A virtual synchronous grid connection collaborative control method for a large-scale photovoltaic direct current collection system

Abstract

The application discloses a kind of virtual synchronous network construction collaborative control methods of large-scale photovoltaic direct current collection system, comprising the following steps: step one: obtaining the network construction support capability index of photovoltaic group string;Wherein, the network construction support capability index is in real time calculated during the MPPT control process to photovoltaic group string;Step two: based on the network construction support capability index, dynamically allocate virtual inertia to stabilize direct current bus voltage;And when monitoring power grid fault crossing signal, switch to direct current voltage support mode;Step three: obtaining the power backup request generated according to the network construction intensity evaluation of alternating current grid;The power backup request is generated based on the network construction intensity evaluation result of alternating current grid;Step four: by the network construction support capability index and the power backup request, construct hierarchical collaborative closed-loop control loop, and execute virtual synchronous network construction collaborative control.The application improves the ability of large-scale photovoltaic direct current collection system in extreme environment continuous network construction and stable direct current voltage.

Description

Technical Field

[0001] This invention belongs to the field of new energy power generation and DC power transmission technology, specifically relating to a virtual synchronous grid-connected collaborative control method for a large-scale photovoltaic DC collection system. Background Technology

[0002] Northwest my country's desert and Gobi regions are rich in solar energy resources and are a core area for new energy development under the "dual-carbon" strategy. However, the harsh natural conditions and weak power grid infrastructure in this region pose severe challenges to the stable operation of large-scale photovoltaic DC collection systems. At the power grid level, desert and Gobi regions are mostly located at the end of the main grid, with a sparse grid structure and severely insufficient system strength and inertia support. Traditional "grid-connected" photovoltaic systems are highly dependent on the power grid to provide a stable voltage and frequency reference. In weak grid environments, they struggle to maintain stable operation and are prone to complex dynamic interactions with grid impedance, inducing broadband oscillations that threaten equipment and system safety. At the natural environment level, severe sandstorms can cause photovoltaic power to drop by seconds, placing extremely high demands on system transient stability; extreme diurnal temperature variations subject power electronic devices to severe thermal stress cycles, accelerating aging; extreme daytime temperatures force equipment to operate at reduced capacity, conflicting with the strong support demands of the power grid during peak midday hours. Furthermore, long-distance DC collection lines cover vast areas, and their electrical parameters drift significantly with factors such as temperature, sandstorms, and aging, leading to severe performance degradation of control strategies based on fixed models. To overcome these challenges, a complete shift from "grid-following operation" to "proactive grid construction" is needed. Photovoltaic DC aggregation systems should no longer function as passive loads or simple current sources for the grid, but rather evolve into "virtual synchronous power sources" capable of actively establishing, regulating, and maintaining AC bus voltage and frequency. The core objective is to enable thousands of dispersed photovoltaic units to aggregate on the grid side through coordinated control, presenting characteristics of a voltage source with inertia, damping, and voltage support capabilities, similar to or even more robust than traditional synchronous generators. This allows them to autonomously provide instantaneous active and reactive power support during disturbances, suppressing frequency and voltage fluctuations and fundamentally enhancing the stability and power quality of weak power grids. Summary of the Invention

[0003] The technical problem to be solved by this invention is the difficulty in achieving continuous grid construction and stable DC voltage control in large-scale photovoltaic DC collection systems under extreme environments.

[0004] To address the aforementioned technical problems, this invention provides a virtual synchronous grid-connected collaborative control method for a large-scale photovoltaic DC collection system, comprising the following steps:

[0005] Step 1: Obtain the grid support capability index of the photovoltaic string; wherein, the grid support capability index is calculated in real time during the MPPT control process of the photovoltaic string;

[0006] Step 2: Based on the aforementioned grid support capability index, dynamically allocate virtual inertia to stabilize the DC bus voltage; and switch to DC voltage support mode when a grid fault ride-through signal is detected.

[0007] Step 3: Obtain the power reserve request generated based on the AC power grid strength assessment; the power reserve request is generated based on the AC power grid strength assessment results;

[0008] Step 4: Construct a hierarchical collaborative closed-loop control loop using the network support capability index and the power reserve request, and execute virtual synchronous network collaborative control.

[0009] In the aforementioned virtual synchronous grid-connected collaborative control method for large-scale photovoltaic DC collection systems, after performing MPPT control on the photovoltaic strings, a virtual capacitor response loop is superimposed to suppress voltage fluctuations in the photovoltaic strings.

[0010] The formula for the virtual capacitor response loop is: ,

[0011] in, This represents the change in the output current of the photovoltaic string. This indicates the output DC voltage of the MPPT. Indicates time; This represents the virtual capacitance coefficient, which is dynamically adjusted based on the string capacity. , This indicates the actual output power of the current photovoltaic string. This indicates the rated output power of the photovoltaic string. Indicates the reference virtual capacitance coefficient;

[0012] A temperature-irradiance feedforward derating factor is used to limit the power output of photovoltaic strings when environmental conditions exceed a threshold. Represented as:

[0013] ,

[0014] in: Indicates real-time solar irradiance; Indicates reference solar irradiance; Indicates the junction temperature of the photovoltaic cell. Indicates external meteorological parameters; Indicates the temperature decay coefficient. Indicates the light intensity attenuation coefficient;

[0015] The network support capability index for: This indicates the dynamic response potential of the photovoltaic string in this circuit; among which, This indicates the maximum power of the photovoltaic string in this circuit. This represents dynamic impedance.

[0016] In the aforementioned virtual synchronous grid-connected collaborative control method for large-scale photovoltaic DC collection systems, step two involves dynamically allocating virtual inertia through VDG control, which includes voltage outer loop control and current inner loop control.

