A virtual synchronous generator grid-connected inverter voltage control method and system

Abstract

This invention discloses a voltage control method and system for a virtual synchronous generator grid-connected inverter, belonging to the field of power electronic control technology. It includes: real-time acquisition of grid-connected point voltage and current, obtaining voltage amplitude feedback through coordinate transformation, and calculating the voltage control deviation by subtracting it from a reference value; approximating the voltage control system as a first-order integrator model, reducing unmodeled dynamics, parameter uncertainties, and external disturbances to a total disturbance and expanding it into new state variables; constructing a linear extended state observer to estimate the voltage state and total disturbance in real time; employing a linear state error feedback control law to generate a virtual control quantity based on the voltage control deviation, and using the disturbance estimate as feedforward compensation to generate a virtual synchronous potential amplitude command; combining the phase angle generated by the VSG power angle frequency loop to synthesize a reference voltage, which is then pulse-width modulated to drive the inverter. This invention significantly improves the voltage dynamic response performance and robustness of the inverter under complex operating conditions through real-time estimation and compensation of the total disturbance.

Description

Technical Field

[0001] This invention belongs to the field of power electronic control technology, and in particular relates to a control method and system for improving the voltage regulation performance and anti-disturbance capability of grid-connected inverters with grid-connected energy storage under complex operating conditions such as grid impact and load disturbance. Background Technology

[0002] With the transformation of the global energy structure and the rapid development of distributed generation technology, a large number of power electronics-based grid-connected inverters have been integrated into power systems. While these inverters provide clean energy to the grid, they pose serious challenges to the grid's voltage stability and reactive power support capabilities during operation, especially under complex conditions such as grid connection, load surges / disconnections, and grid faults.

[0003] Existing virtual synchronous generator (VSG) control strategies, especially in the reactive-voltage (QU) control loop responsible for voltage regulation and reactive power support within the simulated synchronous generator excitation system, typically employ fixed-parameter proportional (P) or proportional-integral (PI) controllers to adjust voltage errors. This control structure, when facing the increasingly complex and dynamic demands of modern power systems, exposes the following inherent and insurmountable flaws: First, regarding transient dynamic response, because the parameters of the PI controller are fixed, it cannot adapt to sudden and drastic changes in the system. When subjected to large-scale disturbances such as grid connection surges, sudden increases in local load, instantaneous load shedding, or short-term grid faults, the control quantity cannot compensate in a timely manner. This directly leads to huge transient overshoot and prolonged dynamic oscillations in the inverter grid connection voltage amplitude during transient processes, significantly extending the system's recovery time to steady state and seriously threatening the system's transient stability margin. Second, regarding the robustness of the control system, traditional PI control is model-dependent. It requires precise system mathematical models and parameters, but in actual operation, factors such as the nonlinearity and internal coupling of the inverter itself, parameter deviations of the LCL filter, and state changes of the energy storage battery all contribute to model uncertainties. Meanwhile, unknown changes in grid impedance and random fluctuations in external loads constitute external disturbances. Fixed PI controllers cannot effectively and proactively cancel these uncertainties and disturbances in real time, leading to a rapid deterioration in control performance when operating conditions change, exhibiting severely insufficient robustness.

[0004] Therefore, this invention discloses a voltage control method, system, and medium for a virtual synchronous generator grid-connected inverter. Summary of the Invention

[0005] In view of the shortcomings of the prior art, the purpose of this invention is to provide a high-performance, robust virtual synchronous generator grid-connected inverter voltage control method and system to solve the core dynamic performance defects of the prior art under complex grid conditions. The ultimate goal is to enable the grid-connected energy storage inverter to have grid support capabilities comparable to traditional synchronous generators.

[0006] To achieve the above objectives, the present invention adopts the following technical solution;

[0007] A voltage control method for a virtual synchronous generator grid-connected inverter includes the following steps:

[0008] The three-phase voltage and current at the grid connection point are collected in real time. After coordinate transformation and amplitude calculation, the actual voltage amplitude feedback is obtained. The difference between the actual voltage amplitude feedback and the voltage reference value is calculated to obtain the voltage control deviation.

[0009] The dynamic characteristics of the voltage control system are approximated as a first-order integrator model, and the unmodeled dynamics of the system, parameter uncertainties and external power grid disturbances are uniformly attributed to the total disturbance acting on the model.

[0010] The total disturbance is expanded into a new state variable, and a linear extended state observer is constructed. Based on the voltage control deviation and the system control gain, the estimated values ​​of the system voltage state and the total disturbance are calculated in real time.

[0011] A linear state error feedback control law is adopted to generate a virtual control quantity based on the voltage control deviation, and the estimated value of the total disturbance is used to feedforward the virtual control quantity to generate a virtual synchronous potential amplitude command.

[0012] The virtual synchronous potential amplitude command is combined with the real-time phase angle generated by the power angle frequency loop of the virtual synchronous generator to synthesize a reference voltage signal in the three-phase stationary coordinate system, and then drive pulses are generated through pulse width modulation to control the inverter.

[0013] As a further aspect of the present invention, the step of obtaining the voltage control deviation includes:

[0014] The three-phase instantaneous voltage at the grid connection point is acquired in real time using voltage and current transformers at a frequency of not less than 10kHz. The time-varying AC signal is then projected onto the grid using a second-order generalized integrator or synchronous rotating coordinate transformation. Obtain the vertical component of the voltage vector in a synchronously rotating coordinate system. and horizontal axis components ;

[0015] After coordinate transformation and amplitude calculation, the actual amplitude feedback of the current grid connection point voltage is obtained. ;

[0016] ;

[0017] Subsequently, a first-order low-pass filter (LPF) was used to... Smoothing is performed to filter out high-frequency ripple generated by inverter PWM modulation;

[0018] Voltage reference value Feedback amount of actual voltage amplitude after smoothing The difference is calculated to obtain the real-time voltage control deviation at the current moment. :

[0019] ;

[0020] Voltage control deviation, as the core driving quantity of the LADRC control law, is used to guide the observation of the total disturbance by LESO and the calculation of voltage compensation by LSEF in subsequent steps.

