Network system coordination control method and device suitable for grid voltage drop

Abstract

This invention discloses a collaborative control method and apparatus for grid systems under voltage sag. The method includes: obtaining a reference value for the reactive current of the converter under low-voltage ride-through based on the voltage sag depth; obtaining a DC-side power correction based on the DC-side voltage deviation, and then combining it with the output power of new energy sources to obtain a reference value for the active power of the converter; when applying grid control based on a virtual synchronous generator to the converter, the power control loop references the converter reactive current reference value and the converter active power reference value to generate a reference value for the output voltage amplitude of the virtual synchronous generator under grid control. Using the above technical solution, the converter reactive current reference value adaptively changes with the voltage sag depth, avoiding adverse effects of sudden reactive power changes on system stability; the converter active power reference value adaptively changes with the DC bus voltage and the output power of new energy sources, suppressing DC bus overvoltage.

Description

Technical Field

[0001] This invention relates to the fields of grid-connected control for new energy power generation and power electronics technology, and in particular to a method and apparatus for coordinated control of grid systems applicable to grid voltage dips. Background Technology

[0002] With the rapid development of new energy power generation technologies, the installed capacity and penetration rate of new energy power generation, including photovoltaic power generation, in the power system continue to increase. The output power of new energy sources is significantly affected by environmental conditions, and large-scale grid connection places higher demands on the safe and stable operation of the power grid. Especially in scenarios with weak grids or high proportions of new energy connections, the grid's requirements for the voltage and frequency support capabilities of new energy sources are constantly increasing.

[0003] Grid-connected converters, by introducing an equivalent voltage source model at the control level, can actively establish and maintain AC side voltage and frequency, which has significant advantages in improving system stability and anti-disturbance capability.

[0004] In medium-voltage DC collector grid-connected new energy power generation systems, new energy power is collected to the medium-voltage DC bus after passing through a DC boost converter, and then connected to the AC grid through a grid-connected converter. This structure has advantages such as high collection efficiency and good system scalability, making it suitable for large-capacity photovoltaic power generation scenarios. However, under grid voltage drop conditions, medium-voltage DC collector systems face a coupling problem between DC-side power balance and AC-side reactive power support: when the photovoltaic array is operating at its maximum power point for a long time, under current capacity constraints, the grid-connected converter cannot simultaneously achieve active power transmission and reactive voltage support, which can easily lead to problems such as DC bus voltage fluctuations, grid current saturation, and deterioration of system dynamic performance.

[0005] In existing technologies, low voltage ride-through control for grid fault conditions mainly focuses on reactive current injection on the grid-connected side. It does not adequately consider the power reserve on the photovoltaic side, the energy coordination on the DC side, and the synergistic relationship between multiple links within the grid control. The power switching or limiting methods are too direct and drastic, resulting in large command abrupt changes during the fault process and an unsmooth system recovery process, which affects grid stability and safe operation of equipment. Summary of the Invention

[0006] Purpose of the invention: This invention provides a collaborative control method and device for grid-connected systems under voltage dips, aiming to solve the problem in existing medium-voltage DC collector grid-connected new energy power generation systems where large abrupt changes in control command and control system operating parameters occur when grid voltage dips, leading to drastic changes in system operation and adversely affecting grid connection stability and equipment safety.

[0007] Technical Solution: This invention provides a grid-connected system collaborative control method applicable to grid voltage sag, comprising: collecting the actual grid connection point voltage amplitude at the grid connection point between the modular multilevel converter and the grid, comparing it with the grid connection point voltage reference value to obtain the voltage sag depth; the grid-connected system includes: a new energy generation module, a modular multilevel converter, and a grid, wherein the modular multilevel converter is a grid-connected modular multilevel converter, with the DC side connected to the new energy generation module through a DC bus and the AC side connected to the grid; based on the voltage sag depth, determining the grid voltage deviation of the current grid voltage relative to the grid rated voltage; obtaining the reactive current correction amount based on the grid voltage deviation, and then superimposing the converter's rated reactive current to obtain the low voltage of the grid. The converter reactive current reference value under low-voltage crossing; when a grid voltage drop occurs, the converter reactive current reference value and the actual grid connection point voltage amplitude are used to calculate the converter reactive power reference value under low-voltage crossing; the actual DC bus voltage collected on the DC bus is compared with the DC bus voltage reference value to obtain the DC side voltage deviation, and the DC side power correction amount is obtained based on the DC side voltage deviation, and then combined with the output power of new energy sources to obtain the converter active power reference value; when applying grid-based control based on virtual synchronous generators to the converter, the power control loop uses the converter reactive current reference value and the converter active power reference value to generate the output voltage amplitude reference value of the virtual synchronous generator under grid-based control.

[0008] Specifically, the ratio of the deviation between the reference value of the grid connection point voltage and the actual voltage amplitude at the grid connection point to the reference value of the grid connection point voltage is used as the voltage drop depth.

[0009] Specifically, a voltage drop is considered to have occurred when the voltage drop depth exceeds a corresponding threshold.

[0010] Specifically, the deviation between the per-unit value and the voltage drop depth when the actual grid connection point voltage is at the rated value is calculated as the grid voltage deviation.

[0011] Specifically, after the grid voltage deviation is limited, the reactive current correction amount is obtained by linear mapping through a proportional coefficient.

[0012] Specifically, the DC-side voltage deviation is input into the PI controller to obtain the DC-side power correction amount. Based on the correction amount of the DC-side power correction, the output power of the new energy source is corrected to obtain the active power reference value of the converter.