[0017] The voltage outer loop control stabilizes the DC bus voltage at the rated target value by simulating the second-order swing equation of a synchronous generator. The second-order swing equation is as follows: ,

[0018] in: This represents the analog electromagnetic power, mapped to the output power of the photovoltaic string. ; This represents the analog mechanical power, mapped to the power output from the bus boost unit to the grid-connected unit. ; Indicates virtual angular displacement; The virtual damping coefficient is represented by J; the virtual moment of inertia is represented by J; the virtual moment of inertia is dynamically allocated based on the weighted sum of the grid support capacity indicators of each photovoltaic string. ,in Indicates the first One photovoltaic string, Indicates the first The grid support capability index of each photovoltaic string. This represents the weighting factor based on line impedance correction. ; Indicates the system reference impedance. This represents the measured value of the line impedance of this branch line;

[0019] The current inner loop control utilizes dynamic virtual impedance to suppress multi-frequency oscillations, ensuring that the DC bus voltage accuracy is controlled within a set range; the dynamic virtual impedance is expressed as: The dynamic virtual impedance is realized through the real-time numerical representation of the time-domain virtual impedance expression: This is used to compensate for the resistive voltage drop and inductive effect of DC lines in desert areas, where: Indicates the impedance of the basic circuit. Indicates dynamic virtual incremental impedance. Indicates the magnitude of the load current. The low-frequency component representing the virtual impedance. The mid-frequency component representing the virtual impedance. High-frequency resonant term of virtual impedance, This represents a complex frequency variable.

[0020] The aforementioned virtual synchronous grid-connected collaborative control method for large-scale photovoltaic DC collection systems uses a dynamic virtual incremental impedance constructed by time-domain superposition of low-frequency components, mid-frequency components, and high-frequency resonant terms.

[0021] The low-frequency component of the virtual impedance The calculation formula is: ,in, THD represents the harmonic distortion rate of the medium-voltage DC bus voltage. This represents the compensation factor for the increase in impedance. Indicates the switching frequency of the device. Represents the virtual harmonic resistance component. Represents the virtual harmonic inductance component;

[0022] The mid-frequency component of the virtual impedance The calculation formula is: ,in Represents the basic virtual inductance. Represents the adaptive coefficient. This indicates dynamic impedance, used to automatically increase inertia and prevent instability when the MPPT support capacity is weak;

[0023] The high-frequency resonant term of the virtual impedance The calculation formula is: ,in: Indicates the resonant angular frequency. Indicates the damping ratio. It represents the resonant damping gain and is positively correlated with the peak THD.

[0024] In the aforementioned virtual synchronous grid-connected collaborative control method for large-scale photovoltaic DC collection systems, step two, the DC voltage support mode, refers to adjusting the high-voltage side voltage control target from a constant DC voltage to a dynamically adjusted DC voltage that varies with power, in order to actively absorb power fluctuations and prevent power backfeeding from impacting the photovoltaic array. The high-voltage side voltage control formula is: ,in, Indicates the actual operating DC voltage. Indicates a constant DC voltage. This represents the voltage droop factor. This indicates fluctuations in grid power.

[0025] The aforementioned virtual synchronous grid construction and collaborative control method for large-scale photovoltaic DC collection systems, wherein the AC grid construction strength is evaluated in real time based on the system short-circuit ratio (SCR) and equivalent inertia time constant. When the short-circuit ratio (SCR) < 3 or the equivalent inertia time constant is less than 3s, the grid construction capability enhancement mode is triggered, and a power reserve request command is sent to the combiner boost unit.

[0026] The formula for the equivalent inertia time constant is: ,in, This represents the total energy stored in the capacitors within the grid-connected layer. Indicates the system's rated capacity. Indicates the first The reference inertia value of each inertia unit. Indicates the first The inertia derating factor of the unit; when the equivalent inertia time constant is lower than the threshold, the system triggers a power reserve request, controlling the bus boost unit to increase the DC bus voltage within the fault detection time to reserve power. The power reserve request formula is: , This indicates the reserve power that needs to be stored. Indicates the current operating voltage. Indicates the voltage increase. This represents the equivalent bus capacitance.

[0027] The aforementioned virtual synchronous grid-building collaborative control method for large-scale photovoltaic DC collection systems introduces a VDG power synchronization control link under the grid-building capability enhancement mode to achieve inertia coupling and transfer between AC and DC power grids.

[0028] The VDG power synchronization control loop is used to convert the AC grid frequency deviation into a DC voltage correction value, as shown in the formula: ,in, Indicates the frequency deviation of the AC power grid. This indicates the DC voltage correction amount. Represents the time constant. Represents a complex frequency variable. Indicates the frequency droop factor; Dynamically adjust based on system inertia ,in, This represents the depreciation factor. Indicates the rated sag coefficient. Represents the equivalent inertia time constant. This represents the reference value of the equivalent inertia time constant.

[0029] In the aforementioned virtual synchronous grid-connected collaborative control method for large-scale photovoltaic DC collection systems, step four involves hierarchical collaborative closed-loop control performing multi-time-scale coordination, including:

[0030] At the first time scale of milliseconds, based on the instantaneous measurements of local voltage and current, MPPT control and virtual inertia adjustment of the photovoltaic strings are performed, and grid construction capability information is exchanged.

[0031] At the second timescale of hundreds of milliseconds, the regional coordination control layer coordinates and allocates the virtual inertia distribution of each bus boost unit based on the regional power imbalance, with the objective function being: ,in, The penalty weighting coefficient representing the voltage deviation; Indicates the first The dynamic deviation of frequency at any given moment; Indicates the first DC voltage deviation at any given time;

[0032] At the third timescale, on the order of seconds, based on power grid dispatch instructions, environmental early warning information, and equipment status data, full-level operational parameter optimization is performed. The objective function for optimization is:

[0033]

[0034] in, This represents the actual active power received. This represents the reference value for active power. This represents the reference value for the equivalent inertia; c1, c2, c3, and c4 represent weighting coefficients; T loss Dust represents the total system heat loss. risk This indicates the dust accumulation risk index.

[0035] The aforementioned virtual synchronous grid-connected collaborative control method for large-scale photovoltaic DC collection systems also includes dynamic switching strategies for extreme operating conditions:

[0036] High-temperature derating coordination strategy: When the ambient temperature exceeds the derating threshold, the system uniformly issues a derating coefficient to reduce power output, while controlling the bus boost unit to increase the virtual damping coefficient to enhance system stability, and controlling the grid-connected unit to reduce the switching frequency of power devices to reduce losses.

[0037] Sandstorm crossing strategy: The system predicts sandstorm events based on environmental monitoring data, including the sandstorm warning stage and the sandstorm approaching stage; during the sandstorm warning stage, the MPPT is controlled to switch to constant voltage mode to maintain low power operation; during the sandstorm approaching stage, the faulty branch is isolated, and the grid-connected unit is controlled to maintain the DC bus voltage and output reactive power support to the AC grid.

