A grid-connected inverter control method based on second-order linear active disturbance rejection
The grid-connected inverter method using second-order linear active disturbance rejection control solves the stability problem of grid-connected inverters with frequency deviation in the existing technology, realizes voltage and frequency synchronization, improves the adaptability and robustness of the system, and ensures the stability and power quality of the grid connection process.
Patent Information
- Application Number
- CN202511743168.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-25
- Publication Date
- 2026-02-24
- Estimated Expiration
- 2045-11-25
AI Technical Summary
Existing grid-connected inverter control strategies cannot fully recover to the rated value when there is a frequency deviation, resulting in current surges, frequency overshoots, and system oscillations. They are particularly inadequate in low-inertia small microgrids and multi-inverter systems, and their response speed is slow and their parameters are not universal.
A grid-connected inverter method based on second-order linear active disturbance rejection control is adopted. Through pre-synchronization control, linear extended state observer and state error feedback control, combined with dual-channel power regulation and frequency domain harmonic quality judgment, voltage and frequency synchronization and dynamic optimization control during grid connection are achieved.
It improves the adaptability and robustness of grid-connected inverters under grid disturbances and load fluctuations, avoids current surges and frequency drift, and ensures the stability and power quality of the grid connection process.
Smart Images

Figure CN121238686B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of grid-connected inverter control, and specifically to a grid-connected inverter control method based on second-order linear active disturbance rejection. Background Technology
[0002] As the proportion of renewable energy integration continues to increase, the stability of inverters operating in the power system is becoming increasingly prominent. Especially in scenarios dominated by high-proportion inverter sources, such as microgrids and photovoltaic energy storage systems, the control strategy of grid-connected inverters directly affects the stability and power quality of the system.
[0003] The widely adopted virtual synchronous generator (VSG) control strategy achieves a certain degree of power frequency regulation response by simulating the inertia and damping of a synchronous generator. However, traditional VSG control only achieves primary frequency regulation, and when frequency deviation occurs, the system cannot fully restore the frequency to the rated value. Furthermore, this control strategy often experiences a surge in current at the moment of grid connection, leading to active power overshoot, frequency overshoot, increased system oscillation, and even malfunctions of system protection, resulting in reduced reliability.
[0004] Some improved control strategies attempt to introduce PI regulation, model predictive control (MPC), or feedforward disturbance compensation, but they often suffer from slow response speed, lack of parameter universality, and grid connection switching delays. Their performance is particularly inadequate in the following special micro-scenarios: islanded grid connection in low-inertia small microgrids, aggravated disturbance propagation at a single node in multi-inverter grid-connected systems, or grid connection switching after a short-term voltage shift. Summary of the Invention
[0005] The purpose of this invention is to provide a grid-connected inverter control method based on second-order linear active disturbance rejection, so as to solve the shortcomings in the prior art.
[0006] To achieve the above objectives, the present invention provides the following technical solution: a grid-connected inverter control method based on second-order linear active disturbance rejection, comprising:
[0007] S100: Acquire the voltage and frequency error signals between the inverter output side and the power grid;
[0008] S200: Before grid connection, a pre-synchronization control model is constructed. The phase difference, frequency difference and voltage amplitude difference between the inverter and the grid are obtained through Park transformation. Phase-locked loop, proportional-integral regulator and modulation ratio regulation are used respectively to achieve voltage amplitude and frequency synchronization.
[0009] S300: During the frequency synchronization control process, a second-order linear active disturbance rejection controller (LADRC) is introduced. The output frequency, frequency change rate and disturbance term are estimated in real time through the linear extended state observer (LESO). Combined with the state error feedback control law (LSEF), the frequency regulation control quantity is generated to drive the inverter output frequency to converge to the grid reference frequency.
[0010] S400: After completing the synchronization of frequency and voltage amplitude, it performs a closed switch operation and switches the control structure to the LADRC main control mode, retains the LESO and LSEF control paths, and eliminates the integral regulator.
[0011] S500: After grid connection is completed, the grid current signal is collected in real time, and a dual-channel LADRC power regulation controller is constructed based on the dq coordinate system. The active power and reactive power are decoupled and tracked separately, and the inverter output current waveform is controlled by disturbance estimation and feedback regulation.
[0012] S600: Performs periodic fast Fourier transform on the current waveform to obtain frequency domain harmonic components, calculates the total harmonic distortion rate and compares it with a set threshold. When the total harmonic distortion rate is lower than the threshold, it is determined to be in a stable grid-connected state without impact.
[0013] Preferably, step S200 includes:
[0014] S201: Acquires the three-phase voltage signal output from the inverter. and the three-phase voltage signal of the power grid The voltage vector magnitude and phase information are obtained by transforming them to a synchronous rotating coordinate system using the Park transformation.
[0015] S202: Based on the transformation results, calculate the instantaneous phase difference Δθ, frequency difference Δω, and voltage amplitude difference ΔU between the inverter output voltage and the grid voltage;
[0016] S203: Construct a phase-frequency dual-channel closed-loop pre-synchronization control model. Phase synchronization uses a phase-locked loop-based correction controller to adjust the phase difference Δθ. Frequency synchronization uses a proportional-integral controller to adjust Δω. Voltage amplitude synchronization uses an amplitude comparator to control the SVPWM modulation ratio to achieve closed-loop regulation of the output voltage amplitude.
