulation fails, the cluster layer quickly redistributes regional power regulation tasks, with other normal terminals compensating for the power gap to avoid regional power imbalance; when a local power grid fails, the system automatically adjusts grid connection control parameters to reduce the scope of disturbance impact and achieve fault tolerance and uninterrupted stable operation. Example 1

[0042] This embodiment takes a distributed photovoltaic-storage grid-connected system in an industrial park as the application scenario. The scenario includes 12 sets of distributed photovoltaic units and 4 sets of energy storage units. Multiple types of new energy power units are distributed and connected to the grid in a decentralized manner. There are problems such as large output fluctuations, multi-source power coupling, large load peak-valley differences, and frequent occurrence of local weak grid conditions. The grid-connected stability regulation system and adaptive control method of the present invention are used for regulation.

[0043] 1. System Deployment and Implementation This embodiment of the system adopts a three-layer distributed architecture. The global scheduling layer is deployed in the park's power distribution dispatch center to realize the monitoring of the grid connection status of the entire park, the updating of stability thresholds, and the global power optimization. The cluster coordination layer is divided into two regional cluster control units according to the park's power distribution area, which respectively manage all photovoltaic and energy storage grid-connected terminals in the north and south areas, and are responsible for regional power decoupling, coordinated allocation, and regional disturbance identification. The terminal execution layer deploys 16 grid-connected regulation terminals, which are matched one-to-one with 12 photovoltaic units and 4 energy storage units to realize the acquisition of individual equipment operating conditions, impedance identification, local adaptive adjustment, and disturbance suppression.

[0044] The system's various modules work together to collect grid-connected operation data at high frequency, with the sampling frequency set to 10kHz to ensure real-time monitoring of operating conditions. The online grid impedance identification module uses a 150Hz high-frequency small signal injection method, which is free from harmonic interference and does not affect normal grid-connected operation. It updates impedance parameters every 20ms to determine the grid operating condition level in real time.

[0045] 2. Specific Adaptive Control Flow Step 1: Full-network operating condition perception and threshold update. The global scheduling layer collects real-time data on the park's power grid frequency, bus voltage, regional load, total photovoltaic and energy storage output, and grid impedance. Based on photovoltaic penetration rate and load fluctuation coefficient, it adaptively updates the voltage stability range to 0.97~1.03pu and the frequency stability range to 49.8~50.2Hz, and dynamically updates the allowable power fluctuation threshold to adapt to the different operating conditions of the park, such as high photovoltaic power generation during the day and low load at night.

[0046] Step 2: Classification and determination of power grid operating conditions. The terminal execution layer identifies the equivalent grid impedance of each grid connection point in real time and calculates the system short-circuit ratio. A short-circuit ratio greater than 3 is determined to be a strong power grid operating condition, between 1.5 and 3 is determined to be a medium-strong power grid operating condition, and less than 1.5 is determined to be a weak power grid operating condition. At the same time, based on the rate of change of renewable energy output and the load fluctuation amplitude, it distinguishes between steady-state, small disturbance, large disturbance, and fault abnormal operating conditions to achieve accurate classification of operating conditions.

[0047] Step 3: Multi-source power decoupling and collaborative allocation. The cluster coordination layer constructs a multi-source power coupling model for photovoltaic and energy storage. The feedforward decoupling algorithm eliminates the coupling interference between photovoltaic active power output fluctuations and energy storage reactive power regulation. Based on global power optimization instructions, and combined with the output capacity of each photovoltaic unit and the charging and discharging margin of the energy storage unit, the particle swarm optimization algorithm is used to dynamically allocate the active and reactive power regulation power of each unit. The photovoltaic unit prioritizes tracking the maximum power point output, while the energy storage unit undertakes the tasks of power smoothing, voltage support, and frequency regulation, achieving optimal grid connection for photovoltaic and energy storage collaboration.

[0048] Step 4: Layered Adaptive Parameter Iteration. Under strong grid conditions, the virtual inertia coefficient is reduced to improve the power response speed of the photovoltaic-storage unit and quickly track load changes. Under weak grid conditions, the virtual inertia and damping coefficients are automatically increased to optimize the droop control coefficient and suppress voltage frequency oscillations. Under steady-state conditions, parameters are finely adjusted to ensure stable grid connection. Under disturbance conditions, parameters are rapidly iterated to enhance disturbance suppression capabilities. Compared to traditional fixed parameter control, after adaptive parameter iteration in this embodiment, the system frequency oscillation amplitude is reduced by 75%, and the voltage deviation is reduced by 82%.

