Multi-source coordinated regulation method for grid-connection scenario with high proportion of new energy

Abstract

The application discloses a multi-source collaborative regulation method for grid-constructed field stations in a high-proportion new energy grid-connected scene and relates to the technical field of new energy power system control. The method comprises the following steps: acquiring multi-source heterogeneous data and constructing a unified time scale dataset; establishing a dynamic equivalent model of the field station, extracting virtual inertia, damping and short-circuit ratio characteristic parameters; dividing three types of scenes, namely, steady-state economic dispatch, transient frequency / voltage support and wide-frequency oscillation suppression, according to the grid state, and respectively constructing a multi-source power economic allocation model, a frequency collaborative response model, a voltage layered collaborative regulation model and a wide-frequency oscillation adaptive suppression model; and obtaining optimal power instructions and issuing them for execution through dynamic coupling constraint set and multi-objective optimization algorithm collaborative solution. The application realizes multi-objective collaborative regulation of wind power, photovoltaic power, energy storage and flexible load in the grid-constructed field station, takes into account economy, transient support and oscillation suppression, and improves the stability and operation efficiency of the high-proportion new energy grid-connected system.

Description

Technical Field

[0001] This invention relates to the field of new energy power system control technology, specifically a multi-source coordinated control method for grid-connected power plants in scenarios with a high proportion of new energy grid connection. Background Technology

[0002] With the continuous increase in the penetration rate of new energy power generation such as wind power and photovoltaics, the power system is gradually showing the "dual high" characteristics of "high proportion of renewable energy and high proportion of power electronic equipment," resulting in a significant decrease in system inertia and damping levels, and a reduction in disturbance rejection capability. Traditional grid-forming control based on phase-locked loop synchronization essentially exhibits current source characteristics, relying on the grid to provide voltage and frequency references, and gradually reveals its instability defects under weak grid conditions. Grid-forming control technology simulates the dynamic characteristics of synchronous generators by controlling power electronic converters, exhibiting voltage source characteristics. It can autonomously establish and maintain grid voltage and frequency, and even achieve black start, making it one of the key technologies for adapting to the high proportion of new energy integration.

[0003] However, existing technologies still have the following shortcomings:

[0004] First, current research mostly focuses on optimizing control strategies for single grid-connected converters or single types of resources. While there are schemes in the literature that involve integrated control of grid-connected energy storage and new energy power plants, as well as collaborative control methods for grid-connected and grid-connected energy storage systems, a complete technical framework for the systematic and collaborative regulation of multiple types of resources (wind power, photovoltaic, energy storage, flexible loads, etc.) within grid-connected power plants is lacking.

[0005] Second, existing collaborative control methods typically target only a single regulation objective, such as focusing solely on grid frequency regulation scenarios or power fluctuation mitigation scenarios. In terms of multi-objective collaborative control of grid-connected converters, existing research mainly addresses transient synchronization stability and current stress constraints under asymmetric faults, failing to comprehensively consider the collaborative needs of multiple scenarios and objectives, such as steady-state economic dispatch, transient frequency / voltage support, and broadband oscillation suppression.

[0006] Third, broadband oscillations are a prominent issue in scenarios with a high proportion of renewable energy connected to the grid. Existing research shows that grid-based control modes have better stability under weak grid conditions, but under strong grid conditions, the active power loop damping is very small, making the output power prone to oscillations. Although there have been practical explorations of suppressing broadband oscillations by configuring grid-based equipment, existing methods lack a technical solution that organically integrates the broadband oscillation suppression capability of grid-based control with multi-source coordinated regulation.

[0007] Fourth, existing methods mostly employ a centralized control architecture, which makes it difficult to balance the global optimality of regulation with the need for rapid local response, and it does not adequately consider the time synchronization and communication latency issues of multi-source heterogeneous data. Therefore, we propose a multi-source collaborative regulation method for grid-connected power plants in scenarios with a high proportion of renewable energy grid connection. Summary of the Invention

[0008] The technical problem to be solved by the present invention is to overcome the existing defects and provide a multi-source coordinated control method for grid-type power plants in high-proportion new energy grid-connected scenarios. It can overcome the technical defects of existing technologies, such as the lack of a systematic coordinated control framework for multi-type resources, the inability to take into account the control needs of multiple scenarios, and the lack of organic integration of broadband oscillation suppression and multi-source coordination. It can effectively solve the problems in the background technology.