[0013] Specifically, the actual system frequency of the converter under grid control is compared with the rated system frequency to obtain the system frequency deviation; the active power correction of the new energy source is obtained based on the system frequency deviation; the power dispatch deviation of the new energy source is obtained based on the deviation between the output power of the new energy source and the reference active power of the new energy source, and the correction of the deviation by the active power correction of the new energy source; the duty cycle correction of the new energy generation module at the next moment is calculated based on the power dispatch deviation of the new energy source, and the duty cycle at the current moment is corrected to obtain the duty cycle of the new energy generation module at the next moment.

[0014] Specifically, if a voltage drop in the power grid is detected, the duty cycle correction for the new energy power generation module at the next moment is set to 0.

[0015] Specifically, the sign of the duty cycle correction of the new energy power generation module at the next moment is determined based on the sign of the product between the new energy output voltage increment and the new energy output power increment.

[0016] This invention also provides a grid-connected system collaborative control device suitable for grid voltage sag, comprising: a voltage sag calculation unit, a reactive power reference value calculation unit, an active power reference value calculation unit, and an execution unit, wherein: the voltage sag calculation unit is used to collect the actual grid connection point voltage amplitude at the grid connection point between the modular multilevel converter and the grid, compare it with the grid connection point voltage reference value, and obtain the voltage sag depth; the grid-connected system includes: a new energy generation module, a modular multilevel converter, and a grid, wherein the modular multilevel converter is a grid-connected modular multilevel converter, with the DC side connected to the new energy generation module through a DC bus and the AC side connected to the grid; the reactive power reference value calculation unit is used to determine the grid voltage deviation between the current grid voltage and the grid rated voltage based on the voltage sag depth; and to obtain the reactive current correction amount based on the grid voltage deviation. The converter's rated reactive current is then superimposed to obtain the converter's reactive current reference value under low-voltage grid ride-through. When a grid voltage drop occurs, the converter's reactive current reference value and the actual grid connection point voltage amplitude are used to calculate the converter's reactive power reference value under low-voltage grid ride-through. The active power reference value calculation unit is used to compare the actual DC bus voltage collected on the DC bus with the DC bus voltage reference value to obtain the DC side voltage deviation. Based on the DC side voltage deviation, the DC side power correction amount is obtained, and then combined with the output power of new energy sources, the converter's active power reference value is obtained. The execution unit is used to generate the output voltage amplitude reference value of the virtual synchronous generator under grid control when applying grid control based on a virtual synchronous generator to the converter. The power control loop references the converter's reactive current reference value and the converter's active power reference value.

[0017] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: the reactive current reference value of the converter adapts to the voltage drop depth, avoiding the adverse effects of sudden reactive power changes on system stability; the active power reference value of the converter adapts to the DC bus voltage and the output power of new energy sources, suppressing DC bus overvoltage; the synergy of the new energy power generation module, the DC side of the converter, and the grid control achieves DC bus voltage stability. Attached Figure Description

[0018] Figure 1 The main circuit and control structure diagram of the medium-voltage DC collector grid-connected new energy power generation system provided by the present invention; Figure 2 A schematic diagram illustrating the determination of low voltage ride-through (LVRT) state provided by the present invention; Figure 3 A flowchart of the perturbation observation method provided by the present invention; Figure 4 A schematic diagram of photovoltaic active power reserve control based on frequency feedback provided by the present invention; Figure 5 A schematic diagram illustrating the reactive power reference value calculation method based on voltage sag depth provided by the present invention; Figure 6 A schematic diagram illustrating the active power reference value calculation method based on DC bus voltage constraint provided by the present invention; Figure 7 A schematic diagram of active power and frequency control in the power control loop provided by the present invention; Figure 8 A schematic diagram of reactive power and voltage control in the power control loop provided by the present invention; Figure 9 The diagram shows the simulation results of the output current and output voltage under the cooperative control scheme provided by this invention when the grid voltage drops by 40% in 2 to 3 seconds. Figure 10 This is a schematic diagram of the output current simulation results when the grid voltage drops by 40% in 2 to 3 seconds without the use of coordinated control. Figure 11 A schematic diagram showing the simulation results of active power, reactive power, DC bus voltage and frequency under two control schemes for a grid voltage drop of 40% in 2 to 3 seconds. Figure 12 A schematic diagram of the simulation results of capacitor current and capacitor voltage of the MMC submodule of the system under cooperative control for a 40% voltage drop in grid voltage over 2 to 3 seconds. Figure 13 This is a schematic diagram of the simulation results of the capacitor current and capacitor voltage of the MMC submodule of the system without the use of collaborative control when the grid voltage drops by 40% in 2 to 3 seconds. Detailed Implementation

[0019] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0020] See Figure 1 This is the main circuit and control structure diagram of the medium-voltage DC collector grid-connected new energy power generation system provided by the present invention.

[0021] In practice, Figure 1 The medium-voltage DC collector grid-connected new energy power generation system (grid system) shown includes: a distributed photovoltaic array (photovoltaics are used as a new energy source in this invention), a DC boost converter, a medium-voltage DC bus, a grid-connected modular multilevel converter, AC side filtering and grid connection lines, a common coupling point, a virtual synchronous generator power and current dual closed-loop control module, a low voltage ride-through detection and power coordination control module, and a carrier phase shift modulation module.

[0022] In practical implementation, the grid-connected converter is a three-phase modular multilevel converter (MMC). Its DC side is connected to a medium-voltage DC collector bus (DC bus), and its AC side is connected to the AC grid through a grid-connected reactor. The grid-connected converter includes three phase units: A, B, and C. Each phase unit consists of an upper arm and a lower arm, which are connected through the output terminal on the AC side. Arm reactors are connected in series at the connection point to limit circulating current and improve system dynamic characteristics. Each arm consists of several sub-modules (SMs) connected in series. The sub-modules (SMs) can adopt a half-bridge structure and internally contain power switching devices and sub-module capacitors. By controlling the activation and bypass states of the sub-modules, the synthesis of multi-level output voltages can be achieved.