[0038] Smooth day-night switching strategy: At sunset, the MPPT is controlled to gradually reduce power according to the preset soft shutdown slope; the system controls the bus boost unit to transfer the virtual inertia to the station power energy storage system. The station power energy storage system receives the transferred inertia parameters and uses the same virtual synchronization protocol to maintain the bus voltage, realizing long-term continuous grid construction.

[0039] The present invention also provides a large-scale photovoltaic DC collection system, comprising:

[0040] A photovoltaic string unit includes multiple photovoltaic strings equipped with controllers, which are used to perform MPPT control and calculate grid support capability indicators;

[0041] The bus boost unit is connected to the photovoltaic string unit and is used to collect the output power of the photovoltaic string unit and form a DC bus. It receives the grid support capability index and dynamically adjusts the virtual inertia based on the grid support capability index to stabilize the DC bus voltage. It also receives power reserve requests and boosts the DC bus voltage within the fault identification time to reserve power.

[0042] The grid connection unit, connected to the busbar booster unit, is used to connect the DC busbar to the AC grid, assess the grid strength of the AC grid, and issue a power reserve request to the busbar booster unit based on the assessment results.

[0043] The hierarchical collaborative control unit is connected to the photovoltaic string unit, the busbar booster unit and the grid-connected unit respectively, and is used to transmit the grid support capability indicators and power reserve requests.

[0044] The aforementioned large-scale photovoltaic DC collection system,

[0045] The controller of the photovoltaic string unit is also used to perform virtual capacitor response control;

[0046] The combiner boost unit includes a DC combiner box, a regional coordinator, and a DCT controller. The DC combiner box is used to collect the output power of the photovoltaic string units to the medium-voltage DC bus. The regional coordinator is used to dynamically allocate virtual inertia according to the grid support capability index. The DCT controller is used to connect the medium-voltage DC bus to the high-voltage DC bus and execute VDG control to maintain the stability of the DC bus voltage by adjusting the virtual rotational inertia and virtual damping.

[0047] The grid-connected unit includes an MMC controller, which has a built-in grid strength evaluator.

[0048] The beneficial effects achieved by this invention are as follows: By dynamically allocating virtual inertia resources across all levels, this invention enables the DC bus voltage to maintain high-precision stability under multi-source power disturbances, improving the system's anti-disturbance capability and power supply reliability. Through VDG control, it realizes the inertia coupling and transfer between AC and DC power grids, enhancing the system's all-weather continuous grid-building operation capability. Addressing the characteristics of desert environments, it ensures the system maintains its grid-building characteristics and fault ride-through capability even under extreme conditions. This invention constructs a three-layer hierarchical collaborative control architecture comprising photovoltaic string units, busbar booster units, and grid-connected units. At each layer, it introduces core mechanisms such as dynamic calculation and uploading of grid-building support capability indicators, virtual inertia adjustment and damping compensation based on virtual DC generators, and grid strength assessment and power reserve request issuance, forming a closed-loop collaborative control system for the forward transmission of grid-building capability indicators and the reverse transmission of power reserve requests. Attached Figure Description

[0049] Figure 1 This is a flowchart of the control method of the present invention;

[0050] Figure 2 This is a system control logic architecture diagram of the present invention;

[0051] Figure 3 This is a block diagram of the virtual capacitor response loop control of the present invention;

[0052] Figure 4 This is a block diagram of dynamic inertia allocation weighted by the VDG control and network support capability index of the present invention.

[0053] Figure 5 This is a block diagram of the network strength assessment and reinforcement control logic of the present invention. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0055] Example 1

[0056] This embodiment provides a virtual synchronous grid-connected collaborative control method for a large-scale photovoltaic DC collection system, including the following steps:

[0057] Step 1: Perform Maximum Power Point Tracking (MPPT) control on the photovoltaic string and calculate the grid support capability index of this photovoltaic string in real time; upload the grid support capability index to the DC boost unit;

[0058] Step 2: Control the bus boost unit to receive the grid support capability index and dynamically allocate virtual inertia according to the grid support capability index to stabilize the DC bus voltage; at the same time, monitor the grid connection status and switch to DC voltage support mode when a grid fault ride is detected;

[0059] Step 3: The system performs grid connection control through the grid connection unit, evaluates the grid strength of the AC power grid, generates a power reserve request based on the evaluation results, and sends the power reserve request to the combiner boost unit;

[0060] Step 4: By combining the grid support capability index with the power reserve request, a hierarchical collaborative closed-loop control is formed to achieve dynamic scheduling of virtual inertia and complete the coordinated operation of the photovoltaic power generation system.

[0061] In step one, the maximum power point is found using the conventional incremental conductance method, and maximum power point tracking control is performed for each photovoltaic string. Then, a virtual capacitor response loop is added, using the DC output voltage of the MPPT as the feedback quantity, to collect the capacity information and operating conditions of the photovoltaic strings in real time, and dynamically adjust the virtual capacitor coefficient according to the capacity of the photovoltaic strings to adaptively suppress the voltage fluctuation of the DC bus.

[0062] The formula for the virtual capacitor response loop is: ,

[0063] in, This represents the change in the output current of the photovoltaic string. This indicates the output DC voltage of the MPPT. Indicates time; Represents the virtual capacitance coefficient. Dynamically adjust based on string capacity: , This indicates the actual output power of the current photovoltaic string. This indicates the rated output power of the photovoltaic string. This represents the baseline virtual capacitance coefficient.

[0064] To ensure equipment safety and reserve grid support capabilities under extreme environmental conditions, a temperature-irradiance feedforward derating factor is introduced into the system's MPPT algorithm. This temperature-irradiance feedforward derating factor is used to proactively limit MPPT power output when encountering high temperatures in desert environments (e.g., temperatures exceeding 70°C) or sudden drops in solar irradiance caused by dust obscuring the solar grid. This prevents thermal runaway of photovoltaic modules and system power oscillations. The formula for the temperature-irradiance feedforward derating factor is: ,

[0065] in: Represents real-time solar irradiance, in units of ; This represents reference solar irradiance, in units of 1. ; Indicates the junction temperature of the photovoltaic cell. This represents external meteorological parameters, set according to actual engineering environmental conditions; Indicates the temperature decay coefficient. Indicates the light intensity attenuation coefficient. , The settings are based on the temperature coefficient and shading degradation characteristics of photovoltaic modules.