[0009] To achieve the above objectives, the present invention provides the following technical solution: a multi-source coordinated control method for grid-connected power plants in high-proportion renewable energy grid-connected scenarios, wherein a grid-connected power plant refers to a renewable energy power generation cluster containing two or more grid-connected converters employing a grid-connected control strategy, characterized by comprising the following steps:

[0010] Step S1: Obtain the operating status data of each new energy power generation unit, energy storage unit and flexible load in the grid-type power station and the grid-side operating parameters, perform time synchronization preprocessing on the data, and construct a unified time-stamped dataset of multi-source heterogeneous data;

[0011] Step S2: Based on the unified time-scaled dataset, establish a dynamic equivalent model of the network-type power station, calculate the virtual inertia, equivalent damping coefficient and multi-node short-circuit ratio distribution of the entire power station, and obtain the equivalent characteristic parameter set of the power station;

[0012] Step S3: Based on the power grid operating status and the equivalent characteristic parameter set of the substations, the control scenarios are divided into steady-state economic operation scenarios, transient frequency / voltage support scenarios, and broadband oscillation suppression scenarios;

[0013] Step S4: For steady-state economic operation scenarios, with the goal of minimizing overall power plant operating losses and maximizing renewable energy absorption rate, construct a multi-source power economic allocation model;

[0014] Step S5: For transient frequency / voltage support scenarios, construct a frequency coordinated response model based on the virtual synchronous machine control strategy, and construct a voltage hierarchical coordinated control model based on reactive power and voltage sensitivity analysis;

[0015] Step S6: For wideband oscillation suppression scenarios, identify the oscillation frequency band based on impedance analysis, construct a wideband oscillation adaptive suppression model, and dynamically adjust the virtual impedance parameters of the grid-type converter;

[0016] Step S7: Based on the variable coupling relationship between the multi-source power economic allocation model, frequency coordinated response model, voltage hierarchical coordinated control model, and broadband oscillation adaptive suppression model, establish a dynamic coupling constraint set that includes active-reactive power coupling constraints and frequency-voltage-oscillation multi-objective conflict constraints; use a multi-objective optimization algorithm to solve the above four models in a coordinated manner under the dynamic coupling constraint set to obtain the optimal power command value of each control resource;

[0017] Step S8: The optimal power command value is sent to the actuators of each new energy power generation unit, energy storage unit and flexible load through the high-speed communication network to perform multi-source coordinated control, and the execution results are fed back in real time to trigger correction.

[0018] Furthermore, in step S1, the time synchronization preprocessing includes: using the Global Positioning System or BeiDou timing signal as a reference, the data from each channel are unified to the same time segment by interpolation resampling, and the transmission delay introduced by the communication link is dynamically compensated; the operating status data includes at least the grid-connected power, terminal voltage, and terminal current of each power generation unit, the state of charge and charging / discharging power of the energy storage unit, as well as the frequency, grid connection point voltage, and line power flow on the grid side.

[0019] Furthermore, in step S2, the specific method for establishing the dynamic equivalent model of the grid-type power station is as follows: based on the port-controlled Hamiltonian model of each grid-type converter, the entire power station is equivalent to a second-order inertia-damping dynamic model, and the equivalent virtual inertia and equivalent damping coefficient of the power station are identified online using the recursive least squares method with forgetting factor; at the same time, according to the impedance network relationship between each grid-connected node in the power station, the short-circuit ratio of each grid-connected node is calculated to form a comprehensive characteristic parameter set including the distribution of inertia, damping and short-circuit ratio.