[0021] As a further aspect of the present invention, the step of approximating the dynamic characteristics of the voltage control system as a first-order integrator model includes:

[0022] Make the voltage amplitude at the grid connection point Quickly and stably track reference values Due to the effect of the inner loop current control, the dynamic characteristics of the entire voltage control system can be approximated as a first-order integrator model containing the total disturbance. The system output state is selected. Voltage amplitude The final control output of LADRC is denoted as .

[0023] The dynamic equation of the system can be approximated as follows:

[0024] ;

[0025] In the formula, The control gain of the system is a parameter that needs to be estimated in advance. Its physical meaning is the final control output of the LADRC. For the rate of change of system state The intensity of the effect; This is the total disturbance of the system.

[0026] As a further aspect of the present invention, the step of uniformly attributing the unmodeled dynamics of the system, parameter uncertainties, and external power grid disturbances to the total disturbance acting on the first-order integrator model includes:

[0027] All unmodeled dynamics, model uncertainties, and external disturbances in the system are uniformly attributed to the total disturbance.

[0028] State and the total disturbance Expand into a new state variable ,Right now Therefore, the first-order system with disturbance is extended to a second-order system, and its extended state-space equation can be written as follows:

[0029] ;

[0030] In the formula, Represents the rate of change of the total disturbance; the LADRC controller estimates this precisely. and ,Will Compensation will be provided to eliminate Impact on the system.

[0031] As a further aspect of the present invention, the step of calculating the estimated value of the system voltage state and the estimated value of the total disturbance in real time includes:

[0032] For the aforementioned extended state-space equations, a linear extended state observer is constructed, whose state estimation equation is:

[0033] ;

[0034] In the formula, and These are the observation system states. and total disturbance of the observation system The estimated value; It is observation error; For state observation gain and For perturbation observation gain, and All are observer gains;

[0035] LESO parameter configuration using the bandwidth method: setting the observer bandwidth to... By placing the two poles of LESO at the same location on the negative real axis, and comparing the coefficients of the characteristic equation, the observer gain... and can be Sure:

[0036] ;

[0037] Wherein, the observer bandwidth Greater than the controller bandwidth This is to ensure that the observer can capture and estimate the dynamic changes and disturbances of the system more quickly.

[0038] As a further aspect of the present invention, the step of employing a linear state error feedback control law to generate a virtual control quantity based on the voltage control deviation includes:

[0039] LSEF uses a linear proportional control law for state error feedback, based on the system voltage error. and controller gain Calculate the virtual control quantity for uncompensated disturbances. :

[0040] ;

[0041] Controller gain The bandwidth method is also used for tuning; the controller bandwidth is set to... The poles of the closed-loop system after disturbance compensation are configured at Location. Controller gain. can be Sure:

[0042] ;

[0043] As a further aspect of the present invention, the controller bandwidth... Set as observer bandwidth 1 / 3 to 1 / 5.

[0044] As a further aspect of the present invention, the step of using the estimated value of the total disturbance to perform feedforward compensation on the virtual control quantity to generate a virtual synchronization potential amplitude command includes:

[0045] Total perturbation observed using LESO Feedforward compensation is applied to the system to eliminate the impact of disturbances on the system. Subtract compensation items Then divide by the system gain. The final control output of LADRC can then be obtained. :

[0046] ;

[0047] Output the final control value of the calculated LADRC. The result is added to the reactive power input and fed into the proportional-integral (PI) circuit to obtain the final control output. The command directly assigns the amplitude of the internal controlled electromotive force of the virtual synchronous generator. ,Right now:

[0048] ;

[0049] This instruction represents the ideal voltage amplitude that should be generated at the midpoint of the internal bridge arm of the inverter in order to maintain stable output voltage.

[0050] After generating the virtual synchronization potential amplitude command, a limiting process is also included:

[0051] ;

[0052] In the formula, This is the command for the amplitude of the virtual synchronous potential after limiting. and These are the upper and lower limits of the system's maximum permissible operating voltage, respectively.

[0053] As a further aspect of the present invention, the step of generating a drive pulse through pulse width modulation to control the inverter includes:

[0054] The real-time phase angle generated by the combined VSG power angle frequency loop Synthesize the reference voltage signal in the three-phase stationary coordinate system:

[0055] ;

[0056] , , The reference voltage signal for the three-phase stationary coordinate system; Three-phase symmetrical sinusoidal fundamental wave;

[0057] This signal directly contains the real-time voltage amplitude correction information of LADRC and the support characteristics of VSG for system frequency.

[0058] The synthesized three-phase reference voltage signal is sent to the space vector pulse width modulation module or the sinusoidal pulse width modulation module. The modulation module generates drive pulses for six power switching transistors based on the reference signal, which drive the main circuit actuator of the inverter to operate. By adjusting the instantaneous duty cycle of the inverter output voltage, the real-time correction and disturbance cancellation of the grid connection point voltage deviation are finally achieved, so that the output voltage always closely follows the reference reference.

[0059] Another aspect of this application provides a voltage control system for a virtual synchronous generator grid-connected inverter, the system comprising: the system comprising:

[0060] The main circuit module includes a three-phase inverter, an LC filter, and a common coupling point connected in sequence.