[0049] Step 5: Differentiated Control for Multiple Operating Conditions. During the daytime, when photovoltaic power generation is high and load is stable, a high-precision power point tracking control is employed to maximize photovoltaic absorption. In the evening, when photovoltaic output declines and load surges cause significant disturbances, model-based predictive control is used to anticipate power gaps and control rapid discharge of energy storage to replenish power, smoothing power fluctuations. During cloudy or rainy weather, when photovoltaic output fluctuates randomly, a fuzzy PID adaptive adjustment is used to fine-tune control parameters in real time, mitigating short-term power fluctuations. In the event of weak grid conditions due to localized line maintenance, a reactive power and voltage coordinated support strategy is activated to improve grid connection voltage stability and prevent grid disconnection due to instability.

[0050] Step 6: Disturbance Classification and Closed-Loop Suppression. For small harmonics and steady-state disturbances caused by grid connection, precise management is achieved through local reactive power compensation and harmonic correction modules at the terminal. For transient power and frequency disturbances caused by load fluctuations and output offsets, rapid suppression is achieved through cluster-level collaborative power regulation. For severe disturbances such as sudden voltage drops and power surges, a multi-level linkage suppression strategy involving global overall scheduling and rapid parameter iteration at the terminal is triggered to quickly restore grid steady state, reducing the disturbance recovery time from the traditional 200ms to less than 50ms.

[0051] Step 7: Fault-Tolerant Coordinated Regulation. During operation, when the output of a single photovoltaic unit drops sharply due to fluctuations in sunlight, the cluster layer quickly redistributes the energy storage output to compensate for the power shortfall and avoid power fluctuations across the entire network. When regional communication is briefly interrupted, the cluster and terminals automatically switch to local autonomous control mode, independently completing stable regulation without global commands. When a single energy storage unit fails and shuts down, the remaining energy storage units automatically expand their capacity to adjust the margin, ensuring regional grid connection stability and achieving seamless operation during local faults.

[0052] 3. Implementation effect verification This embodiment has been running continuously for 3 months in a distributed photovoltaic-storage grid-connected system in an industrial park. The operating data shows that compared with traditional grid-connected control strategies, the present invention can achieve adaptive and stable regulation under all operating conditions, improve the grid-connected voltage qualification rate, improve the frequency stability qualification rate, reduce the power fluctuation amplitude, and improve the renewable energy absorption rate; there is no oscillation or grid disconnection fault under weak grid conditions, and the system's anti-disturbance capability, operating condition adaptability, and fault-tolerant operation capability are greatly improved, fully meeting the stable operation requirements of high-penetration distributed renewable energy cluster grid connection.

[0053] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A distributed new energy power grid-connected stability regulation system, characterized in that: It includes a three-layer distributed collaborative architecture consisting of a top-down global scheduling layer, a cluster coordination layer, and a terminal execution layer; The global scheduling layer is deployed in the regional power grid dispatch center and includes a network-wide status monitoring unit, a stability threshold update unit, a global optimization scheduling unit, and a data storage and analysis unit. It is used to realize comprehensive perception of the network-wide grid connection status, adaptive update of stability thresholds, global power optimization scheduling, and operation data storage and analysis. The cluster coordination layer is deployed at the gateway of the distributed new energy cluster in each area and corresponds to multiple terminal execution layer units, including regional status perception unit, multi-source power decoupling unit, collaborative power allocation unit, regional disturbance identification unit, and communication fault tolerance unit, which are used to realize refined perception of regional operating conditions, multi-source power coupling and decoupling, regional power collaborative allocation, disturbance identification and communication fault tolerance control. The terminal execution layer consists of several distributed new energy grid-connected regulation terminals. Each terminal is matched one-to-one with a distributed new energy power unit, including a real-time operating condition acquisition unit, a grid impedance online identification unit, an adaptive parameter adjustment unit, a local disturbance suppression unit, a grid-connected execution unit, and a fault protection unit. It is used to realize the acquisition of operating conditions of individual equipment, grid impedance identification, adaptive adjustment of control parameters, local disturbance suppression, and fault safety protection. The system is equipped with an adaptive control module, a disturbance closed-loop suppression module, a multi-condition switching module, and a fault-tolerant collaborative module, which are used to realize real-time iteration of control parameters, hierarchical disturbance suppression, multi-condition strategy switching, and fault-tolerant collaborative operation.

2. The distributed new energy power grid-connected stability regulation system according to claim 1, characterized in that, The network status monitoring unit collects real-time information on the network topology, grid impedance parameters, bus voltage, system frequency, network load data, total output of distributed new energy clusters in each region, and grid-connected power quality parameters. The stable threshold update unit adaptively and iteratively updates the core control thresholds for voltage, frequency, power fluctuation, and harmonic limits based on the penetration rate of new energy sources, load fluctuation amplitude, and grid impedance characteristics.