[0020] Furthermore, in step S3, the criteria for classifying each control scenario include: when the absolute value of the grid frequency deviation exceeds the set dead zone value and the frequency change rate is greater than the set threshold, it is determined to be a transient frequency support scenario; when the absolute value of the grid connection point voltage deviation exceeds the set dead zone value, it is determined to be a transient voltage support scenario; when the damping ratio of the subsynchronous oscillation, supersynchronous oscillation, or high-frequency oscillation mode is detected to be lower than the safety threshold, it is determined to be a broadband oscillation suppression scenario; when the equivalent short-circuit ratio of the substation is lower than the set threshold, the trigger priority of the broadband oscillation suppression scenario is increased; when none of the above conditions are met, it is determined to be a steady-state economic operation scenario; the switching between each scenario is equipped with a hysteresis comparison circuit to prevent frequent scenario switching due to signal fluctuations.

[0021] Furthermore, in step S4, the multi-source power economic allocation model takes minimizing the sum of the power generation cost of the new energy power generation unit, the charging and discharging loss cost of the energy storage unit, and the adjustment cost of the flexible load as the objective function, while satisfying the total active power balance constraint within the station, the upper and lower limit constraints of the output of each power generation unit and energy storage unit, the safe operating range constraint of the energy storage charging state, and the transmission capacity constraint of each feeder line.

[0022] Furthermore, in step S5, the specific implementation of the frequency coordinated response model is as follows: when the grid frequency deviation exceeds the preset dead zone, the total frequency response power required by the station is calculated based on the frequency deviation and the frequency change rate, and the total power is allocated to each control unit according to the weighted ratio of the remaining adjustable capacity and response speed of each grid-type converter; for energy storage units, a state of charge balancing factor is introduced in the allocation process, and when the state of charge of energy storage deviates from the ideal operating range, its share of frequency response power is automatically reduced.

[0023] Furthermore, in step S5, the voltage hierarchical coordinated control model includes two levels: station level and unit level. At the station level, the total reactive power adjustment required for the entire station is calculated based on the voltage deviation at the grid connection point using the voltage sensitivity coefficient to reactive power. At the unit level, the initial allocation coefficient is used as the reactive capacity ratio of each control unit, and the optimization objectives are to minimize the voltage deviation at each unit terminal and minimize the reactive circulating current between units. The initial allocation coefficient is then corrected to obtain the reactive power command value for each unit.

[0024] Furthermore, in step S6, the specific implementation of the broadband oscillation adaptive suppression model is as follows: real-time acquisition of three-phase voltage and current signals at the grid connection point; extraction of oscillation characteristic frequencies and amplitudes of subsynchronous, supersynchronous, and high-frequency bands through spectrum analysis; online identification of the equivalent output impedance of the station facing the grid; design of additional virtual impedances for each oscillation frequency band; superposition of additional virtual impedances on the basis of the fundamental frequency virtual impedance; and adaptive adjustment of the amplitude and phase angle of the additional virtual impedances according to the magnitude of the oscillation amplitude of each frequency band, so as to reshape the output impedance characteristics of the station and improve the damping ratio of the oscillation modes of each frequency band.

[0025] Furthermore, in step S7, the optimization objectives of the multi-objective optimization algorithm include: minimizing the absolute value of the time integral of the frequency deviation, minimizing the absolute value of the time integral of the voltage deviation, maximizing the damping ratio of the main mode of broadband oscillation, and minimizing the total operating cost of the power station. An improved non-dominated sorting genetic algorithm is used for multi-objective optimization. Crossover and mutation operators are introduced into the algorithm to accelerate convergence. The entropy weight method is used to select the compromise solution with the highest matching degree with the current power grid operating state from the Pareto optimal solution set obtained by solving the algorithm, which is then used as the final power command value output.

[0026] Furthermore, in step S8, the optimal power command value is encapsulated into process layer bus messages or station control layer messages conforming to the IEC61850 standard according to the type of control resource, and sent out through high-speed fiber optic Ethernet in the station; after receiving the command, each actuator performs power adjustment and uploads the actual response value to the control center in real time through the feedback channel for calculating the control deviation; if the execution deviation of two consecutive control cycles exceeds the preset threshold, the re-solution in step S7 is automatically triggered to form a closed-loop adaptive control.