[0017] S204: In the three-channel control closed loop, set the synchronization threshold respectively. , , Only if the following conditions are met: When the pre-synchronization is completed, the subsequent grid connection switching command can be executed.
[0018] Preferably, step S300 includes:
[0019] S301: Based on the frequency control target of the grid-connected inverter, construct a mathematical model of the controlled object in the frequency control process, and compare the output angular frequency ω with the reference angular frequency. The frequency difference Δω is used as the observation input;
[0020] S302: Establish the LESO state space structure for estimating state variables and extended states;
[0021] S303: The controller output is designed using an estimated state error feedback method;
[0022] S304: Combine the estimated extended state set with the LSEF control law to form a complete second-order linear active disturbance rejection controller (LADRC).
[0023] Preferably, the output control quantity u of the LADRC controller is input to the inverter as the frequency input of the virtual synchronous generator controller, which dynamically drives the internal reference frequency of the inverter to track the grid frequency, thereby controlling the frequency of the inverter output voltage.
[0024] Preferably, step S400 includes:
[0025] S401: Set frequency synchronization threshold Voltage amplitude synchronization threshold ,in: The maximum allowable frequency difference; This represents the maximum permissible voltage amplitude difference.
[0026] And the current inverter output angular frequency With reference angular frequency Inverter output voltage amplitude With grid voltage amplitude Perform difference calculations to form synchronization criteria: ;
[0027] S402: When the synchronization conditions are met simultaneously, the grid connection control logic is started, the grid connection enable signal is output, the closed switch between the inverter and the grid is activated, and the output current is smoothly transitioned to the grid connection state through the shockless switching mechanism.
[0028] S403: When starting grid-connected control, automatically switch the control structure, retain the LADRC main control path, and switch the control structure to the LADRC main control mode.
[0029] S404: After grid connection is completed, the frequency is continuously dynamically adjusted by the LADRC controller to compensate for frequency offset in real time.
[0030] Preferably, step S500 includes:
[0031] S501: After the inverter completes the grid connection switch, it collects the three-phase AC current signal output by the inverter and injected into the grid in real time, and converts it into direct axis and quadrature axis components in the synchronous rotating coordinate system, so as to correspond to the adjustment reference of active power and reactive power respectively.
[0032] S502: Construct a dual-channel power regulation controller based on LADRC, wherein: the first channel takes the difference between the desired active power setpoint and the direct-axis current component as input; the second channel takes the difference between the desired reactive power setpoint and the quadrature-axis current component as input;
[0033] The disturbance term, grid fluctuation and current response error state are estimated by LESO respectively, and the state error feedback law is designed by LSEF to form independent direct-axis and quadrature-axis current control outputs to drive the modulator to adjust the output current waveform.
[0034] S503: The direct-axis current control quantity generated by the LADRC dual-channel controller With cross-axis current control quantity Real-time control of the inverter output current can be achieved by adjusting the PWM duty cycle or current reference command.
[0035] Preferably, step S600 includes:
[0036] S601: After the inverter completes grid connection, a preset harmonic analysis time window is set, the three-phase AC current signal on the grid-connected output side is periodically sampled, and the frequency domain component in the current signal is extracted using the fast Fourier transform algorithm to obtain the amplitude of each order harmonic component of the current.
[0037] S602: Based on frequency domain components, calculate the total harmonic distortion rate of the grid-connected current. It is defined as the ratio of the sum of the squares of the effective values of each harmonic component excluding the fundamental wave to the effective value of the fundamental wave, and the square root of the sum is used to measure the degree of distortion of the current current waveform.
[0038] Preferably, S600 further includes: setting a total harmonic distortion rate (THR) judgment threshold; when the calculated THR of the grid-connected current is less than or equal to the THR judgment threshold, determining that the current grid-connected operation state is a stable grid-connected state without impact, and using the judgment result as a stable operation flag to activate the trigger signal.
[0039] The technical effects and advantages provided by the present invention in the above technical solution are as follows:
[0040] 1. This invention introduces a grid-connected inverter control method based on second-order linear active disturbance rejection control (LADRC), combined with pre-synchronization closed-loop regulation, disturbance observation and feedback control, dual-channel decoupled current regulation, and frequency domain harmonic quality assessment, to achieve dynamic optimization control throughout the entire process from grid connection preparation and synchronization control to grid-connected operation. This method can effectively improve the adaptability and robustness of grid-connected inverters under complex operating conditions such as grid disturbances, load fluctuations, and frequency offsets.
[0041] 2. By introducing LESO observation disturbance, LSEF state feedback and THD-based grid connection quality judgment mechanism, this invention achieves high-precision identification and adaptive adjustment of key states during grid connection, avoiding problems such as current surge, frequency drift and integral saturation. It can be widely applied to grid connection control of photovoltaic power generation, energy storage systems and distributed power sources. Attached Figure Description
[0042] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.