[0023] In practice, multiple photovoltaic arrays are connected to the medium-voltage DC bus via their respective DC-DC boost converters, and the DC bus provides energy input to the converters. The converters operate in grid-connected control mode, independently establishing and maintaining the voltage and frequency of the grid connection point PCC, and are connected to the grid through AC-side filter reactors. pcc and i pcc These are the grid connection point voltage and grid connection point current, respectively, P e and Q e These are the actual active power output of the converter and the actual reactive power output of the converter, respectively.

[0024] See Figure 2 This is a schematic diagram of the low voltage ride-through (LVRT) state determination provided by the present invention.

[0025] In this embodiment of the invention, the actual grid connection point voltage amplitude collected by the PCC at the grid connection point between the modular multilevel converter and the power grid is compared with the grid connection point voltage reference value to obtain the voltage drop depth.

[0026] In this embodiment of the invention, the ratio between the deviation between the grid connection point voltage reference value and the actual grid connection point voltage amplitude and the grid connection point voltage reference value is used as the voltage drop depth.

[0027] In this embodiment of the invention, a grid voltage drop is identified when the voltage drop depth is greater than a corresponding threshold.

[0028] In practical implementation, the grid connection point voltage u pcc After processing (Clark Transformation and absolute value calculation), the actual grid-connected point voltage amplitude V is obtained. pcc Calculate the reference value V of the grid connection point voltage. pcc,ref The deviation △V between them: △V=|V pcc -V pcc,ref |

[0029] The ratio between the deviation ΔV and the grid connection point voltage reference value (which can be set according to the actual application scenario) is used as the voltage drop depth ΔU. LVRT,depth : △U LVRT,depth =△V / V pcc,ref .

[0030] In practice, voltage dip depth can be used to measure the degree of voltage dip in the power grid. In practice, the power grid voltage will have a certain degree of normal fluctuation. Therefore, in order to avoid the wrong identification of low voltage faults in the power grid due to normal fluctuations, a corresponding threshold is set for voltage dip depth. When the voltage dip depth is greater than the threshold, it is identified that a power grid voltage dip and low voltage fault have occurred.

[0031] See Figure 5 This is a schematic diagram of the reactive power reference value calculation method based on voltage drop depth provided by the present invention.

[0032] In this embodiment of the invention, the voltage deviation of the current grid voltage relative to the grid rated voltage is determined based on the voltage drop depth.

[0033] In this embodiment of the invention, the deviation between the per-unit value and the voltage drop depth when the actual grid connection point voltage is the rated value is calculated as the grid voltage deviation.

[0034] In practice, the actual grid connection point voltage is the rated value, and the reference value is usually also directly adopted as the rated value. In this case, the per-unit value is 1, then the grid voltage deviation ΔU = 1 - ΔU LVRT,depth .

[0035] In practical implementation, when the voltage drop depth is obtained based on the ratio between the deviation ΔV and the grid connection point voltage reference value (proportional calculation), then based on the deviation between the voltage drop depth and the per-unit value (when the actual grid connection point voltage is the rated value), the grid voltage deviation ΔU can characterize the remaining degree of the grid current voltage relative to the grid rated voltage. The converter reactive power reference value ultimately changes adaptively mainly based on the change of this parameter.

[0036] In this embodiment of the invention, after the grid voltage deviation is limited, the reactive current correction amount is obtained by linear mapping through a proportional coefficient.

[0037] In practice, the purpose of limiting is to prevent excessively high reactive power reference values ​​during voltage transient disturbances or significant voltage drops. There are various limiting methods, such as multiplying a coefficient by the grid voltage deviation or setting an upper limit, which can be configured according to the specific application.

[0038] In this embodiment of the invention, the reactive current correction amount is obtained based on the grid voltage deviation, and then the rated reactive current of the converter is superimposed to obtain the reference value of the converter reactive current under low-voltage ride-through of the grid.

[0039] In practical implementation, the reactive current reference value I q,LVRT The calculation is as follows: I q,LVRT =K qi △U+ K qi1 I n , Among them, K qi K represents the proportionality coefficient (i.e., the reactive power droop coefficient). qi △U is the reactive current correction, I n K represents the rated reactive current of the converter, which is the current reference flowing from the grid to the power grid and is used to constrain the upper limit of grid-connected reactive power output. qi1 It represents the rated current weighting factor, used to enhance reactive power support capability under large disturbance conditions.

[0040] In this embodiment of the invention, when a grid voltage drop occurs, the converter reactive current reference value and the actual grid connection point voltage amplitude are used to calculate the converter reactive power reference value under low-voltage grid ride-through.

[0041] In practical implementation, considering the proportional relationship between reactive power, reactive current, and grid connection point voltage amplitude in the synchronous rotating coordinate system, the converter reactive current reference value is multiplied by the actual collected grid connection point voltage amplitude, and after coefficient conversion, the converter reactive power reference value Q for the low-voltage ride-through period when a grid voltage drop occurs is generated. ref,LVRT The calculation formula is as follows: Q ref,LVRT=(3 / 2)(V pcc I q,LVRT ), Among them, Q ref,LVRT This indicates the reference value of the converter reactive power during the low-voltage ride-through period when a grid voltage drop occurs.

[0042] In practical implementation, it is important to note that the converter reactive power reference value is calculated under grid control only when a grid voltage dip occurs and low-voltage ride-through is required. When no grid voltage dip occurs, the converter reactive power reference value Q is... ref Take 0.