[0066] The network support capability index for: This indicates the dynamic response potential of the photovoltaic string in this circuit, among which, This indicates the maximum power of the photovoltaic string in this circuit. This represents the dynamic impedance at the point of maximum power.

[0067] In step two, the system performs virtual inertia regulation through the DC Transformer (DCT) within the bus boost unit. The DCT is used to connect the medium-voltage DC bus to the high-voltage DC bus. The DCT has a built-in virtual DC generator control algorithm that calculates and outputs virtual inertia and virtual damping in real time to suppress DC bus voltage fluctuations.

[0068] When a grid fault ride is detected, the DCT switches from a power regulation mode based on virtual inertia to a DC voltage-current droop control mode to maintain DC bus voltage stability.

[0069] The dynamic allocation of virtual inertia is achieved through control based on a virtual DC generator (VDG). The VDG control adopts a dual closed-loop control architecture, which includes voltage outer loop control and current inner loop control.

[0070] The voltage outer loop control stabilizes the DC bus voltage at the rated target value by simulating the second-order swing equation of a synchronous generator. The second-order swing equation is as follows: ,

[0071] in: This represents the analog electromagnetic power, mapped to the output power of the photovoltaic string unit. ; This represents the analog mechanical power, mapped to the power output from the bus boost unit to the grid-connected unit. ; Indicates virtual angular displacement; represents the virtual damping coefficient; J represents the virtual moment of inertia.

[0072] The virtual moment of inertia J has dynamic damping injection: when When the acceleration threshold is exceeded, Instantaneous increase in damping term: ,in This is the enhancement coefficient under strong disturbances in a desert environment. Its value is selected based on the oscillation suppression effect and is used to suppress sudden power oscillations caused by sandstorms.

[0073] The virtual moment of inertia J is dynamically allocated based on the weighted sum of the grid support capacity indicators of each photovoltaic string, using the following formula: ,in Indicates the first One photovoltaic string, Indicates the first The grid support capability index of each photovoltaic string. This represents the weighting factor based on line impedance correction; ; Indicates the system reference impedance. This represents the measured value of the line impedance of this branch.

[0074] The current inner loop control utilizes dynamic virtual impedance to suppress multi-frequency oscillations, ensuring that the DC bus voltage accuracy is controlled within a set range. The formula for the dynamic virtual impedance is: The dynamic virtual impedance is realized through real-time numerical representation of the time-domain virtual impedance expression: This is used to compensate for the resistive voltage drop and inductive effect of DC lines in desert areas; where: Indicates the impedance of the basic circuit. Indicates dynamic virtual incremental impedance. Indicates the magnitude of the load current. The low-frequency component representing the virtual impedance. The mid-frequency component representing the virtual impedance. The high-frequency resonant term representing the virtual impedance. This represents a complex frequency variable.

[0075] The dynamic virtual incremental impedance It is formed by the time-domain superposition of low-frequency components, mid-frequency components and high-frequency resonant terms.

[0076] The low-frequency component is set based on the basic line impedance and the virtual incremental impedance, and is used to compensate for the resistance voltage drop of long-distance DC lines in the desert and the increase in resistance caused by sand and dust accumulation. The low-frequency component of the virtual impedance The calculation formula is: ,in, THD represents the harmonic distortion rate of the medium-voltage DC bus voltage. This represents the compensation factor for the increase in impedance. Indicates the DCT switching frequency. Represents the virtual harmonic resistance component. This represents the virtual harmonic inductance component.

[0077] The intermediate frequency component is set based on the basic virtual inductance and can adaptively adjust according to the branch support capability. This enhances system inertia and prevents system instability caused by operating condition fluctuations when the MPPT network support capability is weak. The intermediate frequency component of the virtual impedance... The calculation formula is: ,in Represents the basic virtual inductance. Represents the adaptive coefficient. This indicates the dynamic impedance of the branch.

[0078] The high-frequency resonant term is used to suppress high-frequency harmonics and line resonances from the bus in real time. Closed-loop suppression is achieved through harmonic detection and gain adjustment. The high-frequency resonant term of the virtual impedance... The calculation formula is: ,in: Indicates the resonant angular frequency. Indicates the damping ratio. This represents the resonant damping gain and is positively correlated with the THD peak value. This represents a complex frequency variable.

[0079] By superimposing and compensating for the above-mentioned multi-band virtual impedance, the effects of line resistance voltage drop, inductance effect, parameter drift and multi-band oscillation are effectively offset, and the bus voltage accuracy is maintained within the preset stable range.

[0080] The dynamic virtual impedance is represented as a well-structured compensation network in the complex frequency domain by superimposing the low-frequency, mid-frequency, and high-frequency components in the time domain. ,

[0081] in: , , ; , , The three components work together in the time and frequency domains to achieve multi-band line parameter drift compensation and oscillation suppression, and to maintain the accuracy of the DC bus voltage within a preset range.

[0082] Furthermore, the DC voltage support mode means adjusting the high-voltage side voltage control target from a constant DC voltage to a DC voltage that dynamically adjusts with power, in order to actively absorb power fluctuations and prevent power backfeeding from impacting the photovoltaic array. The high-voltage side voltage control formula is: ,in, Indicates the actual operating DC voltage. Indicates a constant DC voltage. This represents the voltage droop factor, which is selected based on the voltage support strength requirements. This indicates fluctuations in grid power.

[0083] When a fault is detected on the medium-voltage DC bus, graded current-limiting isolation and coordinated control is executed in sequence, including:

[0084] Within the first preset time period: less than the fault discrimination time, the virtual impedance Zv is instantaneously increased to the current limiting coefficient multiple, the safe current tolerance threshold of the power device is reached, and the rate of increase of the fault current is actively limited.

[0085] Within the second preset time period: During the fault location time, a fault signal is sent to the photovoltaic string. After receiving the fault signal, the photovoltaic string will uniformly reduce its operating power to the preset safe maintenance power.

[0086] After the third preset time: which is greater than the fault location time, after the fault is located, the primary side series circuit breaker of the grid-connected unit selectively isolates the faulty branch, and the non-faulty area works normally. The entire process of grid control does not stop the machine, and the voltage recovery time is selected according to the system recovery speed requirements.

[0087] The system performs grid connection control through the grid connection unit, which is controlled by the Modular Multilevel Converter (MMC) to connect the high-voltage DC bus to the AC power grid.