[0043] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation
[0044] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0045] For examples, please refer to Figure 1 As shown in this embodiment, a grid-connected inverter control method based on second-order linear active disturbance rejection includes:
[0046] S100: Acquire the voltage and frequency error signals between the inverter output side and the power grid;
[0047] S200: Before grid connection, a pre-synchronization control model is constructed. The phase difference, frequency difference and voltage amplitude difference between the inverter and the grid are obtained through Park transformation. Phase-locked loop, proportional-integral regulator and modulation ratio regulation are used respectively to achieve voltage amplitude and frequency synchronization.
[0048] S300: During the frequency synchronization control process, a second-order linear active disturbance rejection controller (LADRC) is introduced. The output frequency, frequency change rate and disturbance term are estimated in real time through the linear extended state observer (LESO). Combined with the state error feedback control law (LSEF), the frequency regulation control quantity is generated to drive the inverter output frequency to converge to the grid reference frequency.
[0049] S400: After completing the synchronization of frequency and voltage amplitude, it performs a closed switch operation and switches the control structure to the LADRC main control mode, retains the LESO and LSEF control paths, and eliminates the integral regulator.
[0050] S500: After grid connection is completed, the grid current signal is collected in real time, and a dual-channel LADRC power regulation controller is constructed based on the dq coordinate system. The active power and reactive power are decoupled and tracked separately, and the inverter output current waveform is controlled by disturbance estimation and feedback regulation.
[0051] S600: Performs periodic fast Fourier transform on the current waveform to obtain frequency domain harmonic components, calculates the total harmonic distortion rate and compares it with a set threshold. When the total harmonic distortion rate is lower than the threshold, it is determined to be in a stable grid-connected state without impact.
[0052] In implementing this invention, to achieve stable control and rapid response of the grid-connected inverter under different operating conditions, it is first necessary to comprehensively monitor and model the key physical quantities during the grid connection process. This step aims to obtain the voltage, frequency, and power error signals between the inverter output side and the grid side, and to construct the master control variable set for subsequent control logic based on these signals. This is the foundation and prerequisite for the entire control strategy.
[0053] Specifically, before an inverter is in normal operation or about to be connected to the grid, there is usually a certain degree of voltage amplitude difference, frequency deviation, and power difference between its output side and the grid side. If these are not effectively adjusted, direct grid connection may lead to inrush current, sudden changes in active power, frequency disturbances, or even grid connection failure. Therefore, the following parameters need to be collected and an error signal constructed:
[0054] Voltage Deviation: The inverter output voltage is set to... The grid voltage is Then voltage error The expression for (voltage amplitude difference) is: This error reflects whether the voltage amplitudes on both sides are synchronized, and is one of the important parameters for grid-connected synchronous regulation.
[0055] Frequency Deviation: The inverter output angular frequency is set to... The reference angular frequency of the power grid is Typically, this is 50 Hz or 60 Hz, then the frequency error (frequency difference) is... for: Frequency error is used to characterize whether the inverter frequency closely follows the grid operation. It is a core input in frequency control loops, such as Linear Active Disturbance Rejection Control (LADRC) technology or PI control.
[0056] Active Power Error: Assume the inverter output active power is... The reference active power of the power grid is ,but: This error reflects whether the current power output meets dispatch requirements, and is particularly helpful in supporting grid frequency regulation.
[0057] Reactive Power Error: Let the inverter output reactive power be... The reference reactive power of the power grid is ,but: Reactive power error is mainly used to control the output voltage amplitude of the inverter, and is often used in conjunction with voltage closed-loop control.
[0058] To improve control accuracy and response speed, this invention also includes a high-speed synchronous sampling device in this step for real-time acquisition of three-phase voltage, current, and power parameters. The sampling frequency is set as follows: ;in, The grid frequency is typically 50 Hz, so the sampling frequency should be no less than 1 kHz to ensure the timeliness and robustness of grid-connected control.
[0059] Acquire three-phase voltage signals from the inverter output side With grid-side voltage signal The signal sampling frequency should be no less than 1 kHz to ensure synchronization accuracy.
[0060] The Park transformation (also known as synchronous rotating coordinate transformation) is used to convert the voltage signal in the three-phase stationary coordinate system to the synchronous rotating coordinate system (dq coordinate system) to obtain the direct-axis component of the voltage vector. Cross-axis components The specific transformation formula is as follows: Where θ represents the transformation angle, which can be obtained through a phase-locked loop. This represents the instantaneous value of the A-phase voltage on the inverter output side; This represents the instantaneous value of the B-phase voltage on the inverter output side; This represents the instantaneous value of the C-phase voltage on the inverter output side; through the above transformation, the voltage vector magnitude and phase angle of the inverter and the grid voltage in the synchronous rotating coordinate system are obtained.
[0061] The voltage amplitude can be calculated using the following formula: The phase angle can be determined by... get.
[0062] The obtained voltage amplitude and phase information in the dq coordinate system were used to calculate three key error signals:
[0063] Instantaneous phase difference Δθ: represents the voltage vector phase error between the inverter and the grid, and is calculated as follows: ; This represents the instantaneous electrical phase angle of the inverter output voltage relative to the reference time base at the current moment. This represents the reference phase angle of the three-phase AC voltage on the grid side.