[0061] The signal acquisition module includes a voltage transformer and a current transformer, which are used to acquire the three-phase voltage and current at the common coupling point in real time at a frequency of not less than 10kHz, and calculate the actual amplitude feedback of the voltage.

[0062] The reactive-voltage control loop of the linear active disturbance rejection control includes:

[0063] The voltage error calculation unit is used to receive the voltage reference value and the actual voltage amplitude feedback, and to calculate the real-time voltage control deviation.

[0064] Linear extended state observer, based on system control gain and observer bandwidth Constructed for real-time estimation of system voltage state and the total disturbance acting on the system ;

[0065] Linear state error feedback control law module, based on controller bandwidth Based on the estimation results from the linearly extended state observer, a virtual synchrotron potential amplitude command is generated. ;

[0066] Limiting protection unit, used to limit the voltage based on DC side voltage utilization. Perform saturation limiting processing to generate a virtual synchronous potential amplitude command after limiting. ;

[0067] The reference voltage synthesis module is used to receive the amplitude command of the limited virtual synchro potential. and real-time phase angle Synthesize a sinusoidal reference voltage signal in a three-phase stationary coordinate system;

[0068] The pulse width modulation drive module is used to generate drive pulses from the three-phase reference voltage signal to control the on / off state of the power switching transistors in the three-phase inverter.

[0069] The parameter tuning configuration module is used to preset and store the observer bandwidth, controller bandwidth and system control gain of the LADRC controller according to the system response speed and disturbance rejection capability requirements, and automatically calculate the LESO gain and LSEF gain according to the bandwidth method relationship.

[0070] In summary, due to the adoption of the above technical solution, the beneficial technical effects of the invention are as follows:

[0071] Because LADRC can provide real-time and accurate feedforward compensation for total disturbances caused by grid connection surges and load surges, the strategy of this invention can effectively suppress transient voltage spikes. Simulation results show that under various strong disturbance conditions such as grid connection, load surges, and load shedding, this strategy can reduce the maximum voltage overshoot on average, greatly improving the grid-connected inverter's ability to withstand voltage surges and the system's transient voltage stability.

[0072] LADRC uses the bandwidth method to configure the dynamic performance of the closed-loop system to a value determined by... The ideal state determined not only accelerates the voltage dynamic response speed, but more importantly, its real-time elimination of total disturbances effectively suppresses the spikes and violent oscillations of shaft current, active power, and reactive power during transient processes, making the transition process smoother and faster, and significantly improving the system damping characteristics.

[0073] The LADRC strategy proposed in this invention operates in a model-independent manner. It estimates and compensates for the total disturbance through an observer, thereby decoupling the control design from the complex and accurate system model. Therefore, its control performance exhibits strong adaptability and anti-interference capability to fluctuations in system parameters (such as LCL filter parameters), changes in grid impedance, and other nonlinear disturbances. It is particularly suitable for grid-connected applications in weak grids with uncertain grid parameters and complex structures, thus overcoming the inherent deficiency of insufficient robustness in traditional VSG.

[0074] LADRC parameter tuning only requires two physically meaningful bandwidth parameters. and This greatly simplifies the complex parameter tuning process. Designers only need to configure according to the desired response speed. Configure according to the desired anti-interference capability ( This allows for the rapid development of high-performance controllers, with design efficiency far exceeding that of traditional PI controllers and complex nonlinear controllers.

[0075] Once the dynamic stability and smoothness of the system are guaranteed, the active and reactive power output of the inverter no longer fluctuate drastically, ultimately ensuring that the three-phase current waveform is more stable and pure under both transient and steady-state conditions, significantly improving the output power quality of the grid-connected inverter. Attached Figure Description

[0076] Figure 1 A schematic diagram of a VSG control flow based on LADRC is provided for an embodiment of the present invention;

[0077] Figure 2 This is a schematic diagram of the VSG system control topology provided in an embodiment of the present invention;

[0078] Figure 3 This is a schematic diagram of a typical reactive voltage loop control topology provided in an embodiment of the present invention;

[0079] Figure 4 A schematic diagram of a reactive voltage loop control topology based on an LADRC controller provided in an embodiment of the present invention;

[0080] Figure 5 This is a schematic diagram of first-order LADRC control provided in an embodiment of the present invention;

[0081] Figure 6 This is a schematic diagram of the dynamic response curve of voltage amplitude under grid-connected and load change conditions provided in an embodiment of the present invention.

[0082] Figure 7 This is a schematic diagram of the q-axis current waveform under sudden load change conditions provided in an embodiment of the present invention;

[0083] Figure 8 This is a schematic diagram of the active power waveform under sudden load changes provided in an embodiment of the present invention.

[0084] Figure 9 This is a schematic diagram of reactive power waveform under load change conditions provided in an embodiment of the present invention. Detailed Implementation

[0085] To make the objectives, technical solutions, and advantages of the invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0086] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0087] The logical framework of this invention is as follows: Figure 1 As shown, after determining the control framework, the core step of this invention lies in how to achieve real-time capture of internal coupling and external shocks of the system through a linear extended state observer (LESO).

[0088] The specific steps involved in its implementation include:

[0089] S100. System Status Acquisition and Voltage Error Calculation:

[0090] In existing technologies, the voltage control of VSGs often relies directly on local measurements. However, in weakly fluctuating power grids with a high proportion of renewable energy integration, the sampled signals often contain high-frequency harmonics and random fluctuations. If directly applied to a PI regulator, this can easily lead to frequent controller overshoot. This invention provides a precise input reference for the subsequent active disturbance rejection of LADRC through high-frequency synchronous acquisition and error preprocessing.