[0088] The system evaluates the AC power grid strength in real time. This strength is determined based on the system short-circuit ratio (SCR) and the equivalent inertia time constant. The SCR can be calculated in real time using the AC voltage, current, and DC bus voltage at the grid connection point. The specific calculation method is well-known to those skilled in the art and will not be elaborated here. The formula for the equivalent inertia time constant is: ,

[0089] in, This represents the total energy stored in the capacitors within the grid-connected layer. Indicates the system's rated capacity. Indicates the first The reference inertia value of each inertia unit. Indicates the first The inertia derating factor of the unit.

[0090] When SCR < 3 or the equivalent inertia time constant is less than 3s, the system automatically triggers the network capability enhancement mode, increases the virtual rotational inertia and virtual damping coefficient to maintain the DC bus voltage stability, and sends a power standby request command to the bus boost unit.

[0091] The MMC introduces a power synchronization control loop based on the VDG principle, building upon conventional vector control. This power synchronization control loop converts AC grid frequency deviation into DC voltage correction, and through droop characteristics and virtual inertia adjustment, achieves inertia coupling and transfer between the AC and DC grids.

[0092] The formula for converting the DC voltage correction amount is: ,in, Indicates the frequency deviation of the AC power grid. This indicates the DC voltage correction amount. Represents the time constant. Represents a complex frequency variable. This represents the frequency droop coefficient.

[0093] Frequency droop coefficient Dynamically adjust based on system inertia: ,

[0094] in, This represents the depreciation factor, characterizing the network strength utilization coefficient. , Indicates the rated sag coefficient. Represents the equivalent inertia time constant. This represents the reference value of the equivalent inertia time constant.

[0095] When the equivalent inertia time constant is lower than the threshold, a power reserve request command is sent to the bus boost unit. The bus boost unit then increases the DC bus voltage to reserve power within the fault detection time, as shown in the formula: ,in, This indicates the reserve power that needs to be stored. Indicates the current operating voltage. Indicates the voltage increase. This represents the equivalent bus capacitance.

[0096] During the process of connecting the high-voltage DC bus to the AC grid, the MMC implements an active rebalancing strategy for the submodule capacitor voltage. By injecting three zero-sequence voltages, the voltage balancing speed of the submodule is increased by more than 40% without increasing the switching losses. This ensures that the voltage of all submodule capacitors remains balanced, strengthens the MMC's fault ride-through capability, and enhances its continuous operation capability during faults.

[0097] The formula for the third zero-sequence voltage is: ,

[0098] in, This represents the third harmonic voltage component of the output. This indicates the voltage deviation value of the submodule. This represents the third harmonic voltage gain coefficient. This represents the fundamental angular frequency of the power grid.

[0099] In the network strength enhancement mode, the MMC actively injects a second-harmonic circulating current into the bridge arms to balance voltage fluctuations in submodule capacitors, reduce capacitor value requirements for submodules, and improve economy and network reliability. The second-harmonic circulating current formula is: ,in Represents the instantaneous value of alternating current. This indicates the amplitude of the alternating current, which is selected based on the capacitance imbalance. The initial phase is determined by the voltage equalization of the submodule. It represents angular frequency.

[0100] In step four, the hierarchical collaborative closed-loop control performs multi-timescale coordination, including:

[0101] At the first timescale of milliseconds, for example, 1-4ms, based on the instantaneous measurements of local voltage and current, MPPT control of the photovoltaic string and virtual inertia adjustment of DCT are performed. At this time, the communication network is used to broadcast the grid support capability indicators in real time and synchronize control parameters.

[0102] At a second timescale on the order of hundreds of milliseconds, such as 50-200 ms, based on the pooled regional power imbalance, a coordinated allocation of virtual inertia is performed for each DCT, with the objective function being: ,in, The penalty weighting coefficient represents the voltage deviation. ; Indicates the first The dynamic deviation of frequency at any given moment; Indicates the first DC voltage deviation at any given time;

[0103] At a third timescale on the order of seconds, such as 1-3 seconds, based on grid dispatch instructions, environmental early warning information, and equipment status data, full-level operational parameter optimization is performed, globally optimizing the operating setpoints of the photovoltaic string unit, the current converter boost unit, and the grid-connected unit. This includes adjusting the derating factor of the photovoltaic string, the voltage setting reference value of the DC bus, and the reactive power support of the MMC; the execution cycle of the operational parameter optimization action is selected according to the response speed requirements of the energy management system.

[0104] In the third timescale at the second level, the full-level operational parameter optimization possesses multi-objective rolling optimization capabilities. The objective function for this rolling optimization capability is a comprehensive weighted optimality of power generation, grid support effect, equipment heat loss, and environmental adaptability. The objective function is:

[0105]

[0106] in, This represents the actual active power received. This represents the reference value for active power. The reference value represents the equivalent inertia; c1, c2, c3, and c4 represent weighting coefficients, selected according to the importance weight of each sub-objective; T loss Dust represents the total system heat loss. risk This indicates the dust accumulation risk index.

[0107] During the transmission of the network support capability indicators and power reserve requests, an event-driven communication mechanism is adopted: when the regional power imbalance or DC bus voltage deviation exceeds its respective set threshold, a coordination command is triggered; the minimum interval between two event communications is not less than a preset suppression time. This event-driven communication mechanism reduces the long-distance communication load in the desert and improves communication reliability while ensuring coordinated control performance.

[0108] The network construction capability index also includes an online optimization algorithm, which is repeatedly executed at a preset period, such as 5 minutes. The optimization algorithm formula is as follows:

[0109] ,

[0110] ,

[0111] in Indicates the first The actual output power of each photovoltaic string Indicates the first Power reference value for each photovoltaic string, This indicates that the system is allocated to the first [unit / item] based on its overall performance indicators. The virtual rotational inertia of the photovoltaic string is based on the first of the system's comprehensive performance indicators. Solar photovoltaic string allocation quotas Indicates the first The optimal virtual moment of inertia reference value corresponding to the photovoltaic string. Online updates enable optimal allocation of network resources across the entire system. , The weighting coefficients are selected based on the principles of prioritizing power generation and ensuring grid connection safety. This indicates a constraint.

[0112] In step four, the forward transfer of the grid support capability index from the photovoltaic string unit to the grid-connected unit is achieved by the grid support capability transfer function, which includes the transfer function from the photovoltaic string unit to the busbar booster unit and the transfer function from the busbar booster unit to the grid-connected unit.