[0064] Frequency difference Δω: The frequency difference is approximated using the rate of phase change;
[0065] Voltage amplitude difference ΔU: calculated from the difference in magnitude of the voltage vectors on both sides.
[0066] The aforementioned error signal serves as the main control variable input for pre-synchronization control, providing fundamental data for the next step of constructing a closed-loop control model.
[0067] The calculated error signal is input to a three-channel synchronous control structure, including a phase synchronization channel, a frequency synchronization channel, and a voltage synchronization channel.
[0068] Phase synchronization control channel: Employs a phase-locked loop (PLL), specifically including a phase detector, low-pass filter, and VCO (voltage-controlled oscillator). The PLL outputs an angular frequency. Adjusted by instantaneous phase difference Δθ: ;in , The gain parameter of the proportional-integral controller, the PLL output angular frequency The phase angle is used for synchronous inverters, where t is the sampling time.
[0069] Frequency synchronization control channel: Adjusts Δω and uses a PI controller to generate frequency correction values. The expression is: This control variable is used to adjust the inverter frequency reference input. This represents the adjustment coefficient of the proportional element in the frequency control controller. This indicates the weight of the integral element in the frequency regulation controller.
[0070] Voltage amplitude synchronization control channel: By comparing ΔU in real time, the SVPWM modulation ratio M is controlled to adjust the inverter output voltage amplitude. The modulation ratio adjustment method is as follows: ;in The initial modulation ratio, The modulation sensitivity coefficient.
[0071] To ensure that the inverter's electrical state is strictly aligned with the grid before grid connection, the following three synchronization criterion thresholds are set in the three-channel closed-loop control system:
[0072] Phase synchronization threshold Typical value is set at 5° (approximately 0.087 radians).
[0073] Frequency synchronization threshold Typical value is set to 0.1 Hz;
[0074] Voltage amplitude synchronization threshold Set to 2% of the grid rated voltage.
[0075] The logic for determining whether synchronization is complete is defined as follows: ;
[0076] Only when all three conditions above are met simultaneously is the system determined to enter the synchronization state and a "grid connection allowed" logic command is sent to the controller.
[0077] Dynamic modeling is performed on the frequency regulation process of a grid-connected inverter within a virtual synchronous generator control framework. This model is used to adjust the inverter's output angular frequency ω to track the grid reference angular frequency. Treating the frequency regulation system as a typical second-order controlled object, its dynamic behavior can be simplified to the following differential equation: ;in: This represents the second derivative of the inverter output angular frequency ω with respect to time, where u is the control input (determined by the controller output). The control gain represents the linear proportional relationship between the control input and the frequency response, which can be obtained through modeling or identification; f(t) is the total disturbance term, which represents the equivalent disturbance including modeling error, external disturbances (such as load fluctuations and grid disturbances), and unmodeled dynamics.
[0078] Introducing state variables: (Angular frequency) Let be the rate of change of frequency; then the state-space expression is: This model forms the basis of the controlled objects in LADRC design.
[0079] To estimate the true state and the perturbation term, an Extended State Observer (LESO) is designed. The perturbation term f(t) is introduced as part of the extended state, and we set: ;
[0080] Therefore, the extended dynamic can be written in third-order form: In the formula, Represents state variables The first derivative with respect to time, Represents state variables The first derivative with respect to time, Represents state variables The first derivative with respect to time;
[0081] because Generally unmeasurable, it is therefore treated as a disturbance uncertainty in the observer. LESO is used to estimate... , , Given the value of , construct the following observer: Where: y=ω is the controlled variable, that is, the actual measured inverter output angular frequency; , , They are respectively for , , The estimated values are frequency, rate of change of frequency, and disturbance term; Representing state variables respectively , , Rate of change over time 、 、 The observer gain coefficient for LESO depends on the reference angular frequency. set up: ;in, The typical value range is 20~200 rad / s, which can be adjusted according to the response requirements.
[0082] The LESO structure estimates the current frequency state and disturbance terms in real time through simple linear calculations, exhibiting high robustness and low computational complexity, making it suitable for real-time control scenarios.
[0083] Using LESO estimation results , , A linear state error feedback controller (LSEF) is constructed to generate the control input u of the frequency control loop. The control law is defined as follows: ;in: The reference angular frequency given to the power grid (e.g., 2π×50 rad / s); For frequency error estimation; For estimation of the rate of change of frequency; For disturbance estimation; , The feedback gain parameter of the controller can be set according to the closed-loop pole placement principle; To control the gain, it is usually obtained through identification, but it can also be set as an empirical value.
[0084] The control law can be structurally viewed as having three contributing parts:
[0085] Adjustment of frequency error ;
[0086] Regulation of dynamic inertia ;
[0087] Active suppression of external disturbances and modeling errors .
[0088] This feedback control strategy can respond quickly when disturbances occur, effectively maintaining frequency stability.