[0043] See Figure 6 This is a schematic diagram of the active power reference value calculation method based on DC bus voltage constraint provided by the present invention.

[0044] In this embodiment of the invention, the actual DC bus voltage collected on the DC bus is compared with the DC bus voltage reference value to obtain the DC side voltage deviation. Based on the DC side voltage deviation, the DC side power correction amount is obtained, and then combined with the output power of new energy sources, the active power reference value of the converter is obtained.

[0045] In this embodiment of the invention, the DC-side voltage deviation is input to the PI controller to obtain the DC-side power correction amount. Based on the correction of the output power of the new energy source, the active power reference value of the converter is obtained.

[0046] In practical implementation, the DC side voltage deviation e vdc The calculation is as follows: e vdc =V dc,ref -V dc , Among them, V dc This represents the actual DC bus voltage, V. dc,ref This indicates the reference value for the DC bus voltage.

[0047] In practical implementation, the DC side voltage deviation e vdc Characterizing the energy balance state of the DC bus, the DC side voltage deviation e vdc The DC-side power correction ΔP is obtained by inputting the PI controller. vdc .

[0048] In practical implementation, the output power P of the new energy source pv (In this embodiment of the invention, photovoltaic) can be calculated using the output voltage and output current of the new energy source.

[0049] In practical implementation, the converter active power reference value P ref It is generated jointly by the DC-side power correction and the output power of new energy sources, as shown in the following formula: P ref =P pv -△P vdc .

[0050] In practice, under grid control, the converter active power reference value calculated above is executed regardless of whether a grid voltage drop occurs.

[0051] In practical implementation, the converter's active power reference value can adaptively change with the DC bus voltage and the output power of renewable energy sources, overcoming the problems of DC bus voltage rise and sudden changes in renewable energy output power during grid voltage dips, thus improving the safety and stability of the medium-voltage DC collector grid-connected renewable energy power generation system under fault conditions. Simultaneously, even during periods of normal grid voltage, the converter's active power reference value can adaptively adjust to an appropriate value, effectively transmitting renewable energy power to the grid.

[0052] In practice, the calculation of the converter's active power reference value is achieved through the collaboration of the new energy power generation module and the DC side.

[0053] See Figure 4 This is a schematic diagram of the photovoltaic active power reserve control based on frequency feedback provided by the present invention.

[0054] In this embodiment of the invention, the actual system frequency of the converter under grid control is compared with the rated system frequency to obtain the system frequency deviation; the active power correction amount of new energy is obtained based on the system frequency deviation.

[0055] In practical implementation, the actual system frequency is compared with the rated system frequency, and the calculation formula is as follows: △f=f n -f, Where Δf represents the system frequency deviation, f n f represents the rated system frequency, and f represents the actual system frequency.

[0056] In practice, the system frequency deviation can be amplified by the frequency power proportionality coefficient to form the active power correction amount ΔP of the new energy source. f The calculation formula is as follows: △P f =K f △f, Among them, K f This represents the frequency-power ratio coefficient.

[0057] In this embodiment of the invention, the new energy power scheduling deviation is obtained based on the deviation between the output power of the new energy source and the reference active power of the new energy source, as well as the correction of the deviation by the active power correction amount of the new energy source.

[0058] In practical implementation, the active power correction amount △P of new energy sources f It represents the adjustment amount of frequency support demand feedback to power. Combined with the active power of new energy benchmark, it is used to adjust the output power of new energy within the allowable range, so that the new energy side participates in the active power regulation and control of the grid system. In other words, the new energy power scheduling deviation represents the adjustment and correction of the output power of new energy.

[0059] In practical implementation, the formula for calculating the deviation of new energy power dispatch is as follows: P cmd =P pv -P base -△P f , Among them, P cmd P represents the deviation in the dispatch of renewable energy power. pv P represents the output power of the new energy source. base This indicates the benchmark active power of new energy sources.

[0060] See Figure 3 This is a flowchart of the perturbation observation method provided by the present invention.

[0061] In practical implementation, the benchmark active power P of new energy sources base The settings are based on maximum power point tracking achieved through the perturbation observation method.

[0062] In practical implementation, let the output power and output voltage of the new energy source at time k-1 and time k be P, respectively. pv (k-1), P pv (k), V pv (k-1) and V pv (k), based on Figure 3 The perturbation observation method shown calculates the maximum output power P of the new energy power generation module in real time under the current environmental conditions such as light intensity and temperature. mppt And P cmd It can be set to P cmd <P mppt Generally speaking, P cmd It can be set close to P mppt This is used to continuously increase the output power of the new energy power generation module during subsequent duty cycle updates, thereby achieving the maximum output power P. mppt This ensures that the operating point is near the maximum power point.

[0063] In this embodiment of the invention, the duty cycle correction amount of the new energy power generation module at the next moment is calculated based on the new energy power dispatch deviation, and the duty cycle at the current moment is corrected to obtain the duty cycle of the new energy power generation module at the next moment.

[0064] In practical implementation, the renewable energy power dispatch deviation represents the difference between the current renewable energy output power and the sum of the reference power and frequency support requirements. When the actual renewable energy output power is higher than the reference power and frequency support requirements, the power dispatch deviation is positive, generating a corresponding duty cycle correction signal (via PWM) to reduce the photovoltaic output power. When the actual renewable energy output power is lower than the reference power and frequency support requirements, the power dispatch deviation is negative, generating a duty cycle correction signal in the opposite direction to increase the photovoltaic output power. Through this mechanism, the dynamic response of renewable energy output power to the frequency support requirements represented by the system frequency deviation Δf is achieved.