[0027] Compared with the prior art, the beneficial effects of the present invention are as follows: This multi-source coordinated control method for grid-connected power plants in high-proportion renewable energy grid-connected scenarios has the following advantages:

[0028] 1. This invention realizes the unified and coordinated control of multiple types of resources such as wind power, photovoltaic, energy storage, and flexible loads in grid-type power stations, overcomes the limitations of existing technologies that only target a single type of resource or a single device, and fills the technical gap in the systematic and coordinated control of multiple types of resources in grid-type power stations.

[0029] 2. This invention achieves multi-objective collaborative optimization of grid-connected power stations under different power grid operating conditions by constructing adaptive control models for three scenarios: steady-state economic dispatch, transient frequency / voltage support, and broadband oscillation suppression. Hysteresis comparison links are provided between each scenario, effectively preventing frequent scenario switching due to signal fluctuations and improving the stability and robustness of the control system.

[0030] 3. This invention combines impedance analysis with virtual impedance reshaping strategy to achieve adaptive suppression of subsynchronous, supersynchronous and high-frequency oscillations, effectively improves the broadband oscillation stability of high-proportion new energy grid-connected systems, and overcomes the technical defects of grid-type control with low active power loop damping and easy oscillation under strong grid conditions.

[0031] 4. By establishing a dynamic coupling constraint set of active-reactive power coupling constraints and frequency-voltage-oscillation multi-objective conflict constraints, this invention achieves organic synergy among multiple scenarios and multiple objectives rather than simple superposition or alternating optimization, significantly improving the overall optimality of the control scheme.

[0032] 5. This invention adopts a hierarchical architecture combining station-level optimization and unit-level allocation, balancing global optimality with local rapid response requirements, thus improving the reliability and real-time performance of the control system. Based on unified time-scaled data acquisition and online parameter identification, this invention eliminates the need for large-scale modifications to existing stations. Utilizing the IEC 61850 standard communication protocol, it is compatible with existing smart substation communication systems, demonstrating excellent engineering applicability and widespread application value. Attached Figure Description

[0033] Figure 1 This is a flowchart of the workflow of the present invention. Detailed Implementation

[0034] 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. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0035] Please see Figure 1 This invention provides a technical solution: a multi-source coordinated control method for grid-connected power plants in scenarios with a high proportion of renewable energy grid connection. A grid-connected power plant refers to a renewable energy power generation cluster containing two or more grid-connected converters employing a grid-connected control strategy. The method is characterized by the following steps:

[0036] Step S1: Obtain the operating status data of each new energy power generation unit, energy storage unit, and flexible load in the grid-type power station, along with the grid-side operating parameters. Perform time synchronization preprocessing on the data to construct a unified time-stamped dataset of multi-source heterogeneous data. The time synchronization preprocessing includes: using the Global Positioning System or BeiDou timing signal as a reference, unifying the data from each source to the same time segment through interpolation resampling, and dynamically compensating for the transmission delay introduced by the communication link. The operating status data includes at least the grid-connected power, terminal voltage, and terminal current of each power generation unit, the state of charge and charging / discharging power of the energy storage unit, and the grid-side frequency, grid connection point voltage, and line power flow.

[0037] Step S2: Based on the unified time-scaled dataset, establish a dynamic equivalent model of the grid-type power station, calculate the overall virtual inertia, equivalent damping coefficient, and short-circuit ratio distribution of the power station, and obtain the equivalent characteristic parameter set of the power station. The specific method for establishing the dynamic equivalent model of the grid-type power station is as follows: Based on the port-controlled Hamiltonian model of each grid-type converter, the entire power station is equivalent to a second-order inertia-damping dynamic model, and the equivalent virtual inertia and equivalent damping coefficient of the power station are identified online using the recursive least squares method with forgetting factor; at the same time, according to the impedance network relationship between each grid-connected node in the power station, the short-circuit ratio of each grid-connected node is calculated to form a comprehensive characteristic parameter set including the distribution of inertia, damping, and short-circuit ratio.