[0089] The LESO and LSEF mentioned above are combined to form a complete second-order linear active disturbance rejection controller (LADRC). Its workflow is as follows:
[0090] Collect the inverter output angular frequency ω;
[0091] Real-time estimation of current frequency status via LESO Rate of change of frequency Disturbance terms ;
[0092] The estimation results are input into the LSEF controller to calculate the control input u.
[0093] The control quantity u is sent to the inverter frequency control loop to achieve closed-loop self-adjustment of the frequency.
[0094] The controller outputs the dynamic adjustment result and inputs it to the virtual synchronous generator control module to adjust the frequency reference value in the inverter voltage controller.
[0095] The aforementioned LESO (Linear Extended State Observer) design allows for real-time estimation of key state variables in the frequency control loop. These include the following three:
[0096] First extended state : Represents the estimated value of the current frequency, corresponding to the angular frequency on the inverter output side;
[0097] Second extended state : Represents the estimated rate of change of frequency, corresponding to the first derivative of frequency, i.e., acceleration or inertial response;
[0098] Third extended state : Represents the estimated equivalent total disturbance, which includes external disturbances, model uncertainties, load fluctuations, grid voltage disturbances, and other factors. It is the core basis for dynamic compensation control.
[0099] This extended state set is output by the LESO module and updated in real time for each control cycle. It has continuous and dynamic response characteristics and is suitable for direct use in frequency controllers.
[0100] The controller output u generated above is used as the input to the frequency control loop and directly fed into the inverter main control module or the virtual synchronous generator (VSG) frequency setting module. This control quantity is used to correct the mechanical torque or frequency command in the VSG model, thereby driving the internal frequency response of the inverter.
[0101] In actual control, the inverter will gradually adjust its output angular frequency ω according to the value of the controller output u, making it gradually approach the value of the controller output u. Ultimately, frequency convergence is achieved.
[0102] Since the controller already includes a disturbance compensation term. Therefore, under conditions such as grid frequency disturbances and sudden changes in active load, the control system can still maintain rapid frequency stability without relying on redundant standby units or multi-level lag compensation.
[0103] This closed-loop control structure also supports "online adjustment," meaning that the LESO state estimate and the controller output are updated synchronously and in real time during frequency control execution. The control system recalculates the state estimate at each sampling period (e.g., 1 millisecond). , , And regenerate u, thereby achieving a continuous, stable, and smooth frequency tracking control process.
[0104] After the frequency and voltage amplitude pre-synchronization control stage is completed, in order to ensure that the frequency and voltage consistency between the inverter and the grid side meets the grid connection conditions, it is necessary to first set the synchronization error threshold parameters for determining grid connection permission, as follows:
[0105] Frequency synchronization threshold, denoted as This value is used to determine the maximum permissible deviation between the inverter output frequency and the grid frequency. A typical setting for this value is 0.1 Hz, meaning that if the difference between the inverter frequency and the grid frequency does not exceed 0.1 Hz, they can be considered synchronized.
[0106] Voltage amplitude synchronization threshold, denoted as This is used to determine the maximum permissible error between the inverter's output voltage amplitude and the grid voltage amplitude. It is generally set to 2% of the grid's rated voltage (e.g., 220V). ≈4.4 V.
[0107] In actual implementation, the following two synchronization differences are calculated respectively:
[0108] Frequency difference Δω = current inverter output frequency minus grid frequency;
[0109] Voltage amplitude difference ΔU = Current inverter output voltage amplitude minus grid voltage amplitude.
[0110] When both of the above errors simultaneously meet the following conditions:
[0111] The absolute value of the frequency difference is less than or equal to ;
[0112] The absolute value of the voltage amplitude difference is less than or equal to ,like If the inverter meets the grid connection conditions, a "grid connection allowed" control signal is generated, and subsequent grid connection control operations are executed.
[0113] Once the above synchronization criteria are met, the grid connection control logic process is immediately initiated, including the following key actions:
[0114] Grid connection permission signal issued: The main controller outputs a "grid connection permission" control signal to the grid connection execution module, with the logic value set to 1 (high level), which is used to activate the closed switch connected to the power grid, such as a relay or electronic switching device;
[0115] Close the power path between the inverter and the grid: The inverter output side and the grid side are physically connected through the closing of the switch, the current path is formally established, and active power is sent to the grid.
[0116] Implementing a shockless switching control strategy: To avoid inrush current during switching, this invention introduces a "current smoothing control mechanism," which includes the following two points:
[0117] Gradually increase the output current a short period of time (e.g., 5-10 ms) before grid connection to bring it close to the target value after grid connection and reduce current surges.
[0118] Maintaining the inverter output voltage amplitude and phase unchanged allows the output current to transition naturally towards the grid voltage vector direction, preventing harmonic surges or tripping.
[0119] This strategy achieves "soft grid connection" by coordinating the PWM modulator and phase-locked loop (PLL) inside the inverter, without the need for an external buffer module.
[0120] Simultaneously with the grid connection control startup, the internal structure of the controller also undergoes synchronous switching. The frequency control structure of this invention previously included two branches: one is the main control path based on second-order linear active disturbance rejection (LADRC), and the other is a dynamic PI control branch used to accelerate the frequency response during the pre-synchronization phase.