[0113] The transfer function from the photovoltaic string unit to the bus boost unit characterizes the impact of power disturbances on the medium-voltage bus voltage, and the formula is: ,

[0114] in, This indicates a small-signal disturbance in the medium-voltage DC bus voltage. This represents the change in power input to this stage. This represents the equivalent sum of the output capacitance of each photovoltaic string unit and the line-to-ground capacitance. It represents the equivalent resistance of each branch line, including line loss, contact resistance, and resistance to dust and dirt buildup.

[0115] The transfer function from the bus boost unit to the grid-connected unit is: ,in Small-signal disturbances in the high-voltage DC bus voltage, where n represents the turns ratio. Represents the virtual impedance transfer function. This represents the load impedance transfer function.

[0116] The full-level collaborative compensation control strategy adopted in this invention can optimize the overall stability characteristics of the system, enabling the system to have the phase margin and amplitude margin required by the standard, and ensuring that the system still has good dynamic stability and anti-disturbance capability under the weak power grid conditions with a short-circuit ratio of 2.5 in desert areas.

[0117] To adapt to the extreme operating conditions of the desert environment, the virtual synchronous grid-connected collaborative control method for the large-scale photovoltaic DC collection system also includes a dynamic switching strategy for extreme operating conditions:

[0118] High-temperature derating coordination strategy: When the temperature exceeds the derating threshold, the system controls the photovoltaic string unit to reduce the power output according to the unified derating factor issued. At the same time, it controls the bus boost unit to increase the virtual damping factor to the enhancement level to suppress oscillations caused by power changes and maintain system stability. The grid-connected unit is controlled to synchronously reduce the switching frequency of power devices to reduce losses.

[0119] Sandstorm crossing strategy based on network capability preservation: The system predicts sandstorms in advance through satellite remote sensing and local visibility sensors, and the system is divided into two stages of switching:

[0120] The first stage is the early warning stage: control the MPPT to switch to constant voltage mode to maintain low power operation;

[0121] The second stage is the sandstorm arrival stage: the bus boost unit isolates the faulty string, while maintaining the transmission of the grid construction capability index of the non-faulty branch, and the grid connection unit maintains the DC bus voltage and outputs reactive power to support the grid voltage; the DC bus voltage fluctuation during the switching process is less than the safety threshold.

[0122] The hierarchical switching logic of the sandstorm crossing mode strategy is implemented by a fuzzy controller. The input variables of the fuzzy controller include visibility, dust accumulation rate, and DC bus voltage fluctuation, and the output variable is the switching mode command. The fuzzy controller has preset fuzzy rules. When the visibility is lower than the emergency threshold and the dust accumulation rate exceeds the set threshold, the DCT voltage support mode and MPPT constant voltage mode are triggered. The switching sequence is synchronized through preset timestamps to ensure a smooth switching across the entire network. The power fluctuation at the moment of switching is selected based on the impact limit.

[0123] In the sandstorm crossing mode strategy that maintains network construction capability, the MMC output reactive power priority mode is defined by the following formula:

[0124] ,in, This indicates the maximum reactive power that can be output. Indicates the rated apparent power. This represents the actual output power of the photovoltaic string. During a sudden drop, the reactive power output is maximized to support the grid voltage. The reactive current response time is selected based on the switching speed and stress capability of the MMC device. The ratio of the increase in the support voltage recovery speed and the ratio of the reduction in voltage drop depth are determined according to the requirements of the overall engineering design.

[0125] Day / night switching smoothing strategy: At sunset, the MPPT gradually reduces power according to a preset soft shutdown slope. The DCT transfers the virtual inertia to the station's power storage system within the MMC station. The capacity is selected based on the required nighttime support duration. The energy storage system receives the inertia parameters transferred by the DCT and maintains the bus voltage using the same virtual synchronous machine protocol, achieving long-term continuous grid connection. The virtual inertia transfer smoothness index is selected based on mechanical shock limits.

[0126] The day-night switching smooth measurement includes the MPPT soft shutdown sequence: at sunset, the power is gradually reduced according to the soft shutdown slope selected by the project, and the DCT virtual capacitor is increased simultaneously to compensate for the voltage change rate caused by the power loss. When the actual output power of the photovoltaic string is lower than the dormancy threshold, the MPPT switches to dormancy mode, and the DCT voltage control loop gain is increased according to the voltage drop suppression requirements to maintain the stability of the bus. The voltage deviation throughout the process is less than the project accuracy requirements.

[0127] Example 2

[0128] This embodiment provides a large-scale photovoltaic DC collection system, including:

[0129] A photovoltaic string unit includes multiple photovoltaic strings equipped with controllers, which are used to perform MPPT control and calculate grid support capability indicators;

[0130] The bus boost unit is used to collect the output power of the photovoltaic string unit and form a DC bus, receive the grid support capability index, and dynamically adjust the virtual inertia based on the grid support capability index to stabilize the DC bus voltage, receive power reserve requests, and boost the DC bus voltage within the fault detection time to reserve power.

[0131] The grid connection unit is used to connect the DC bus to the AC grid, assess the grid strength, and issue a power reserve request to the bus boost unit based on the assessment results.

[0132] The hierarchical collaborative control unit connects the photovoltaic string unit, the busbar booster unit, and the grid-connected unit, and is used to transmit the grid support capability indicators and power reserve requests.

[0133] The controller of the photovoltaic string unit is also used to perform virtual capacitor response control.

[0134] The combiner unit includes a DC combiner box, a regional coordinator, and a DCT controller. The DC combiner box is used to collect the output power of the photovoltaic string units to the medium-voltage DC bus; the regional coordinator is used to dynamically allocate virtual inertia according to the grid support capability index; the DCT controller is used to connect the medium-voltage DC bus to the high-voltage DC bus and execute VDG control to maintain the stability of the DC bus voltage by adjusting the virtual rotational inertia and virtual damping.

[0135] The DCT controller is an isolated dual active bridge cascade structure. Its primary side uses series input to withstand medium-voltage DC, and the secondary side uses parallel output to the high-voltage DC bus. Each isolated dual active bridge module independently executes VDG control. The distributed consensus algorithm of virtual inertia is realized through a high-speed fiber optic ring network between isolated dual active bridge modules. Power balance and synchronous response of each module are required. The current sharing error between modules is selected according to the engineering accuracy requirements.

[0136] The grid connection unit includes an MMC controller, which has a built-in grid strength evaluator for real-time evaluation of the grid strength of the AC power grid.