[0038] Step S3: Based on the grid operating status and the equivalent characteristic parameter set of the substation, the control scenarios are divided into steady-state economic operation scenarios, transient frequency / voltage support scenarios, and wideband oscillation suppression scenarios. The criteria for classifying each control scenario include: when the absolute value of the grid frequency deviation exceeds the set dead zone value and the frequency change rate is greater than the set threshold, it is determined to be a transient frequency support scenario; when the absolute value of the voltage deviation at the grid connection point exceeds the set dead zone value, it is determined to be a transient voltage support scenario; when the damping ratio of the subsynchronous oscillation, supersynchronous oscillation, or high-frequency oscillation mode is detected to be lower than the safety threshold, it is determined to be a wideband oscillation suppression scenario; when the equivalent short-circuit ratio of the substation is lower than the set threshold, the trigger priority of the wideband oscillation suppression scenario is increased; when none of the above conditions are met, it is determined to be a steady-state economic operation scenario; a hysteresis comparison circuit is provided for the switching between each scenario to prevent frequent scenario switching due to signal fluctuations.

[0039] Step S4: For the steady-state economic operation scenario, with the goal of minimizing the overall operating loss of the power station and maximizing the renewable energy absorption rate, a multi-source power economic allocation model is constructed. The multi-source power economic allocation model takes minimizing the sum of the power generation cost of renewable energy generation units, the charging and discharging loss cost of energy storage units, and the adjustment cost of flexible loads as the objective function, while satisfying the total active power balance constraint within the power station, the upper and lower limit constraints of the output of each power generation unit and energy storage unit, the safe operating range constraint of the energy storage charging state, and the transmission capacity constraint of each feeder line.

[0040] Step S5: For transient frequency / voltage support scenarios, a frequency coordinated response model is constructed based on the virtual synchronous machine control strategy, and a voltage hierarchical coordinated control model is constructed based on reactive power and voltage sensitivity analysis. The specific implementation of the frequency coordinated response model is as follows: When the grid frequency deviation exceeds the preset dead zone, the total frequency response power required by the power station is calculated based on the frequency deviation and frequency change rate, and the total power is allocated to each control unit according to the weighted ratio of the remaining adjustable capacity and response speed of each grid-type converter. For energy storage units, a state of charge balancing factor is introduced during the allocation process. When the state of charge of energy storage deviates from the ideal operating range, its share of frequency response power is automatically reduced. The voltage hierarchical coordinated control model includes two levels: power station level and unit level. At the power station level, the total reactive power adjustment required by the entire power station is calculated based on the voltage deviation at the grid connection point using the voltage sensitivity coefficient to reactive power. At the unit level, the reactive power capacity ratio of each control unit is used as the initial allocation coefficient, and the initial allocation coefficient is corrected with the minimum voltage deviation at each unit end and the minimum reactive power circulating current between units as the optimization objectives, to obtain the reactive power command value of each unit.

[0041] Step S6: For wideband oscillation suppression scenarios, the oscillation frequency bands are identified based on impedance analysis, and a wideband oscillation adaptive suppression model is constructed. The virtual impedance parameters of the grid-connected converter are dynamically adjusted. The specific implementation of the wideband oscillation adaptive suppression model is as follows: real-time acquisition of three-phase voltage and current signals at the grid connection point; extraction of oscillation characteristic frequencies and amplitudes of subsynchronous, supersynchronous, and high-frequency bands through spectrum analysis; online identification of the equivalent output impedance of the substation facing the grid; design of additional virtual impedances for each oscillation frequency band; superposition of additional virtual impedances on the basis of the fundamental frequency virtual impedance; and adaptive adjustment of the amplitude and phase angle of the additional virtual impedances according to the magnitude of the oscillation amplitude of each frequency band to reshape the output impedance characteristics of the substation and improve the damping ratio of the oscillation modes of each frequency band.