[0065] In practical implementation, the formula for calculating the duty cycle correction at the next moment is as follows: △D(k)=K D y(k)P cmd (k), Where △D(k) represents the duty cycle correction at time k (time k is used to characterize the next time step), K D y(k) represents the duty cycle adjustment step size coefficient, used to limit the single adjustment amplitude to ensure the smoothness and stability of the control process. y(k) represents the sign determination at time k. cmd (k) represents the new energy power dispatch deviation at time k.

[0066] In this embodiment of the invention, the sign of the duty cycle correction of the new energy power generation module at the next moment is determined based on the sign of the product between the new energy output voltage increment and the new energy output power increment.

[0067] In practical implementation, to determine the direction of duty cycle adjustment, this invention introduces a direction discrimination signal based on disturbance observation. By calculating the sign of the product of the photovoltaic side voltage increment and power increment, a sign discrimination function y(k) is constructed: y(k)=sign(dV×dP), Here, the function sign represents the sign function, and dV and dP represent the increment of the new energy output voltage and the increment of the new energy output power, respectively. Mathematically, these can be obtained by calculating the derivatives of the new energy output voltage and the new energy output power.

[0068] In practical implementation, when dV×dP is greater than 0, it indicates that the current operating point of the new energy source is on the same side as the maximum power point, and the output direction signal is positive; when dV×dP is less than 0, it indicates that the operating point has passed the maximum power point, and the output direction signal is negative. This achieves adaptive discrimination of the duty cycle correction direction.

[0069] In practical implementation, during the duty cycle update process, the duty cycle correction for the next time step is superimposed with the current duty cycle to form a new duty cycle command. Based on the duty cycle for the next time step, this command affects the output power of the new energy source. The discrete update relationship is as follows: D(k) = D(k-1) + ΔD(k), Where D(k) represents the duty cycle at time k, which represents the duty cycle at the next time step, and D(k-1) represents the duty cycle at time k-1, which represents the duty cycle at the current time step.

[0070] In practical implementation, under normal operating conditions, the operating point usually corresponds to the vicinity of the maximum power point. However, after introducing frequency support demand (i.e., Δf system frequency deviation versus the new energy power dispatch deviation P), the operating point becomes more critical. cmd The impact of this, which in turn affects the duty cycle and ultimately the output power of the new energy source, can be determined by the reference power P. base and frequency adjustment requirements (by ΔP) f The characteristics of the new energy source are jointly determined, thereby achieving orderly regulation of the output power of the new energy source.

[0071] In this embodiment of the invention, if a voltage drop in the power grid is detected, the duty cycle correction amount of the new energy power generation module at the next moment is set to 0.

[0072] In practice, under normal operating conditions, the duty cycle is continuously adjusted according to the aforementioned update mechanism to ensure that renewable energy operates within its optimal operating range while meeting power dispatch requirements. During low-voltage ride-through, the duty cycle remains unchanged, maintaining the stable value from the moment before the low-voltage fault occurred. This is to prevent the photovoltaic power from continuing to rise during grid voltage dips due to attempts to increase renewable energy output, which could lead to a rise in DC bus voltage or even overvoltage risks.

[0073] In this embodiment of the invention, when applying grid-based control based on a virtual synchronous generator to the converter, the power control loop references the converter reactive current reference value and the converter active power reference value to generate the output voltage amplitude reference value of the virtual synchronous generator under grid-based control.

[0074] In practical implementation, in order to enable the medium-voltage DC collector grid-connected new energy power generation system to have the ability to autonomously establish voltage and frequency similar to a synchronous generator, a grid-connected control scheme based on a virtual synchronous generator (VSG) is introduced into the grid-connected converter.

[0075] In practical implementation, VSG control adopts a dual closed-loop structure of power and current. The outer loop is a power control loop based on a virtual synchronous generator model, used to generate system voltage phase angle and amplitude references. The inner loop is a current closed-loop control loop, used to quickly adjust the converter output current to track the commands given by the outer loop. Overall, this enables the converter to achieve grid-connected operation without relying on an external power grid.

[0076] See Figure 7 This is a schematic diagram of active power and frequency control in the power control loop provided by the present invention.

[0077] The following section is the active power and frequency control section.

[0078] In practical implementation, the active power control channel is used to simulate the rotor motion characteristics of a synchronous generator. Frequency deviation feedback and a converter active power reference value P are introduced. ref This forms a virtual mechanical input power P. m : P m =P ref +K p (ω ref -ω), Among them, K p ω represents the active frequency proportionality coefficient. ref ω and ω represent the system reference angular frequency and the system current angular frequency, respectively.

[0079] The rotational dynamics of the virtual synchronous generator are described by the following equivalent oscillation equation: J(dω / dt)=P m -P e -D(ω-ω0), Where J represents the virtual moment of inertia coefficient, D represents the virtual damping coefficient, and ω0 represents the system's rated angular frequency. This equation allows for the introduction of equivalent inertia and damping effects during power disturbances or load changes, suppressing rapid frequency variations.

[0080] Integrating the current angular frequency ω of the system after the above correction process (1 / s in the figure represents the integration operation) yields the system voltage phase angle θ. The system voltage phase angle θ serves as the phase reference for the output voltage of the grid-type converter, used for subsequent voltage reference generation and coordinate transformation.

[0081] See Figure 8 This is a schematic diagram of reactive power and voltage control in the power control loop provided by the present invention.

[0082] The following section is the reactive power and voltage control section.