[0091] To solve the above problem, step S100 specifically adopts the following steps:

[0092] S110. The control method of the present invention is applied to a grid-connected inverter with grid-connected energy storage, and its main circuit structure is as follows: Figure 2 As shown. The system mainly includes a three-phase inverter, an LC filter, and a common coupling point (PCC). Voltage transformers (PTs) and current transformers (CTs) are used to acquire the three-phase instantaneous voltage at the grid connection point (PCC) in real time at a frequency of not less than 10kHz. The time-varying AC signal is projected onto the grid using a second-order generalized integrator (SOGI) or synchronous rotating coordinate transformation (Park transform). In a synchronously rotating coordinate system, the vertical component representing the voltage vector is obtained. and horizontal axis components .

[0093] S120. After coordinate transformation and amplitude calculation, the actual amplitude feedback of the current grid connection point voltage is obtained. :

[0094] ;

[0095] Subsequently, a first-order low-pass filter (LPF) with a cutoff frequency slightly higher than the system control bandwidth is used to... Smoothing is performed to filter out high-frequency ripple generated by inverter PWM modulation, ensuring the continuity of feedback status.

[0096] S130. Will and Difference calculation to determine real-time voltage deviation As the driving source of the control system, the voltage control deviation at the current moment is obtained:

[0097] ;

[0098] This deviation will serve as the core driving variable for the LADRC controller, guiding the subsequent steps of LESO's observation of the total disturbance intensity and LSEF's calculation of the voltage compensation.

[0099] S200. Establish a model of the controlled object with disturbances:

[0100] Traditional PI control treats the controlled object as a black box or a simplified linear model, neglecting the nonlinear characteristics caused by the hysteresis of the inner loop current response and changes in line impedance. This invention establishes a disturbance-inclusive model, making all system uncertainties explicit and providing a mathematical basis for disturbance compensation.

[0101] To solve the above technical problem, step S200 specifically includes the following steps:

[0102] S210. Adjust the voltage amplitude at the grid connection point. Quickly and stably track reference values Due to the effect of the inner loop current control, the dynamic characteristics of the entire voltage control system can be approximated as a first-order integrator model containing the total disturbance. The system output state is selected. Voltage amplitude The final control output of the LADRC controller is denoted as The dynamic equations of the system can be approximated as follows:

[0103] ;

[0104] In the formula, The control gain of the system is a parameter that needs to be estimated in advance. Its physical meaning is the final control output of the LADRC. For the rate of change of system state The intensity of the effect; The total disturbance of the system is the sum of all unmodeled dynamics, model uncertainties and external disturbances (such as power grid fluctuations, load changes, etc.) in the system.

[0105] S220. Based on the core concept of the LADRC controller, the state... and the total disturbance Expand into a new state variable ,Right now Therefore, the first-order system with disturbance is extended to a second-order system, and its extended state-space equation can be written as follows:

[0106] ;

[0107] In the formula, Represents the rate of change of the total disturbance. LADRC estimates this rate of change through precise estimation. and ,Will Compensation will be provided to eliminate Impact on the system.

[0108] S300. Real-time estimation of total disturbance:

[0109] Existing technologies are reactive in handling disturbances, requiring the PI (Programmable Logic Controller) to activate only after an error occurs. This invention expands the invisible total disturbance into a new state variable using LESO (Learning on Disturbance Optimization), enabling proactive disturbance sensing and addressing the problem of regulation lag at its root.

[0110] To solve the problems mentioned above, S300 adopts the following steps:

[0111] S310. The Linear Extended State Observer (LESO) is the core of LADRC, used to estimate the system state accurately in real time. Total disturbance The design of a first-order LADRC is as follows: Figure 5 As shown, for the above first-order extended system, its state estimation equation is as follows:

[0112] ;

[0113] In the formula, and These are the observation system states. and total disturbance of the observation system The estimated value; It is observation error; For state observation gain and For perturbation observation gain, and All are observer gains.

[0114] S320. To simplify parameter tuning and improve engineering practicality, this invention employs the bandwidth method for LESO parameter configuration. The observer bandwidth is set to... By placing the two poles of LESO at the same position on the negative real axis, that is... By comparing the coefficients of the characteristic equation, the state observation gain... and perturbation observation gain All by Sure:

[0115] ;

[0116] Observer bandwidth The selection of the appropriate value is crucial to ensuring LESO performance, as it determines the speed at which the observer tracks the state and disturbances. Typically, it requires... Greater than the controller bandwidth (Right now This ensures that the observer can capture and estimate the dynamic changes and disturbances of the system more quickly.

[0117] S400. Disturbance feedforward compensation control:

[0118] Traditional control methods struggle to balance response speed and stability, and excessive proportional gain can cause oscillations. This invention utilizes LSEF to dynamically feedforward compensate for estimated disturbances, transforming the nonlinear object into an easily controllable integral element through active cancellation, thus achieving overshoot-free regulation.

[0119] To solve the problems mentioned above, S400 adopts the following steps:

[0120] S410.LSEF is responsible for basing its estimates on LESO results. and Generate control quantity LSEF employs a linear proportional control law for state error feedback and integrates disturbance compensation terms. First, the virtual control quantity for uncompensated disturbances is calculated. This quantity is based on the system voltage error. and controller gain :

[0121] ;

[0122] S420. Total perturbation observed using LESO Feedforward compensation is applied to the system to eliminate the impact of disturbances. Subtract compensation items Then divide by the system gain. This will give you the final control output of LADRC:

[0123] ;

[0124] S430. Controller Gain The bandwidth method is also used for tuning; the controller bandwidth is set to... The poles of the closed-loop system after disturbance compensation are configured at Location. Controller gain. can be Sure:

[0125] ;

[0126] S440. By adjusting the controller bandwidth ( Typically set to the observer bandwidth (1 / 3 to 1 / 5), ensuring sufficient stability margin in the control loop. Because the compensated object possesses integral characteristics, LSEF eliminates the need for an integral stage to ensure zero steady-state error tracking of the system in steady state, significantly improving the inverter's voltage steady-state accuracy and transient recovery speed under harsh conditions such as the Northwest Power Grid.