[0042] Step S7: Based on the variable coupling relationship between the multi-source power economic allocation model, frequency coordinated response model, voltage hierarchical coordinated control model, and broadband oscillation adaptive suppression model, establish a dynamic coupling constraint set that includes active-reactive power coupling constraints and frequency-voltage-oscillation multi-objective conflict constraints; use a multi-objective optimization algorithm to solve the above four models collaboratively under the dynamic coupling constraint set to obtain the optimal power command value for each control resource. The optimization objectives of the multi-objective optimization algorithm include: minimizing the absolute value of the time integral of frequency deviation, minimizing the absolute value of the time integral of voltage deviation, maximizing the damping ratio of the main mode of broadband oscillation, and minimizing the total operating cost of the power station; use an improved non-dominated sorting genetic algorithm for multi-objective optimization, introduce crossover and mutation operators to accelerate convergence, and use the entropy weight method to select the compromise solution with the highest matching degree with the current power grid operating state from the Pareto optimal solution set obtained by the solution as the final power command value output;

[0043] Step S8: The optimal power command value is sent to the actuators of each new energy power generation unit, energy storage unit, and flexible load through the high-speed communication network to perform multi-source coordinated control. The execution results are fed back in real time to trigger correction. The optimal power command value is encapsulated into process layer bus messages or station control layer messages conforming to the IEC 61850 standard according to the control resource type and sent through the high-speed fiber optic Ethernet in the station. After receiving the command, each actuator performs power adjustment and uploads the actual response value to the control center in real time through the feedback channel for calculating the control deviation. If the execution deviation of two consecutive control cycles exceeds the preset threshold, the re-solution of step S7 is automatically triggered to form closed-loop adaptive control.

[0044] The present invention provides the following embodiments:

[0045] Example 1: The method of this invention is applied to a grid-type renewable energy power station comprising 20 wind turbines (total capacity 100MW), a 10MW photovoltaic array, and a 20MW / 40MWh energy storage system. Each unit within the power station adopts a grid-type control strategy. Under steady-state operating conditions, the system automatically operates in a steady-state economic dispatch scenario. The multi-source power economic allocation model optimizes the power output of the entire power station based on the generation cost of each unit and the SOC status of the energy storage, reducing operating losses by approximately 8% and increasing the renewable energy absorption rate to over 98%. When a frequency drop event occurs in the grid, the system quickly switches to a transient frequency support scenario. The frequency collaborative response model completes the calculation and allocation of total frequency response power within 200ms, and the energy storage units dynamically adjust their output based on the SOC balancing factor, effectively supporting grid frequency recovery.

[0046] Example 2: In the aforementioned power station, when a 25Hz subsynchronous oscillation characteristic is detected at the grid connection point, the system automatically switches to a broadband oscillation suppression scenario. The broadband oscillation adaptive suppression model extracts the oscillation characteristic frequency through spectrum analysis, superimposes an additional virtual impedance for the 25Hz frequency band on the basis of the fundamental frequency virtual impedance, and dynamically adjusts the impedance amplitude according to the oscillation amplitude. After suppression is implemented, the oscillation mode damping ratio in this frequency band increases from 0.02 to 0.12, and the oscillation amplitude is attenuated by more than 90%, effectively ensuring the safe and stable operation of the power station and the power grid.

[0047] Example 3: In the aforementioned power station, when the power grid experiences both frequency and voltage deviations simultaneously, the system operates in a transient frequency / voltage support scenario. The dynamic coupling constraint set comprehensively considers the mutual influence between frequency response power and voltage regulation reactive power, and the multi-objective optimization algorithm selects a compromise solution on the Pareto optimal frontier. Simulation results show that, compared to independent control methods, the collaborative control method of this invention shortens the frequency recovery time by approximately 30% and reduces the voltage deviation integral value by approximately 25%, achieving coordinated optimization of frequency support and voltage regulation.