[0121] To avoid frequency fluctuations or control conflicts caused by PI integral term accumulation or adjustment lag after grid connection, the following operations should be performed simultaneously with the start of grid connection control:
[0122] Shutting down the dynamic PI control branch: Disconnecting the PI output from the main controller fusion path by controlling the logic gate;
[0123] Activate LADRC independent operation mode: retain the LESO output state set (i.e., frequency, rate of change of frequency, disturbance estimate) to feed back to LSEF, and calculate the control quantity as the only frequency regulation input;
[0124] The control quantity fusion method for this structure switching can be implemented by the following methods:
[0125] Set the control output fusion coefficient α=1, meaning the controller fully adopts LADRC output and ignores PI output; or directly set the PI output to zero, thus removing it from the control process.
[0126] The switching process is executed using software instructions and state variables to ensure continuous output and smooth control during switching.
[0127] After the grid connection switch is completed, the controller enters a stable operation phase, where frequency regulation is performed solely by the LADRC controller. Its operating mechanism is as follows:
[0128] LESO state updates in real time: The controller continuously calls the Linear Extended State Observer (LESO) to estimate the current frequency, rate of frequency change, and disturbances;
[0129] LSEF feedback real-time adjustment: Using a feedback control law, the frequency control quantity is dynamically calculated based on the current state estimate to ensure that the frequency closely follows the grid reference frequency;
[0130] Disabling the integral adjustment channel: The system no longer uses any PI or other integral controllers, completely avoiding integral saturation problems;
[0131] Dynamic compensation capability: When factors such as load changes, grid disturbances or voltage fluctuations cause frequency deviation, the LADRC controller automatically compensates and adjusts through disturbance estimation terms to maintain stable frequency operation near the reference value.
[0132] After grid connection is completed, the inverter continuously injects three-phase AC current into the grid. In order to control it, it is necessary to first transform its three-phase AC current signal into a synchronous rotating coordinate system (i.e., the dq coordinate system).
[0133] Signal Acquisition: Real-time acquisition of three-phase current signals from the inverter output side via current sensors or sampling modules. ;
[0134] Clark transformation: transforms three-phase static coordinate current to a two-phase static coordinate system;
[0135] Park transformation: The current in the two-phase static coordinate system is further projected onto a synchronous rotating reference coordinate system, namely the dq coordinate system. The rotational angular velocity of this coordinate system is consistent with the grid frequency, and the angle input is provided by the phase-locked loop (PLL).
[0136] Ultimately, two key control variables were obtained:
[0137] Direct-axis current component : Reflects the active power component of the inverter output;
[0138] quadrature axis current components : Reflects the reactive power component of the inverter output.
[0139] These two current components are used for subsequent comparison of active and reactive power control targets and feedback control, respectively.
[0140] After obtaining the current feedback in the dq coordinate system, a dual-channel controller is constructed to control the current feedback in the dq coordinate system. and ,make Tracking preset reference value ,make Tracking preset reference value This enables dynamic and adjustable tracking of active and reactive power.
[0141] Dual-channel controller input definition:
[0142] First control channel input ;
[0143] Second control channel input ;
[0144] These are the errors between the currently measured current and the set reference value.
[0145] For each control channel, a set of second-order linear extended state observers (LESO) are constructed to estimate the true state and equivalent disturbance of the control system.
[0146] For the d-axis channel:
[0147] State variables ;
[0148] Extended state =Total disturbance, including load changes, grid disturbances, modeling errors, etc.;
[0149] LESO estimation output: ;in, Represents the actual state variable The estimation results; Represents the actual state variable The estimation results; Represents the actual state variable The estimation results;
[0150] For the q-axis channel:
[0151] State variables ;
[0152] Extended state =Total disturbance;
[0153] Based on the above LESO output results, a linear state error feedback controller (LSEF) is constructed:
[0154] The control law is defined as: .
[0155] The control quantity output by the LADRC controller mentioned above and This is converted into command inputs for controlling the inverter output current through a control mapping relationship. The control path includes the following two typical methods:
[0156] PWM duty cycle adjustment method:
[0157] Will and The resulting vector is synthesized into a voltage control vector in a two-phase static coordinate system.
[0158] The space vector PWM (SVPWM) algorithm is used to map the control vector to the PWM duty cycle of the inverter;
[0159] By controlling the PWM drive signal and adjusting the on-time of the switching devices, precise control of the shape and amplitude of the output current can be achieved.
[0160] Current reference value method:
[0161] Will and As a reference current command;
[0162] Input to the current inner loop controller (such as PR control or current loop PI);
[0163] It generates a current-controlled voltage signal to drive the inverter modulator.
[0164] Regardless of the method used, the controller uses the dq current component as the feedback closed-loop variable to form a complete current inner-loop control structure, ensuring that the output current closely follows the set target.
[0165] After the inverter completes the grid connection switch (i.e., physically connects to the grid and begins to inject current into it), the current quality analysis module is started in real time to continuously sample the grid-connected AC current output by the inverter.
[0166] Sampling signal source: Acquiring the three-phase output current of the inverter. High-precision current sensors can be used to collect this data.