[0083] The reactive power control channel is used to simulate the excitation regulation process of a synchronous generator, achieving grid-based control of voltage amplitude. The converter reactive power reference value Q is used. ref The actual output reactive power Q of the converter e The resulting deviation is compared and then processed by an integral adjustment mechanism: E Q =(K q / s)( Q ref -Q e ); E U =D q (U n -U o ), Where 1 / s represents the integration operation, E Q K represents the excitation compensation correction amount. q E represents the integral adjustment gain coefficient. U D represents the voltage deviation droop compensation component. q U represents the reactive voltage droop damping coefficient. n U represents the rated reference voltage of the AC system. o This represents the actual output voltage at the grid connection point. The converter reactive power reference value Q... ref Take Q when a grid voltage drop occurs. ref,LVRT The value is 0 when there is no voltage drop in the mains.

[0084] The adjustment value obtained above is superimposed on the reference voltage to form the reference value E of the virtual synchronous generator output voltage amplitude: E=E0+ E Q + E U , Where E0 represents the no-load reference voltage amplitude of the virtual synchronous generator.

[0085] The following section is the dual closed-loop control section for voltage and current.

[0086] The virtual synchronous generator control uses the output voltage amplitude reference value E generated by the outer loop and the corrected system voltage phase angle θ as the basis for constructing a synchronous rotating coordinate system. By performing coordinate transformation on the converter output voltage and output current, the voltage components (v) in the dq coordinate system can be obtained. d and v q ) and current component (i d and i q ), and calculate the actual output active power P of the converter. e and the actual output reactive power Q of the converter e : P e =(3 / 2)(v d i d+v q i q ); Q e =(3 / 2)( v q i d -v d i q ).

[0087] In the network coordinate system, the voltage vector is usually aligned with the d-axis, so that v q Approximately equal to 0, thus the current reference value can be calculated based on the power command: i d ref =(2 / 3)(P ref / v d ), i q ref =-(2 / 3)(Q ref / v d ).

[0088] To verify the dynamic response characteristics of the control method proposed in this invention under strong disturbance conditions, based on the established simulation model of the medium-voltage DC collector grid photovoltaic power generation system, time-domain simulation analysis was conducted on the key electrical quantities of the system during the entire process of grid voltage drop and fault recovery.

[0089] This embodiment sets up the following disturbance scenario: the grid voltage experiences a symmetrical drop of -0.4 pu during 2 to 3 seconds, and recovers to its rated value after 3 seconds. The rated DC bus voltage is set to 3kV, and the rated AC line voltage is 1500V. Simulation comparisons are performed on two schemes: the LVRT collaborative control method described in this invention and the traditional VSG network control method alone. See Figure 9 The output current and output voltage simulation results are shown in the diagram, which is based on the grid voltage dropping by 40% in 2 to 3 seconds and the collaborative control scheme provided by this invention.

[0090] Figure 9 The output current and voltage response were demonstrated. During steady-state operation, the inverter output current exhibited a standard sinusoidal waveform with stable amplitude. During a 2-second voltage drop, the output current experienced a brief fluctuation, but the waveform maintained its sinusoidal characteristics without significant harmonic distortion. Upon fault recovery at 3 seconds, the current only experienced a brief overshoot before quickly recovering to its steady-state amplitude, verifying the controllability and waveform quality of the inverter output current. During steady-state operation, the inverter output voltage also exhibited a standard sinusoidal waveform with stable amplitude. During a 2-second voltage drop, the output voltage amplitude briefly decreased, but the waveform maintained its sinusoidal characteristics. Upon fault recovery at 3 seconds, the voltage quickly recovered to its rated value without significant harmonic distortion, verifying the stability and dynamic adjustment capability of the inverter output voltage.

[0091] See Figure 10 The diagram shows the simulation results of the output current when the grid voltage drops by 40% in 2 to 3 seconds without the use of cooperative control, as provided by this invention.

[0092] like Figure 10 The figure shows the output current response under the same operating conditions without the coordinated control method. When a voltage drop occurs, although the system can maintain the basic voltage output, a significant current surge and oscillation process occurs at the moment of fault recovery. Specifically, at the 3-second voltage recovery instant, the peak output current is significantly higher than the steady-state value, accompanied by obvious damped oscillations, and the recovery time is relatively long. This phenomenon indicates that without a coordinated mechanism, the grid-connected converter's ability to output active power suddenly increases when the grid voltage recovers, while the photovoltaic side maintains a high power injection. The energy accumulated on the DC side is released in a short time, leading to a momentary overshoot of active power, which in turn triggers current surges and power angle oscillations. This dynamic process reflects that the coupling relationship between DC-side energy and AC-side power has not been effectively coordinated under traditional control.

[0093] By comparison Figure 9 and Figure 10 It can be seen that the collaborative control method described in this invention has significant advantages in the entire process of grid voltage dip and recovery: First, during voltage dips, it can prioritize the allocation of current capacity for reactive power support while limiting active power output, thereby achieving grid-connected current-constrained operation; Second, during the fault recovery phase, through the smooth transition mechanism of active power reference value, it significantly reduces the current overshoot amplitude and oscillation degree, and improves the dynamic stability of the system; Third, through the closed-loop control of DC bus voltage and coordinated control of active power, it effectively suppresses the recovery impact risk caused by DC side energy accumulation.

[0094] See Figure 11 The diagram shows the simulation results of active power, reactive power, DC bus voltage and frequency under two control schemes, where the grid voltage drops by 40% in 2 to 3 seconds.

[0095] The figure shows the dynamic response curves of active power, reactive power, DC bus voltage, and system frequency from 0 to 6 seconds. The voltage drop phase is from 2 to 3 seconds, and the voltage recovers to the rated value at 3 seconds.