[0048] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A multi-source coordinated control method for grid-connected power plants in scenarios with a high proportion of renewable energy grid connection, wherein a grid-connected power plant refers to a renewable energy power generation cluster containing two or more grid-connected converters employing a grid-connected control strategy, characterized in that... Includes the following steps: Step S1: Obtain the operating status data of each new energy power generation unit, energy storage unit and flexible load in the grid-type power station and the grid-side operating parameters, perform time synchronization preprocessing on the data, and construct a unified time-stamped dataset of multi-source heterogeneous data; Step S2: Based on the unified time-scaled dataset, establish a dynamic equivalent model of the network-type power station, calculate the virtual inertia, equivalent damping coefficient and multi-node short-circuit ratio distribution of the entire power station, and obtain the equivalent characteristic parameter set of the power station; Step S3: Based on the power grid operating status and the equivalent characteristic parameter set of the substations, the control scenarios are divided into steady-state economic operation scenarios, transient frequency / voltage support scenarios, and broadband oscillation suppression scenarios; Step S4: For steady-state economic operation scenarios, with the goal of minimizing overall power plant operating losses and maximizing renewable energy absorption rate, construct a multi-source power economic allocation model; Step S5: For transient frequency / voltage support scenarios, construct a frequency coordinated response model based on the virtual synchronous machine control strategy, and construct a voltage hierarchical coordinated control model based on reactive power and voltage sensitivity analysis; Step S6: For wideband oscillation suppression scenarios, identify the oscillation frequency band based on impedance analysis, construct a wideband oscillation adaptive suppression model, and dynamically adjust the virtual impedance parameters of the grid-type converter; Step S7: Based on the variable coupling relationship between the multi-source power economic allocation model, frequency coordinated response model, voltage hierarchical coordinated control model, and broadband oscillation adaptive suppression model, establish a dynamic coupling constraint set that includes active-reactive power coupling constraints and frequency-voltage-oscillation multi-objective conflict constraints; use a multi-objective optimization algorithm to solve the above four models in a coordinated manner under the dynamic coupling constraint set to obtain the optimal power command value of each control resource; Step S8: The optimal power command value is sent to the actuators of each new energy power generation unit, energy storage unit and flexible load through the high-speed communication network to perform multi-source coordinated control, and the execution results are fed back in real time to trigger correction.

2. The multi-source coordinated control method for grid-type power stations in a high-proportion renewable energy grid-connected scenario as described in claim 1, characterized in that: In step S1, time synchronization preprocessing includes: using the Global Positioning System or BeiDou timing signal as a reference, the data from each channel are unified to the same time segment by interpolation resampling, and the transmission delay introduced by the communication link is dynamically compensated; the operating status data includes at least the grid-connected power, terminal voltage, and terminal current of each power generation unit, the state of charge and charging / discharging power of the energy storage unit, as well as the frequency, grid connection point voltage, and line power flow on the grid side.

3. The multi-source coordinated control method for grid-type power stations in a high-proportion renewable energy grid-connected scenario as described in claim 1, characterized in that: In step S2, the specific method for establishing the dynamic equivalent model of the grid-type power station is as follows: based on the port-controlled Hamiltonian model of each grid-type converter, the entire power station is equivalent to a second-order inertia-damping dynamic model, and the equivalent virtual inertia and equivalent damping coefficient of the power station are identified online using the recursive least squares method with forgetting factor; at the same time, according to the impedance network relationship between each grid-connected node in the power station, the short-circuit ratio of each grid-connected node is calculated to form a comprehensive characteristic parameter set including the distribution of inertia, damping and short-circuit ratio.

4. The multi-source coordinated control method for grid-type power stations in a high-proportion renewable energy grid-connected scenario as described in claim 1, characterized in that: In step S3, the criteria for classifying each control scenario include: when the absolute value of the grid frequency deviation exceeds the set dead zone value and the frequency change rate is greater than the set threshold, it is determined to be a transient frequency support scenario; when the absolute value of the grid connection point voltage deviation exceeds the set dead zone value, it is determined to be a transient voltage support scenario; when the damping ratio of the subsynchronous oscillation, supersynchronous oscillation, or high-frequency oscillation mode is detected to be lower than the safety threshold, it is determined to be a broadband oscillation suppression scenario; when the equivalent short-circuit ratio of the substation is lower than the set threshold, the trigger priority of the broadband oscillation suppression scenario is increased; when none of the above conditions are met, it is determined to be a steady-state economic operation scenario; the switching between each scenario is equipped with a hysteresis comparison circuit to prevent frequent scenario switching due to signal fluctuations.