[0137] To ensure the system's continuous grid-building capability around the clock, the MMC controller is equipped with a grid-building energy storage smoothing unit in the station power system: it is composed of a small-capacity supercapacitor and a large-capacity lithium battery. During sunset, it takes over the virtual inertia support. Its control adopts the same VDG protocol as mentioned above to achieve seamless switching of inertia sources and support the long-term continuous grid-building operation of the converter valve.

[0138] The MPPT controller, DCT controller, and MMC controller all employ a hardware-in-the-loop digital pre-simulation mechanism: before the actual issuance of control commands, the system state for a future duration is simulated and predicted in parallel within a field-programmable gate array. If the simulated DC bus voltage or current fluctuation exceeds a safety threshold, the key parameters in the original control command are automatically corrected, and the corrected command is then issued and executed. This proactively mitigates the oscillation risk caused by the weak damping of the desert power grid, enhancing the system's anti-oscillation capability in a weak desert power grid environment. The prediction error is selected based on the model accuracy.

[0139] The hierarchical collaborative control unit includes a blockchain light node authentication function: each regional controller stores the hash values ​​of key control parameters: virtual moment of inertia, virtual damping coefficient, and derating factor, to prevent malicious network attacks in desert areas from tampering with virtual inertia parameters, ensuring that control commands are trustworthy and traceable, and the blockchain confirmation time is selected according to security level requirements.

[0140] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "up," "down," "front," "rear," "left," "right," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only used to explain the relative positional relationship and movement between components in a specific posture. If the specific posture changes, the directional indication will also change accordingly. These terms are used only for the convenience of describing the invention and for simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting the invention.

[0141] Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0142] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0143] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A virtual synchronous grid-based collaborative control method for a large-scale photovoltaic DC collection system, characterized in that... This includes the following steps: Step 1: Obtain the grid support capability index of the photovoltaic string; wherein, the grid support capability index is calculated in real time during the MPPT control process of the photovoltaic string; Step 2: Based on the aforementioned grid support capability index, dynamically allocate virtual inertia to stabilize the DC bus voltage; and switch to DC voltage support mode when a grid fault ride-through signal is detected. Step 3: Obtain the power reserve request generated based on the AC power grid strength assessment; the power reserve request is generated based on the AC power grid strength assessment results; Step 4: Construct a hierarchical collaborative closed-loop control loop using the network support capability index and the power reserve request, and execute virtual synchronous network collaborative control.

2. The virtual synchronous grid-connected collaborative control method for a large-scale photovoltaic DC collection system as described in claim 1, characterized in that, In step one, after performing MPPT control on the photovoltaic string, a virtual capacitor response loop is superimposed to suppress voltage fluctuations in the photovoltaic string. The formula for the virtual capacitor response loop is: , in, This represents the change in the output current of the photovoltaic string. This indicates the output DC voltage of the MPPT. Indicates time; This represents the virtual capacitance coefficient, which is dynamically adjusted based on the string capacity. , This indicates the actual output power of the current photovoltaic string. This indicates the rated output power of the photovoltaic string. Indicates the reference virtual capacitance coefficient; A temperature-irradiance feedforward derating factor is used to limit the power output of photovoltaic strings when environmental conditions exceed a threshold. Represented as: ， in: Indicates real-time solar irradiance; Indicates reference solar irradiance; Indicates the junction temperature of the photovoltaic cell. Indicates external meteorological parameters; Indicates the temperature decay coefficient. Indicates the light intensity attenuation coefficient; The network support capability index for: This indicates the dynamic response potential of the photovoltaic string in this circuit; among which, This indicates the maximum power of the photovoltaic string in this circuit. This represents dynamic impedance.

3. The virtual synchronous grid-connected collaborative control method for a large-scale photovoltaic DC collection system as described in claim 1, characterized in that, In step two, the dynamic allocation of virtual inertia is achieved through VDG control, which includes voltage outer loop control and current inner loop control. The voltage outer loop control stabilizes the DC bus voltage at the rated target value by simulating the second-order swing equation of a synchronous generator. The second-order swing equation is as follows: , in: This represents the analog electromagnetic power, mapped to the output power of the photovoltaic string. ; This represents the analog mechanical power, mapped to the power output from the bus boost unit to the grid-connected unit. ; Indicates virtual angular displacement; The virtual damping coefficient is represented by J; the virtual moment of inertia is represented by J; the virtual moment of inertia is dynamically allocated based on the weighted sum of the grid support capacity indicators of each photovoltaic string. ,in Indicates the first One photovoltaic string, Indicates the first The grid support capability index of each photovoltaic string. This represents the weighting factor based on line impedance correction. ; Indicates the system reference impedance. This represents the measured value of the line impedance of this branch line; The current inner loop control utilizes dynamic virtual impedance to suppress multi-frequency oscillations, ensuring that the DC bus voltage accuracy is controlled within a set range; the dynamic virtual impedance is expressed as: The dynamic virtual impedance is realized through the real-time numerical representation of the time-domain virtual impedance expression: This is used to compensate for the resistive voltage drop and inductive effect of DC lines in desert areas, where: Indicates the impedance of the basic circuit. Indicates dynamic virtual incremental impedance. Indicates the magnitude of the load current. The low-frequency component representing the virtual impedance. The mid-frequency component representing the virtual impedance. High-frequency resonant term of virtual impedance, This represents a complex frequency variable.

4. The virtual synchronous grid-connected collaborative control method for a large-scale photovoltaic DC collection system as described in claim 3, characterized in that, The dynamic virtual incremental impedance is formed by the time-domain superposition of low-frequency components, mid-frequency components and high-frequency resonant terms; The low-frequency component of the virtual impedance The calculation formula is: ,in, THD represents the harmonic distortion rate of the medium-voltage DC bus voltage. This represents the compensation factor for the increase in impedance. Indicates the switching frequency of the device. Represents the virtual harmonic resistance component. Represents the virtual harmonic inductance component; The mid-frequency component of the virtual impedance The calculation formula is: ,in Represents the basic virtual inductance. Represents the adaptive coefficient. This indicates dynamic impedance, used to automatically increase inertia and prevent instability when the MPPT support capacity is weak; The high-frequency resonant term of the virtual impedance The calculation formula is: ,in: Indicates the resonant angular frequency. Indicates the damping ratio. It represents the resonant damping gain and is positively correlated with the peak THD.