[0127] S450. Output the calculated final control value of LADRC. The result is added to the reactive power input and fed into the proportional-integral (PI) circuit to obtain the final control output. The command directly assigns the amplitude of the internal controlled electromotive force of the virtual synchronous generator. ,Right now:

[0128] ;

[0129] This instruction represents the ideal voltage amplitude that should be generated at the midpoint of the internal bridge arm of the inverter in order to maintain stable output voltage.

[0130] S460. Considering the physical limitations of the power converter (such as DC-side bus voltage utilization), the generated amplitude command... Amplitude limiting is applied:

[0131] ;

[0132] In the formula, This is the command for the amplitude of the virtual synchronous potential after limiting. and These represent the upper and lower limits of the system's maximum permissible operating voltage. This step ensures that, under extreme disturbance conditions, the compensation amount provided by the algorithm will not cause the PWM modulation to enter the overmodulation region.

[0133] S500. Reference Voltage Synthesis and Modulation:

[0134] Traditional VSGs exhibit severe output potential fluctuations under disturbances, which can easily lead to overcurrent in power electronic devices. The commands generated by this invention are purified by LADRC, which not only maintains voltage rigidity but also effectively suppresses high-frequency noise caused by disturbance compensation, thus improving the smoothness of power output.

[0135] The specific steps include:

[0136] S510. Real-time phase angle generated by combining the VSG power angle frequency loop (calculated from the active power-frequency droop characteristic and rotor motion equation). Synthesize the reference voltage signal in the three-phase stationary coordinate system:

[0137] ;

[0138] , , The reference voltage signal for the three-phase stationary coordinate system; It is a three-phase symmetrical sinusoidal fundamental wave; this signal directly contains the real-time correction information of the voltage amplitude by LADRC and the support characteristics of the system frequency by VSG.

[0139] S520. The synthesized three-phase reference voltage signal is sent to the Space Vector Pulse Width Modulation (SVPWM) or Sinusoidal Pulse Width Modulation (SPWM) module. The modulation module generates drive pulses for six power switches based on the reference signal, driving the inverter main circuit actuators to operate. By adjusting the instantaneous duty cycle of the inverter output voltage, real-time correction and disturbance cancellation of the grid connection point voltage deviation are ultimately achieved, ensuring that the output voltage always closely follows the reference.

[0140] Parameter tuning:

[0141] The performance and stability of the LADRC controller of this invention are determined by its key parameters. Controller bandwidth and observer bandwidth The decision is made jointly. Proper parameter tuning is crucial to ensuring the superior performance of LADRC. This embodiment primarily employs the bandwidth method for LADRC parameter tuning, which combines controller parameters with two core design parameters. and This correlation greatly simplifies the complex parameter debugging process.

[0142] System gain The estimate, It is the input gain coefficient of the controlled object (simplified QV-loop model), representing the control quantity. The effect on the rate of change of system state. In practical systems, Preliminary estimates can be made through experimental testing or based on system theoretical models. Due to the rapid response of the inverter's inner-loop current and voltage controllers, it can be approximated as... Includes the inner loop gain and the inverter DC-side voltage. The impact. Accurate. It is estimated that this will improve the observation accuracy of LESO and the compensation effect of LSEF.

[0143] Controller bandwidth The determination of controller bandwidth. The controller proportional gain determines the response speed of the LADRC closed-loop system. Directly by Decide: . The larger the value, the farther the closed-loop poles of the system are from the imaginary axis, resulting in a faster system response and shorter settling time. However, an excessively large value... This increases the controller's sensitivity to high-frequency noise and measurement errors, and may also reduce the system's phase margin, increase the risk of oscillation, and place higher demands on hardware computing power. Therefore, The choice requires a trade-off between speed and robustness. In this embodiment, considering the system sampling frequency and dynamic response requirements, the preferred method is... .

[0144] Observer bandwidth The determination of the observer bandwidth. The LESO determines the system state and total disturbance. The estimation speed and accuracy. Observer gain includes state observation gain. and perturbation observation gain , respectively by Sure: To ensure that LESO can track and estimate disturbances in a timely and accurate manner, thereby enabling LSEF to achieve effective disturbance compensation, Must be greater than This embodiment has been verified through multiple simulations and experiments, and the following settings have been established. and The ratio coefficient between : Experience shows that when The range of values ​​is At this point, the system performance reaches its optimal balance. This embodiment preferably... ,Right now According to the selected ,but Determined as .

[0145] Gain parameter calculation. Based on the bandwidth parameters determined above, calculate the LADRC gain:

[0146] ;

[0147] ;

[0148] ;

[0149] By tuning using the bandwidth method, the LADRC controller achieves a physically meaningful... and With these two parameters, a reasonable configuration of system poles was achieved, ensuring the speed and disturbance rejection robustness of the voltage loop.

[0150] Example 2:

[0151] A voltage control system for a virtual synchronous generator grid-connected inverter, the system being used to execute the voltage control method described in Example 1. The system includes:

[0152] The main circuit module includes a three-phase inverter, an LC filter (or LCL filter), and a common coupling point (PCC) connected in sequence. This module is used to convert DC-side electrical energy into AC-side electrical energy and feed it into the grid or to power local loads.