5. The multi-source coordinated control method for grid-type power stations in a high-proportion renewable energy grid-connected scenario as described in claim 1, characterized in that: In step S4, the multi-source power economic allocation model takes minimizing the sum of the power generation cost of the new energy power generation unit, the charging and discharging loss cost of the energy storage unit, and the adjustment cost of the flexible load as the objective function, while satisfying the total active power balance constraint within the station, the upper and lower limit constraints of the output of each power generation unit and energy storage unit, the safe operating range constraint of the energy storage charging state, and the transmission capacity constraint of each feeder line.

6. The multi-source coordinated control method for grid-type power stations in a high-proportion renewable energy grid-connected scenario as described in claim 1, characterized in that: In step S5, the specific implementation of the frequency coordinated response model is as follows: when the grid frequency deviation exceeds the preset dead zone, the total frequency response power required by the station is calculated based on the frequency deviation and the frequency change rate, and the total power is allocated to each control unit according to the weighted ratio of the remaining adjustable capacity and response speed of each grid-type converter; for energy storage units, a state of charge balancing factor is introduced in the allocation process, and when the state of charge of energy storage deviates from the ideal operating range, its share of frequency response power is automatically reduced.

7. The multi-source coordinated control method for grid-type power stations in a high-proportion renewable energy grid-connected scenario as described in claim 1, characterized in that: In step S5, the voltage hierarchical coordinated control model includes two levels: station level and unit level. At the station level, the total reactive power adjustment required for the entire station is calculated based on the voltage deviation at the grid connection point using the voltage sensitivity coefficient to reactive power. At the unit level, the initial allocation coefficient is the proportion of reactive power capacity of each control unit. The optimization objectives are to minimize the voltage deviation at each unit terminal and the reactive power circulating current between units. The initial allocation coefficient is then corrected to obtain the reactive power command value of each unit.

8. The multi-source coordinated control method for grid-type power stations in a high-proportion renewable energy grid-connected scenario as described in claim 1, characterized in that: In step S6, the specific implementation of the broadband oscillation adaptive suppression model is as follows: real-time acquisition of three-phase voltage and current signals at the grid connection point; extraction of oscillation characteristic frequencies and amplitudes of subsynchronous, supersynchronous, and high-frequency bands through spectrum analysis; online identification of the equivalent output impedance of the station facing the grid; design of additional virtual impedances for each oscillation frequency band; superposition of additional virtual impedances on the basis of the fundamental frequency virtual impedance; adaptive adjustment of the amplitude and phase angle of the additional virtual impedances according to the magnitude of the oscillation amplitude of each frequency band; reshaping the output impedance characteristics of the station; and improving the damping ratio of the oscillation modes of each frequency band.

9. The multi-source coordinated control method for grid-type power stations in a high-proportion renewable energy grid-connected scenario as described in claim 1, characterized in that: In step S7, the optimization objectives of the multi-objective optimization algorithm include: minimizing the absolute value of the time integral of the frequency deviation, minimizing the absolute value of the time integral of the voltage deviation, maximizing the damping ratio of the main mode of broadband oscillation, and minimizing the total operating cost of the power station. An improved non-dominated sorting genetic algorithm is used for multi-objective optimization. Crossover and mutation operators are introduced into the algorithm to accelerate convergence. The entropy weight method is used to select the compromise solution with the highest matching degree with the current power grid operating state from the Pareto optimal solution set obtained by solving the algorithm, which is then used as the final power command value output.

10. The multi-source coordinated control method for grid-type power stations in a high-proportion renewable energy grid-connected scenario as described in claim 1, characterized in that: In step S8, the optimal power command value is encapsulated into process layer bus messages or station control layer messages conforming to the IEC 61850 standard according to the type of control resource, and sent through the high-speed fiber optic Ethernet in the station. After receiving the command, each actuator performs power adjustment and uploads the actual response value to the control center in real time through the feedback channel for calculating the control deviation. If the execution deviation of two consecutive control cycles exceeds the preset threshold, the re-solution in step S7 is automatically triggered to form a closed-loop adaptive control.

Images