[0167] Sampling period and window: Set the time window for harmonic analysis, typically one or more power grid cycles (e.g., 20ms~200ms). The sampling frequency is recommended to be no less than 10 kHz to meet the accuracy requirements of harmonic components.
[0168] FFT Transform Execution: The sampled data is fed into the Fast Fourier Transform (FFT) algorithm to extract the frequency domain characteristics of the current, obtaining the fundamental component and several harmonic amplitudes (e.g., 2nd to 25th). , where n represents the nth harmonic.
[0169] This transformation can be executed in a microcontroller (such as a DSP or FPGA) by calling the standard FFT algorithm library, with the number of sampling points being an integer power of 2 (such as 256, 512, 1024) to improve computational efficiency.
[0170] After obtaining the frequency domain components, calculate the total harmonic distortion (THD) of the current grid-connected current according to the following standard formula: ;in, This represents the amplitude of the fundamental component of the current (50 Hz). It represents the amplitude of the nth harmonic component; THD is expressed as a percentage (%) and is used to quantitatively represent the degree of distortion between the current current waveform and the ideal sine wave.
[0171] To eliminate the spectral leakage caused by phase instability, the sampled signal can be windowed before the FFT operation, such as using a Hanning window or a Blackman window.
[0172] Set a THD threshold εTHD as the criterion for judging grid-connected stability. The typical value range is 3%~7%, and the recommended value is 5%. That is, when the THD of the grid-connected current is less than or equal to 5%, the inverter output current is considered to be stable, the waveform is smooth, and no obvious impact is generated.
[0173] The decision logic is as follows:
[0174] If THD≤εTHD, then it is determined that the current state is a stable grid connection without impact.
[0175] The control system sets the grid-connected stability flag Sstable=1 as the trigger condition for subsequent control processes.
[0176] If THD > εTHD, continue the adjustment process and delay entering the next stage of control.
[0177] This mechanism ensures that the grid connection process is not only a physical connection that meets the requirements of voltage and frequency synchronization, but also has the ability to inject high-quality current, thereby meeting power quality standards.
[0178] Once the grid connection is stable (i.e., Sstable = 1), the controller is allowed to initiate the following advanced scheduling control policies:
[0179] Maximum Power Point Tracking (MPPT): In photovoltaic inverter applications, the MPPT algorithm can be activated based on the voltage-current characteristics of the photovoltaic array to output the optimal power in real time;
[0180] Active power dispatch: Dynamically adjust the active power output of the inverter according to the grid load command or microgrid dispatch command.
[0181] For example, using Maximum Power Point Tracking (MPPT) control specifically includes:
[0182] When the photovoltaic array operates under a light intensity of 1000 W / m², its voltage-current characteristic curve exhibits an approximately single-peak feature, with the maximum power point located at approximately 30 V and 8 A, at which point the power output is approximately 240 W. If a fixed voltage or current is used, the output may only be 180 W to 200 W, resulting in a decrease in energy utilization of approximately 20%. However, when using MPPT algorithms (such as the perturbation-observation method or incremental conductance method), the controller continuously adjusts the DC operating voltage on the inverter input side, keeping the photovoltaic array operating near 30 V, tracking and maintaining the maximum power point in real time, thereby stabilizing the output at around 240 W.
[0183] Furthermore, when the ambient light intensity drops to 600 W / m², the maximum power point of the photovoltaic array will shift, potentially reaching approximately 140W at 28 V and 5 A. The MPPT algorithm in the controller can automatically identify this characteristic change and quickly adjust the operating voltage to 28 V, avoiding additional power loss caused by operating at a fixed voltage.
[0184] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.
Claims
1. A grid-connected inverter control method based on second-order linear active disturbance rejection, characterized in that: include: S100: Acquire the voltage and frequency error signals between the inverter output side and the power grid; S200: Before grid connection, a pre-synchronization control model is constructed. The phase difference, frequency difference and voltage amplitude difference between the inverter and the grid are obtained through Park transformation. Phase-locked loop, proportional-integral regulator and modulation ratio regulation are used respectively to achieve voltage amplitude and frequency synchronization. S300: During the frequency synchronization control process, a second-order linear active disturbance rejection controller (LADRC) is introduced. The output frequency, frequency change rate and disturbance term are estimated in real time through the linear extended state observer (LESO). Combined with the state error feedback control law (LSEF), the frequency regulation control quantity is generated to drive the inverter output frequency to converge to the grid reference frequency. S400: After completing the synchronization of frequency and voltage amplitude, it performs a closed switch operation and switches the control structure to the LADRC main control mode, retains the LESO and LSEF control paths, and eliminates the integral regulator. The S400 step includes: S401: Set frequency synchronization threshold Voltage amplitude synchronization threshold ,in: The maximum allowable frequency difference; This represents the maximum permissible voltage amplitude difference. And the current inverter output angular frequency With reference angular frequency Inverter output voltage amplitude With grid voltage amplitude Perform difference calculations to form synchronization criteria: ; S402: When the synchronization conditions are met simultaneously, the grid connection control logic is started, the grid connection enable signal is output, the closed switch between the inverter and the grid is activated, and the output current is smoothly transitioned to the grid connection state through the shockless switching mechanism. S403: When starting grid-connected control, automatically switch the control structure, retain the LADRC main control path, and switch the control structure to the LADRC main control mode. S404: After grid connection is completed, the frequency is continuously dynamically adjusted by the LADRC controller to compensate for frequency offset in real time; S500: After grid connection is completed, the grid current signal is collected in real time, and a dual-channel LADRC power regulation controller is constructed based on the dq coordinate system. The active power and reactive power are decoupled and tracked separately, and the inverter output current waveform is controlled by disturbance estimation and feedback regulation. S600: Performs periodic fast Fourier transform on the current waveform to obtain frequency domain harmonic components, calculates the total harmonic distortion rate and compares it with a set threshold. When the total harmonic distortion rate is lower than the threshold, it is determined to be in a stable grid-connected state without impact.