[0153] The signal acquisition module includes:

[0154] Voltage transformers (PT) and current transformers (CT) are used to acquire the three-phase instantaneous voltage and three-phase instantaneous current at the common coupling point (PCC) in real time at a frequency of not less than 10kHz.

[0155] The coordinate transformation and calculation unit is used to project AC signals onto a coordinate plane using a second-order generalized integrator (SOGI) or a synchronous rotating coordinate transformation (Park transform). By synchronously rotating the coordinate system, the vertical component of the voltage can be obtained. Horizontal axis components And calculate the actual amplitude feedback of the current grid connection point voltage. .

[0156] Low-pass filter (LPF) is used to provide voltage amplitude feedback. Smoothing is performed to filter out high-frequency ripples generated by PWM modulation.

[0157] The virtual synchronous generator (VSG) power angle-frequency control loop is used to simulate the rotor motion equations and active power-frequency droop characteristics of a synchronous generator, and calculates the real-time phase angle required by the system based on the active power deviation. This provides frequency support for the system.

[0158] A reactive-voltage control loop based on Linear Active Disturbance Rejection Control (LADRC) is provided, comprising:

[0159] Voltage error calculation unit, used to receive voltage reference value and the actual amplitude feedback of voltage Calculate real-time voltage control deviation ;

[0160] Linear Extended State Observer (LESO), based on system control gain and observer bandwidth Constructed for real-time estimation of system voltage state and the total disturbance acting on the system ;

[0161] Linear State Error Feedback (LSEF) control law module, based on controller bandwidth The LESO estimation results, through proportional control and disturbance feedforward compensation, generate the final output voltage acting on the inverter. (i.e., the internal controlled electromotive force amplitude command of the virtual synchronous generator) ); Limiting protection unit, used to limit the voltage based on the DC side voltage utilization rate. Perform saturation limiting processing to generate a virtual synchronous potential amplitude command after limiting. .

[0162] The reference voltage synthesis module is used to receive the amplitude command of the limited virtual synchronous potential. and the real-time phase angle Synthesize a sinusoidal reference voltage signal in a three-phase stationary coordinate system. , , .

[0163] The pulse width modulation (PWM) drive module is used to generate six drive pulse signals from the three-phase reference voltage signal through space vector pulse width modulation (SVPWM) or sinusoidal pulse width modulation (SPWM) algorithms, and control the on and off of the power switching transistors in the three-phase inverter.

[0164] The parameter tuning configuration module is used to preset and store the observer bandwidth of the LADRC controller according to the system response speed and disturbance rejection requirements. Controller bandwidth and system control gain It automatically calculates LESO gain and LSEF gain based on the bandwidth relationship. ).

[0165] Example 3:

[0166] To verify the superiority of the LADRC-based VSG reactive-voltage control strategy proposed in this invention, a complete grid-connected inverter system model was established on the MATLAB / Simulink simulation platform.

[0167] As a comparison, a typical reactive voltage loop is as follows: Figure 3 As shown, the reactive power channel corresponds to the excitation regulation stage of the synchronous machine, mainly responsible for stabilizing the voltage amplitude and regulating reactive power. The controller first compares the reactive power reference value with the actual reactive power, and simultaneously compares the voltage reference value with the grid connection point voltage to obtain the reactive power error and voltage error. The voltage error is amplified by the regulating gain and then superimposed with the reactive power error to form a comprehensive regulation signal. The two are added together and then passed through a proportional-integral stage to obtain the final control output. This regulation signal serves as the voltage reference in the inner loop control, and then regulates the reactive power through the inverter to support the grid voltage. The mathematical model is as follows:

[0168] ;

[0169] In the formula, Voltage amplitude, This is a reference value for reactive power. For real-time reactive power feedback, This is the voltage reference value. This is the measured value of the voltage at point PCC. This is the reactive power droop factor. The integral coefficient is... It is a complex frequency variable.

[0170] Considering the shortcomings of traditional reactive voltage loops in lacking disturbance observation and compensation, this invention introduces a LADRC controller into the reactive voltage loop to improve the ability to suppress model uncertainties and disturbances. This method utilizes reactive error... The reactive power response is controlled, but the voltage error is no longer adjusted proportionally; instead, it is fed into the LADRC controller for more precise and robust voltage deviation compensation, effectively enhancing the control performance. The control topology is as follows: Figure 4 As shown, the final output voltage model is:

[0171] ;

[0172] In the formula, For LADRC controller output, is the integral coefficient.

[0173] By setting typical complex operating conditions and comparing them with the traditional VSG control strategy, the main parameters of the simulation system are shown in the table below.

[0174] Table 1. Main parameters of the simulation system:

[0175] parameter numerical values Remark DC side voltage 800 Inverter supply voltage Grid voltage 380 RMS value of grid line voltage Rated frequency 50 Filter inductor 2.5 Filter capacitor 40 Moment of inertia 0.05 VSGP-f ring parameters Damping coefficient 10 VSGP-f ring parameters LADRC controller bandwidth 100 LADRC Key Parameters LADRC observer bandwidth 460 LADRC Key Parameters

[0176] The simulation conditions are set as follows:

[0177] Operating Condition I (Grid Connection Impact): Inverter in It is put into operation in the power grid.

[0178] Operating Condition II (Sudden Load Increase): In At that time, a resistive load with 50% of the rated power was suddenly added to point PCC.

[0179] Operating Condition III (Load Removal): During the time... The sudden load was removed.

[0180] Figure 6 The inverter output voltage amplitudes under different operating conditions are shown for traditional VSG control and LADRC+VSG strategy. The dynamic response curve.