3. The distributed new energy power grid-connected stability regulation system according to claim 1, characterized in that, The multi-source power decoupling unit constructs a power coupling mathematical model for various types of new energy power units in the area, and eliminates cross-coupling interference between active and reactive power regulation through a feedforward decoupling algorithm; the collaborative power allocation unit combines global scheduling instructions with the regulation capabilities of each terminal unit, and uses a particle swarm optimization algorithm to dynamically allocate the active and reactive power regulation power of each unit.

4. The distributed new energy power grid-connected stability regulation system according to claim 1, characterized in that, The online grid impedance identification unit uses a high-frequency small-signal injection method to identify the equivalent grid impedance at the grid connection point in real time, and combines the system short-circuit ratio to classify the operating conditions of strong grid, medium-strong grid, and weak grid; the adaptive parameter adjustment unit dynamically iterates the droop control coefficient, virtual inertia coefficient, damping coefficient, and power limiting parameter.

5. A distributed new energy power grid-connected adaptive control method, applied to the grid-connected stability regulation system according to any one of claims 1-4, characterized in that, Includes the following steps: S1. Real-time perception of the entire network status and adaptive update of stability thresholds: Collect data on the entire network topology, system frequency, bus voltage, load, new energy penetration rate, and power quality, and adaptively iteratively update the core control thresholds for grid connection stability. S2. Online identification of grid impedance and classification of operating conditions: The grid impedance is identified online by high-frequency small signal injection method. The grid is classified into strong and weak operating conditions by combining the short-circuit ratio. The operating conditions are classified into steady state, small disturbance, large disturbance, and fault abnormal operation by combining the output and load fluctuations. S3. Multi-source power coupling decoupling and regional power collaborative allocation: Construct a multi-source power coupling model, eliminate power coupling interference through a feedforward decoupling algorithm, and dynamically allocate the power regulation of each distributed new energy unit based on global scheduling instructions; S4. Hierarchical Adaptive Parameter Iterative Control: Based on the power grid operating conditions and disturbance levels, the control parameters of the global scheduling layer, cluster coordination layer, and terminal execution layer are optimized respectively to achieve dynamic iteration of inertia, damping, and droop coefficient. S5. Multi-condition differentiated adaptive stability control: Dynamically switches exclusive control strategies for strong grid steady state, weak grid, load / output fluctuation, and fault abnormal conditions to achieve precise regulation across all operating conditions. S6. Disturbance graded closed-loop suppression: Grid connection disturbances are divided into three levels: steady-state small amplitude, transient moderate, and sudden severe. Layered closed-loop suppression is implemented to quickly suppress grid connection disturbances. S7. Fault and Communication Anomaly Tolerance Coordinated Adjustment: Monitors communication and equipment operating status, switches to local autonomous control mode when communication is abnormal, and compensates for power gaps through other units when a single unit fails, so as to achieve fault-tolerant and stable operation of the system.

6. The distributed new energy power grid-connected adaptive control method according to claim 5, characterized in that, In step S2, a short-circuit ratio greater than 3 is determined to be a strong power grid condition, a short-circuit ratio of 1.5 to 3 is determined to be a medium-strong power grid condition, and a short-circuit ratio less than 1.5 is determined to be a weak power grid condition; the output and load change rate is less than 5% for steady-state conditions, 5% to 15% for small-amplitude disturbance conditions, and greater than 15% for large-amplitude disturbance conditions.

7. The distributed new energy power grid-connected adaptive control method according to claim 5, characterized in that, In step S4, the virtual inertia is reduced and the power response speed is improved under strong grid conditions; the virtual inertia and damping coefficient are increased and the droop coefficient is optimized under weak grid conditions to suppress voltage frequency oscillations; and the parameters are rapidly iterated under disturbance conditions to enhance disturbance suppression capabilities.

8. The distributed new energy power grid-connected adaptive control method according to claim 5, characterized in that, In step S5, high-precision power point tracking control is adopted for the steady-state operation of the strong power grid; inertia damping adaptive control + reactive power and voltage coordinated support strategy is adopted for the weak power grid operation; fuzzy PID adaptive adjustment is adopted for the small fluctuation operation; model prediction and advance control is adopted for the large disturbance operation; and current limiting, voltage support and islanding fault-tolerant control are adopted for the fault operation.

9. The distributed new energy power grid-connected adaptive control method according to claim 5, characterized in that, In step S6, steady-state minor disturbances are mitigated through local reactive power compensation and harmonic correction at the terminal; transient moderate disturbances are smoothed through cluster-level collaborative power regulation; and sudden severe disturbances are rapidly restored to grid steady state through a multi-level linkage approach of global coordination and terminal parameter iteration.

Images