5. The virtual synchronous grid-connected collaborative control method for a large-scale photovoltaic DC collection system as described in claim 1, characterized in that, In step two, the DC voltage support mode means adjusting the high-voltage side voltage control target from a constant DC voltage to a DC voltage that dynamically adjusts with power, in order to actively absorb power fluctuations and prevent power backfeeding from impacting the photovoltaic array. The high-voltage side voltage control formula is: ,in, Indicates the actual operating DC voltage. Indicates a constant DC voltage. This represents the voltage droop factor. This indicates fluctuations in grid power.

6. The virtual synchronous grid-connected collaborative control method for a large-scale photovoltaic DC collection system as described in claim 1, characterized in that, The AC grid strength is evaluated in real time based on the system short-circuit ratio (SCR) and the equivalent inertia time constant. When the SCR < 3 or the equivalent inertia time constant is less than 3s, the grid strength enhancement mode is triggered, and a power reserve request command is sent to the combiner boost unit. The formula for the equivalent inertia time constant is: ,in, This represents the total energy stored in the capacitors within the grid-connected layer. Indicates the system's rated capacity. Indicates the first The reference inertia value of each inertia unit. Indicates the first The inertia derating factor of the unit; when the equivalent inertia time constant is lower than the threshold, the system triggers a power reserve request, controlling the bus boost unit to increase the DC bus voltage within the fault detection time to reserve power. The power reserve request formula is: , This indicates the reserve power that needs to be stored. Indicates the current operating voltage. Indicates the voltage increase. This represents the equivalent bus capacitance.

7. The virtual synchronous grid-connected collaborative control method for a large-scale photovoltaic DC collection system as described in claim 6, characterized in that, In the enhanced grid construction capability mode, a VDG power synchronization control loop is introduced to realize the inertia coupling and transfer of AC and DC power grids. The VDG power synchronization control loop is used to convert the AC grid frequency deviation into a DC voltage correction value, as shown in the formula: ,in, Indicates the frequency deviation of the AC power grid. This indicates the DC voltage correction amount. Represents the time constant. Represents a complex frequency variable. Indicates the frequency droop factor; Dynamically adjust based on system inertia ,in, This represents the depreciation factor. Indicates the rated sag coefficient. Represents the equivalent inertia time constant. This represents the reference value of the equivalent inertia time constant.

8. The virtual synchronous grid-connected collaborative control method for a large-scale photovoltaic DC collection system as described in claim 1, characterized in that, In step four, the hierarchical collaborative closed-loop control performs multi-timescale coordination, including: At the first time scale of milliseconds, based on the instantaneous measurements of local voltage and current, MPPT control and virtual inertia adjustment of the photovoltaic strings are performed, and grid construction capability information is exchanged. At the second timescale of hundreds of milliseconds, the regional coordination control layer coordinates and allocates the virtual inertia distribution of each bus boost unit based on the regional power imbalance, with the objective function being: ,in, The penalty weighting coefficient representing the voltage deviation; Indicates the first The dynamic deviation of frequency at any given moment; Indicates the first DC voltage deviation at any given time; At the third timescale, on the order of seconds, based on power grid dispatch instructions, environmental early warning information, and equipment status data, full-level operational parameter optimization is performed. The objective function for optimization is: ， in, This represents the actual active power received. This represents the reference value for active power. This represents the reference value for the equivalent inertia; c1, c2, c3, and c4 represent weighting coefficients; T loss Dust represents the total system heat loss. risk This indicates the dust accumulation risk index.

9. The virtual synchronous grid-connected collaborative control method for a large-scale photovoltaic DC collection system as described in claim 8, characterized in that, It also includes dynamic switching strategies for extreme operating conditions: High-temperature derating coordination strategy: When the ambient temperature exceeds the derating threshold, the system uniformly issues a derating coefficient to reduce power output, while controlling the bus boost unit to increase the virtual damping coefficient to enhance system stability, and controlling the grid-connected unit to reduce the switching frequency of power devices to reduce losses. Sandstorm crossing strategy: The system predicts sandstorm events based on environmental monitoring data, including the sandstorm warning stage and the sandstorm arrival stage; During the sandstorm warning phase, the MPPT is controlled to switch to constant voltage mode to maintain low power operation; during the sandstorm approach phase, the faulty branch is isolated, and the grid-connected unit is controlled to maintain the DC bus voltage and output reactive power support to the AC grid. Smooth day-night switching strategy: At sunset, the MPPT is controlled to gradually reduce power according to the preset soft shutdown slope; the system controls the bus boost unit to transfer the virtual inertia to the station power energy storage system. The station power energy storage system receives the transferred inertia parameters and uses the same virtual synchronization protocol to maintain the bus voltage, realizing long-term continuous grid construction.

10. A large-scale photovoltaic DC collection system, characterized in that, include: A photovoltaic string unit includes multiple photovoltaic strings equipped with controllers, which are used to perform MPPT control and calculate grid support capability indicators; The bus boost unit is connected to the photovoltaic string unit and is used to collect the output power of the photovoltaic string unit and form a DC bus. It receives the grid support capability index and dynamically adjusts the virtual inertia based on the grid support capability index to stabilize the DC bus voltage. It also receives power reserve requests and boosts the DC bus voltage within the fault identification time to reserve power. The grid connection unit, connected to the busbar booster unit, is used to connect the DC busbar to the AC grid, assess the grid strength of the AC grid, and issue a power reserve request to the busbar booster unit based on the assessment results. The hierarchical collaborative control unit is connected to the photovoltaic string unit, the busbar booster unit and the grid-connected unit respectively, and is used to transmit the grid support capability indicators and power reserve requests.

11. The large-scale photovoltaic DC collection system as described in claim 10, characterized in that, The controller of the photovoltaic string unit is also used to perform virtual capacitor response control; The combiner boost unit includes a DC combiner box, a regional coordinator, and a DCT controller. The DC combiner box is used to collect the output power of the photovoltaic string units to the medium-voltage DC bus. The regional coordinator is used to dynamically allocate virtual inertia according to the grid support capability index. The DCT controller is used to connect the medium-voltage DC bus to the high-voltage DC bus and execute VDG control to maintain the stability of the DC bus voltage by adjusting the virtual rotational inertia and virtual damping. The grid-connected unit includes an MMC controller, which has a built-in grid strength evaluator.

Images