2. The grid-connected inverter control method based on second-order linear active disturbance rejection as described in claim 1, characterized in that: The S200 step includes: S201: Acquires the three-phase voltage signal output from the inverter. and the three-phase voltage signal of the power grid The voltage vector magnitude and phase information are obtained by transforming them to a synchronous rotating coordinate system using the Park transformation. S202: Based on the transformation results, calculate the instantaneous phase difference between the inverter output voltage and the grid voltage. Frequency difference and voltage amplitude difference ; S203: Construct a phase-frequency dual-channel closed-loop pre-synchronization control model. Phase synchronization is achieved by adjusting the phase difference using a correction controller based on a phase-locked loop. Frequency synchronization is achieved using a proportional-integral controller. The voltage amplitude is adjusted synchronously by controlling the SVPWM modulation ratio through an amplitude comparator to achieve closed-loop regulation of the output voltage amplitude; S204: In the three-channel control closed loop, set the synchronization threshold respectively. , , Only if the following conditions are met: When the pre-synchronization is completed, the subsequent grid connection switching command can be executed.
3. The grid-connected inverter control method based on second-order linear active disturbance rejection as described in claim 2, characterized in that: The S300 step includes: S301: Based on the frequency control target of the grid-connected inverter, construct a mathematical model of the controlled object in the frequency control process, and compare the output angular frequency ω with the reference angular frequency. frequency difference As an observation input; S302: Establish the LESO state space structure for estimating state variables and extended states; S303: The controller output is designed using an estimated state error feedback method; S304: Combine the estimated extended state set with the LSEF control law to form a complete second-order linear active disturbance rejection controller (LADRC).
4. The grid-connected inverter control method based on second-order linear active disturbance rejection as described in claim 3, characterized in that: The output control quantity u of the LADRC controller is input to the inverter as the frequency input of the virtual synchronous generator controller, which dynamically drives the internal reference frequency of the inverter to track the grid frequency, thereby controlling the frequency of the inverter output voltage.
5. The grid-connected inverter control method based on second-order linear active disturbance rejection as described in claim 1, characterized in that: The S500 step includes: S501: After the inverter completes the grid connection switch, it collects the three-phase AC current signal output by the inverter and injected into the grid in real time, and converts it into direct axis and quadrature axis components in the synchronous rotating coordinate system, so as to correspond to the adjustment reference of active power and reactive power respectively. S502: Construct a dual-channel power regulation controller based on LADRC, wherein: the first channel takes the difference between the desired active power setpoint and the direct-axis current component as input; the second channel takes the difference between the desired reactive power setpoint and the quadrature-axis current component as input; The disturbance term, grid fluctuation and current response error state are estimated by LESO respectively, and the state error feedback law is designed by LSEF to form independent direct-axis and quadrature-axis current control outputs to drive the modulator to adjust the output current waveform. S503: The direct-axis current control quantity generated by the LADRC dual-channel controller With cross-axis current control quantity Real-time control of the inverter output current can be achieved by adjusting the PWM duty cycle or current reference command.
6. The grid-connected inverter control method based on second-order linear active disturbance rejection as described in claim 5, characterized in that: The S600 step includes: S601: After the inverter completes grid connection, a preset harmonic analysis time window is set, the three-phase AC current signal on the grid-connected output side is periodically sampled, and the frequency domain component in the current signal is extracted using the fast Fourier transform algorithm to obtain the amplitude of each order harmonic component of the current. S602: Based on frequency domain components, calculate the total harmonic distortion rate of the grid-connected current. It is defined as the ratio of the sum of the squares of the effective values of each harmonic component excluding the fundamental wave to the effective value of the fundamental wave, and the square root of the sum is used to measure the degree of distortion of the current current waveform.
7. The grid-connected inverter control method based on second-order linear active disturbance rejection as described in claim 6, characterized in that: The S600 further includes: setting a total harmonic distortion rate (THR) judgment threshold; when the calculated THR of the grid-connected current is less than or equal to the THR judgment threshold, the current grid-connected operation state is determined to be a stable grid-connected state without impact, and the judgment result is used as a stable operation flag to activate the trigger signal.
Citation Information
Patent Citations
Double-closed-loop control method for direct-current bus voltage of wind power grid-connected inverter
CN114243753A
Wind power system maximum power tracking control method based on improved LADRC
CN118815658A