[0096] like Figure 11The comparison of active power response is shown. Without coordinated control, the active power exhibits a significant negative overshoot during voltage recovery, with a peak deviation exceeding 30% of the rated power, accompanied by a continuous oscillation process. The power recovery time is long, and the dynamic process damping is insufficient. In contrast, with the coordinated control scheme proposed in this invention, the active power actively drops and remains within the restricted range during voltage dips, smoothly recovers during the recovery phase, the maximum overshoot amplitude is controlled within 5% of the rated power, the oscillation decays rapidly, and the power recovery process is continuously controllable, significantly improving the power surge problem during the fault exit phase.

[0097] like Figure 11 The reactive power response is shown in the comparison. Without coordinated control, a significant reverse impact occurs at the moment of voltage recovery, the reactive power swing is large, the dynamic oscillation duration is long, and it is prone to voltage fluctuations. When the control scheme provided by this invention is adopted, the reactive power continuously maps to generate a reference value based on the voltage drop depth, rapidly increases during the drop period to support the grid voltage, and smoothly drops back during the recovery phase. No significant reverse impact occurs, the dynamic process is stable, and the effectiveness of the reactive power priority allocation mechanism is verified.

[0098] like Figure 11 The comparison of DC bus voltage response is shown. During steady-state operation, the DC bus voltage stabilizes at approximately 3000V. With the traditional control scheme, significant fluctuations occur in the DC bus voltage during both the grid voltage dip and recovery phases, especially a large downward surge during the 3-second voltage recovery, resulting in a significantly increased voltage deviation accompanied by some oscillations. This indicates that a relatively violent energy exchange occurs on the DC side during the fault recovery phase. In contrast, with the cooperative control scheme provided by this invention, the DC bus voltage experiences only minor disturbances during the voltage dip phase and can quickly recover to near the 3000V steady-state value. During the voltage recovery phase, voltage fluctuations are significantly reduced, the transition process is smoother, and the system can re-stabilize at the rated DC voltage level in a shorter time. Simulation results show that the method provided by this invention can effectively suppress the rapid accumulation and release of DC side energy during low-voltage ride-through, thereby reducing DC bus voltage fluctuations and improving the DC side operational stability of the system.

[0099] like Figure 11 The system frequency response comparison is shown. Without coordinated control, the frequency drops to around 48Hz at the moment of voltage recovery, with a frequency deviation exceeding 2Hz and accompanied by obvious oscillations. However, with the control scheme provided by this invention, the frequency deviation is controlled within 0.1Hz and quickly converges to the rated frequency, significantly enhancing system damping.

[0100] See Figure 12 The figure shows the simulation results of the capacitor current and capacitor voltage of the MMC submodule of the system under collaborative control, which is based on the grid voltage dropping by 40% in 2 to 3 seconds.

[0101] like Figure 12 Simulation results of the system's MMC submodule capacitor current and voltage were presented when the grid voltage dropped by 40% over 2 to 3 seconds. During the 2-second voltage drop, the capacitor current experienced a brief, large fluctuation but quickly converged to the steady-state range. Upon fault recovery at 3 seconds, the current only fluctuated briefly before stabilizing rapidly, verifying the dynamic controllability of the submodule capacitor current. During steady-state operation, the submodule capacitor voltage remained around 500V with fluctuations of less than 5%. During the 2-second voltage drop, the capacitor voltage experienced a brief spike before quickly returning to the steady-state range; upon fault recovery at 3 seconds, the voltage only dropped briefly before quickly returning to steady state, verifying the stability of the submodule capacitor voltage and the effectiveness of the voltage equalization control scheme.

[0102] See Figure 13 The diagram shows the simulation results of the capacitor current and capacitor voltage of the MMC submodule of the system when the grid voltage drops by 40% in 2 to 3 seconds without the use of collaborative control.

[0103] Figure 13 As shown by the submodule capacitor current, at the instant of voltage recovery, the submodule capacitor current generates a significant impact and is accompanied by continuous oscillation. The oscillation decay time is significantly prolonged, which indicates that the energy inside the bridge arm undergoes a concentrated redistribution phenomenon during the recovery phase. Figure 13 As shown by the submodule capacitor voltage, the submodule capacitor voltage exhibits significant overshoot and undershoot during voltage recovery. The voltage fluctuation amplitude is significantly higher than the response level when using the cooperative control strategy, and the oscillation duration is relatively long, reflecting the amplification effect of the power mismatch between the DC and AC sides at the submodule level.

[0104] Analysis of the internal energy mechanism of the MMC reveals that traditional VSG control fails to coordinate photovoltaic power and DC energy during voltage dips, leading to energy accumulation in the bridge arm capacitors. Upon voltage recovery, the active power surges instantaneously, causing a dramatic redistribution of energy in the bridge arms, manifested as capacitor current surges and capacitor voltage oscillations. This invention, by implementing an active power voltage drop during grid voltage dips, constrains the rate of change of DC energy, thereby suppressing circulating current fluctuations within the bridge arms and submodule capacitor voltage offsets. Figure 12 and Figure 13 It can be seen that after adopting the collaborative control scheme proposed in this invention, the impact amplitude of the submodule capacitor current is significantly reduced, the fluctuation amplitude of the capacitor voltage is significantly reduced, the oscillation decay time is shortened, and the energy distribution inside the MMC is more balanced.

[0105] In summary, the simulation results demonstrate that, under the condition of a sustained -0.4 pu voltage dip in the power grid, the collaborative control method for a medium-voltage DC-DC grid-connected photovoltaic power generation system proposed in this invention can effectively coordinate the active power reserve control on the photovoltaic side, the closed-loop regulation of the DC bus voltage, and the dual closed-loop power and current control scheme of the grid-connected MMC. During the voltage dip, the system achieves grid-connected current-limited operation through reactive power priority support and active power voltage reduction, while suppressing the rise of the DC bus voltage. During the fault recovery phase, the active power reference value gradually recovers, effectively avoiding current surges and power spikes. Simulation results show that the control method can maintain system operational stability throughout the entire process of grid voltage dips and recovery, improve the grid connection point voltage support capability and DC side energy balance level, and has good engineering application value.