[0181] Operating Condition I (Grid Connection Impact): In At the moment the inverter is switched on and connected to the grid, the system is subjected to current and voltage surges. Under traditional VSG control, the maximum voltage overshoot reaches 12.77%; however, after adopting the LADRC+VSG strategy of this invention, the maximum overshoot is significantly reduced to 7.94%, an improvement of approximately 37.8%. This indicates that LADRC can effectively suppress voltage surges by estimating and compensating for grid disturbances at the moment of grid connection.

[0182] Operating Condition II (Sudden Load Increase): In When the load suddenly increases, the voltage drop under traditional VSG control is large and the recovery time is long; the voltage drop under the LADRC+VSG strategy is smaller, the recovery speed is faster, and the maximum overshoot is reduced by 42.19%.

[0183] Operating Condition III (Load Removal): In When the load is removed, the LADRC+VSG strategy has a stronger effect on suppressing the instantaneous voltage rise, and the maximum overshoot is reduced by 43.35%.

[0184] The results show that the LADRC+VSG control strategy proposed in this invention has significantly better voltage dynamic performance than traditional PI control under all transient conditions, can greatly reduce voltage overshoot / undershoot, and has a smoother dynamic response and faster recovery.

[0185] To further evaluate the improvement of LADRC on the internal dynamics of the system, Axis current, active power and reactive power The responses were compared and analyzed.

[0186] Axis current response: Shaft current affects voltage amplitude Key inner-loop variables. For example... Figure 7 As shown, during a sudden load change, the traditional VSG control... The shaft current exhibited significant oscillations. Under the LADRC+VSG strategy... The amplitude of shaft current fluctuations was significantly reduced, with the maximum overshoot decreasing by 55.99% and 53.1% during load surges and removals, respectively. This indicates that LADRC's accurate estimation of the total disturbance enables the outer voltage loop to generate more precise control commands, thereby improving the stability of the inner loop current.

[0187] Active power Response: Although LADRC primarily operates on the QV ring, its performance also affects the Pf ring due to the system's strong coupling. For example... Figure 8 As shown, the active power of a traditional VSG at the instant of disturbance A distinct spike appears. (LADRC+VSG) The change curves are smoother, and the spikes are effectively suppressed. The maximum overshoot is reduced by 47.6% and 50.4%, respectively, indicating that LADRC helps to suppress transient fluctuations in active power caused by system coupling.

[0188] reactive power Response: Reactive Power The response directly reflects the regulating capability of the QV loop. For example... Figure 9 As shown, the reactive power of a traditional VSG during load abrupt changes... Significant spikes and large steady-state shifts appear. LADRC+VSG can rapidly suppress these. Peak loads were reduced, and the maximum overshoot during load surges and removals was reduced by 66.1% and 57.4%, respectively, significantly improving reactive power fluctuations and steady-state deviations.

[0189] The sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0190] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A voltage control method for a virtual synchronous generator grid-connected inverter, characterized in that, Includes the following steps: The three-phase voltage and current at the grid connection point are collected in real time. After coordinate transformation and amplitude calculation, the actual voltage amplitude feedback is obtained. The difference between the actual voltage amplitude feedback and the voltage reference value is calculated to obtain the voltage control deviation. The dynamic characteristics of the voltage control system are approximated by a first-order integrator model, and the unmodeled dynamics of the system, parameter uncertainties and external power grid disturbances are uniformly attributed to the total disturbance acting on the first-order integrator model. The total disturbance is expanded into a new state variable, and a linear extended state observer is constructed. Based on the voltage control deviation and the system control gain, the estimated values ​​of the system voltage state and the total disturbance are calculated in real time. A linear state error feedback control law is adopted to generate a virtual control quantity based on the voltage control deviation, and the estimated value of the total disturbance is used to feedforward the virtual control quantity to generate a virtual synchronous potential amplitude command. The virtual synchronous potential amplitude command is combined with the real-time phase angle generated by the power angle frequency loop of the virtual synchronous generator to synthesize a reference voltage signal in the three-phase stationary coordinate system, and then drive pulses are generated through pulse width modulation to control the inverter.

2. The voltage control method for a virtual synchronous generator grid-connected inverter according to claim 1, characterized in that, The step of obtaining the voltage control deviation includes: The three-phase instantaneous voltage at the grid connection point is acquired in real time using voltage transformers and current transformers at a frequency of not less than 10kHz. The AC signal is projected onto the synchronous rotating coordinate system through a second-order generalized integrator or synchronous rotating coordinate transformation to obtain the vertical and horizontal components of the voltage vector. Calculate the actual amplitude feedback of the current grid connection point voltage based on the vertical axis component and the horizontal axis component; A first-order low-pass filter is used to smooth the actual amplitude feedback to filter out the high-frequency ripple generated by the inverter pulse width modulation. The real-time voltage control deviation at the current moment is obtained by subtracting the voltage reference value from the smoothed actual amplitude feedback value.

3. The voltage control method for a virtual synchronous generator grid-connected inverter according to claim 1, characterized in that, The steps to approximate the dynamic characteristics of a voltage control system as a first-order integrator model include: The system output state is selected as the voltage amplitude at the grid connection point. The final control output of the linear active disturbance rejection controller is denoted as the control quantity. The system dynamic equation is approximately expressed as: the rate of change of the voltage amplitude at the grid connection point is equal to the system control gain multiplied by the control quantity plus the total disturbance. Among them, the system control gain is a pre-estimated parameter that characterizes the strength of the effect of the control quantity on the rate of change of the voltage amplitude at the grid connection point.

4. The voltage control method for a virtual synchronous generator grid-connected inverter according to claim 1, characterized in that, The steps to unify the unmodeled dynamics of the system, parameter uncertainties, and external power grid disturbances into a total disturbance acting on the first-order integrator model include: All unmodeled dynamics, model uncertainties, and external disturbances in the system are uniformly attributed to the total disturbance. The voltage amplitude at the grid connection point is taken as the first state variable, and the total disturbance is expanded into the second state variable, thereby expanding the first-order system with disturbance into a second-order system, forming the extended state space equation; The extended state-space equation describes the rate of change of the first state variable as being composed of the second state variable and the system control gain multiplied by the control quantity. The rate of change of the second state variable is characterized by the rate of change of the total disturbance, and the system output is the first state variable.

5. The voltage control method for a virtual synchronous generator grid-connected inverter according to claim 1, characterized in that, The steps for calculating the estimated values ​​of the system voltage state and the total disturbance in real time include: A linear extended state observer is constructed for the extended state-space equation. Its state estimation equation includes: the rate of change of the first estimated state is composed of the second estimated state plus the system control gain multiplied by the control quantity plus the state observation gain multiplied by the observation error; the rate of change of the second estimated state is composed of the disturbance observation gain multiplied by the observation error. The observation error is the difference between the actual output of the system and the first estimated state. The observer parameters are configured using the bandwidth method: the observer bandwidth is set, and the two poles of the linearly extended state observer are placed at the same position on the negative real axis, so that the state observation gain is equal to twice the observer bandwidth and the disturbance observation gain is equal to the square of the observer bandwidth. The observer bandwidth is greater than the controller bandwidth to ensure that the observer can estimate the dynamic changes and disturbances of the system more quickly.

6. The voltage control method for a virtual synchronous generator grid-connected inverter according to claim 1, characterized in that, The method of using a linear state error feedback control law to generate virtual control quantities based on voltage control deviation includes: A linear proportional control law is used for state error feedback, and the virtual control quantity for uncompensated disturbances is calculated based on the system voltage error and controller gain. The controller gain is tuned using the bandwidth method: the controller bandwidth is set, and the poles of the closed-loop system after disturbance compensation are configured at the controller bandwidth on the negative real axis, so that the controller gain is equal to the controller bandwidth.

7. The voltage control method for a virtual synchronous generator grid-connected inverter according to claim 6, characterized in that, The controller bandwidth is set to 1 / 3 to 1 / 5 of the observer bandwidth.

8. The voltage control method for a virtual synchronous generator grid-connected inverter according to claim 1, characterized in that, The step of using the estimated total disturbance to feedforward compensate the virtual control quantity to generate a virtual synchronous potential amplitude command includes: The system is fed forward to compensate for the total disturbance estimate observed by the linear extended state observer. The virtual control quantity with uncompensated disturbance is subtracted from the total disturbance estimate and then divided by the system control gain to obtain the final control quantity output of the linear active disturbance rejection controller. The final control output is added to the reactive power channel and then fed into the proportional-integral stage to obtain the final control output, which is then directly assigned to the internal controlled electromotive force amplitude command of the virtual synchronous generator. After generating the virtual synchronization potential amplitude command, a limiting process is also included: the virtual synchronization potential amplitude command is saturated and limited according to the DC side voltage utilization rate to generate a limited virtual synchronization potential amplitude command.

9. The voltage control method for a virtual synchronous generator grid-connected inverter according to claim 8, characterized in that, The step of generating drive pulses to control the inverter via pulse width modulation includes: By combining the real-time phase angle generated by the power angle frequency loop of the virtual synchronous generator, a reference voltage signal in the three-phase stationary coordinate system is synthesized. The reference voltage signal contains the real-time correction information of the voltage amplitude by the linear active disturbance rejection controller and the support characteristics of the virtual synchronous generator for the system frequency. The synthesized three-phase reference voltage signal is sent to the space vector pulse width modulation module or the sinusoidal pulse width modulation module to generate drive pulses for the power switching transistors, thereby driving the main circuit actuator of the inverter to achieve real-time correction and disturbance cancellation of the grid connection point voltage deviation.

10. A voltage control system for a virtual synchronous generator grid-connected inverter, characterized in that, The system for performing the method as described in any one of claims 1 to 9 includes: The main circuit module includes a three-phase inverter, a filter, and a common coupling point connected in sequence. The signal acquisition module, including voltage transformers and current transformers, is used to acquire the three-phase voltage and current at the common coupling point in real time at a frequency of not less than 10kHz, and calculate the actual amplitude feedback of the voltage. The reactive power-voltage control loop of linear active disturbance rejection control includes: The voltage error calculation unit is used to receive the voltage reference value and the actual voltage amplitude feedback, and to calculate the real-time voltage control deviation. A linear extended state observer, built based on the system control gain and observer bandwidth, is used to estimate the system voltage state and the total disturbances acting on the system in real time. The linear state error feedback control law module generates virtual synchronous potential amplitude commands based on the controller bandwidth and the estimation results of the linear extended state observer. The limiting protection unit is used to perform saturation limiting processing on the virtual synchronous potential amplitude command according to the DC side voltage utilization rate, and generate the limited virtual synchronous potential amplitude command. The reference voltage synthesis module is used to receive the amplitude command of the limited virtual synchronous potential and the real-time phase angle, and synthesize a sinusoidal reference voltage signal in the three-phase stationary coordinate system. The pulse width modulation drive module is used to generate drive pulses from the three-phase reference voltage signal to control the on / off state of the power switching transistors in the three-phase inverter. The parameter tuning configuration module is used to preset and store the observer bandwidth, controller bandwidth and system control gain of the linear active disturbance rejection controller according to the system response speed and disturbance rejection capability requirements, and automatically calculate the linear extended state observer gain and linear state error feedback control law gain according to the bandwidth method relationship.

Images