[0106] This invention also provides a grid-connected system collaborative control device suitable for grid voltage sag, comprising: a voltage sag calculation unit, a reactive power reference value calculation unit, an active power reference value calculation unit, and an execution unit, wherein: the voltage sag calculation unit is used to collect the actual grid connection point voltage amplitude at the grid connection point between the modular multilevel converter and the grid, compare it with the grid connection point voltage reference value, and obtain the voltage sag depth; the grid-connected system includes: a new energy generation module, a modular multilevel converter, and a grid, wherein the modular multilevel converter is a grid-connected modular multilevel converter, with the DC side connected to the new energy generation module through a DC bus and the AC side connected to the grid; the reactive power reference value calculation unit is used to determine the grid voltage deviation between the current grid voltage and the grid rated voltage based on the voltage sag depth; and to obtain the reactive current correction amount based on the grid voltage deviation. The converter's rated reactive current is then superimposed to obtain the converter's reactive current reference value under low-voltage grid ride-through. When a grid voltage drop occurs, the converter's reactive current reference value and the actual grid connection point voltage amplitude are used to calculate the converter's reactive power reference value under low-voltage grid ride-through. The active power reference value calculation unit is used to compare the actual DC bus voltage collected on the DC bus with the DC bus voltage reference value to obtain the DC side voltage deviation. Based on the DC side voltage deviation, the DC side power correction amount is obtained, and then combined with the output power of new energy sources, the converter's active power reference value is obtained. The execution unit is used to generate the output voltage amplitude reference value of the virtual synchronous generator under grid control when applying grid control based on a virtual synchronous generator to the converter. The power control loop references the converter's reactive current reference value and the converter's active power reference value.

[0107] In specific implementation, the methods, steps or functions performed by the units of the grid system cooperative control device for grid voltage dips provided by the present invention can refer to the grid system cooperative method for grid voltage dips provided by the present invention.

Claims

1. A collaborative control method for a grid system applicable to voltage sags, characterized in that, include: The actual grid connection point voltage amplitude collected at the grid connection point between the modular multilevel converter and the grid is compared with the grid connection point voltage reference value to obtain the voltage drop depth; the grid system includes: new energy power generation module, modular multilevel converter and grid, the modular multilevel converter is a grid-type modular multilevel converter, the DC side is connected to the new energy power generation module through DC bus, and the AC side is connected to the grid; Based on the voltage drop depth, the grid voltage deviation between the current grid voltage and the grid rated voltage is determined; based on the grid voltage deviation, the reactive current correction is obtained, and then the converter rated reactive current is superimposed to obtain the converter reactive current reference value under low-voltage ride-through. When a grid voltage drop occurs, the converter reactive current reference value and the actual grid connection point voltage amplitude are used to calculate the converter reactive power reference value under low-voltage grid ride-through. The actual DC bus voltage collected on the DC bus is compared with the DC bus voltage reference value to obtain the DC side voltage deviation. The DC side voltage deviation is input into the PI controller to obtain the DC side power correction amount. Based on the DC side power correction amount, the output power of the new energy is corrected to obtain the active power reference value of the converter. When applying grid-based control based on virtual synchronous generators to the converter, the power control loop references the converter reactive current reference value and the converter active power reference value to generate the output voltage amplitude reference value of the virtual synchronous generator under grid-based control. The system frequency deviation is obtained by comparing the actual system frequency of the converter under grid control with the rated system frequency. The active power correction of the new energy source is then obtained based on the system frequency deviation. The new energy power dispatch deviation is obtained based on the deviation between the new energy output power and the new energy reference active power, and the correction of the deviation by the new energy active power correction. The duty cycle correction of the new energy power generation module at the next moment is calculated based on the new energy power dispatch deviation, and the current duty cycle is corrected to obtain the duty cycle of the new energy power generation module at the next moment. If a grid voltage drop is identified, the duty cycle correction of the new energy power generation module at the next moment is set to 0. The sign of the duty cycle correction of the new energy power generation module at the next moment is determined based on the sign of the product between the new energy output voltage increment and the new energy output power increment.

2. The grid networking system collaborative control method for grid voltage sag according to claim 1, characterized in that, The obtained voltage drop depth includes: The ratio of the deviation between the reference value of the grid connection point voltage and the actual voltage amplitude at the grid connection point to the reference value of the grid connection point voltage is used as the voltage drop depth.

3. The grid networking system collaborative control method for grid voltage sag according to claim 2, characterized in that, When a grid voltage drop occurs, it includes: A voltage drop is considered to have occurred when the voltage drop depth exceeds the corresponding threshold.

4. The grid-connected system collaborative control method for voltage dips according to claim 2, characterized in that, Determining the grid voltage deviation between the current grid voltage and the grid rated voltage includes: The deviation between the per-unit value and the voltage drop depth when the actual grid connection point voltage is at the rated value is calculated and used as the grid voltage deviation.

5. The grid networking system collaborative control method for grid voltage sag according to claim 4, characterized in that, The reactive current correction amount obtained based on the grid voltage deviation includes: After limiting the voltage deviation of the power grid, the reactive current correction amount is obtained by linear mapping through a proportional coefficient.

6. A grid-connected system collaborative control device suitable for grid voltage dips, characterized in that, A grid system cooperative control method for grid voltage sag according to any one of claims 1 to 5.

Images