Operation control method of wind power generation system

Abstract

The invention relates to the technical field of wind power generation, and discloses a wind power generation system operation control method comprising the following steps: S1, data acquisition; s2, state estimation; s3, generating a control strategy; and S4, executing and switching. According to the operation control method of the wind power generation system, the advanced extended state observer and the generalized extended state observer accurately estimate dynamic interference caused by wind speed sudden change and power grid voltage fluctuation; at a generator end, an advanced extended state observer processes rotating speed and current signals, a five-order state observation model is built, and rotating speed compensation amount and torque compensation amount are output in real time; at a power grid end, the generalized extended state observer processes a direct current bus voltage and a power grid side current signal, and a four-order state observation model is constructed, so that a voltage compensation amount is output; therefore, the rotating speed control error of the generator end is effectively reduced, the direct-current bus voltage of the power grid end is stabilized, and the utilization efficiency of wind energy and the quality of electric energy are further improved.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power generation, and particularly to an operation control method for a wind power generation system. Background Art

[0002] The operation control of a wind power generation system is a core technology that realizes wind energy capture, power conversion, and stable grid connection by adjusting the rectifier on the generator side and the inverter on the grid side in real time. The key lies in precisely coping with uncertain factors such as wind speed fluctuations and grid disturbances to ensure the high-efficiency power generation and reliable operation of the system. Existing technologies mainly rely on sensors to collect signals such as rotational speed, voltage, and current, and then the controller generates drive instructions to adjust the converter. However, there are key technical bottlenecks under complex operating conditions.

[0003] Currently, the technology mainly faces two major problems: First, the traditional observer has insufficient estimation accuracy for dynamic disturbances such as sudden changes in wind speed and fluctuations in grid voltage, which leads to large control errors in the rotational speed on the generator side and unstable DC bus voltage on the grid side, thus affecting the utilization efficiency of wind energy and the quality of electrical energy. Second, under low wind speed conditions, the doubly-fed wind turbine generator set is restricted by the rotor side voltage, resulting in a relatively high minimum grid connection speed, usually greater than or equal to 600 revolutions per minute. Moreover, the existing control strategy lacks an efficient mode switching mechanism, which makes it unable to fully utilize low wind speed resources. Therefore, the present invention proposes an operation control method for a wind power generation system. Summary of the Invention

[0004] (1) Technical Problems to be Solved

[0005] Aiming at the deficiencies of the existing technology, the present invention provides an operation control method for a wind power generation system, which has the advantages of improving the disturbance estimation accuracy and control robustness through a high-order observer, variable coefficient sliding mode control, and dual-mode switching technology, and solves the problems that the traditional observer has insufficient estimation accuracy for dynamic disturbances such as sudden changes in wind speed and fluctuations in grid voltage, which leads to large control errors in the rotational speed on the generator side and unstable DC bus voltage on the grid side, thus affecting the utilization efficiency of wind energy and the quality of electrical energy. Second, under low wind speed conditions, the doubly-fed wind turbine generator set is restricted by the rotor side voltage, resulting in a relatively high minimum grid connection speed, usually greater than or equal to 600 revolutions per minute. Moreover, the existing control strategy lacks an efficient mode switching mechanism, which makes it unable to fully utilize low wind speed resources.

[0006] (2) Technical Solutions

[0007] To achieve the above purpose of improving the disturbance estimation accuracy and control robustness through a high-order observer, variable coefficient sliding mode control, and dual-mode switching technology, the present invention provides the following technical solutions: An operation control method for a wind power generation system, including the following steps:

[0008] S1. Data acquisition: High-precision sensors are used to obtain the generator operation parameters in real time, as follows:

[0009] An incremental encoder is used to collect the generator speed signal, and the sampling frequency is ≥10 kHz;

[0010] A Hall effect voltage sensor is configured to collect the DC bus voltage and the three-phase grid voltage signals;

[0011] A Rogowski coil current sensor is used to collect the three-phase stator current and the rotor excitation current signals;

[0012] S2. State estimation:

[0013] In the machine-side control module, the speed and current signals are processed by a high-order extended state observer to construct a fifth-order state observation model. This model includes speed observation, speed differential observation, torque disturbance and its first / second derivative observations. The recursive difference algorithm is used to estimate the dynamic disturbance caused by sudden wind speed changes in real time, and the speed compensation amount and torque compensation amount are output;

[0014] In the grid-side control module, the DC bus voltage and grid-side current signals are processed by a generalized extended state observer to construct a fourth-order state observation model, and the total disturbance caused by grid voltage fluctuations and nonlinear loads is estimated in real time, and the voltage compensation amount is output;

[0015] S3. Control strategy generation:

[0016] Machine-side rectifier control: Based on the disturbance compensation amount output by the high-order extended state observer, a machine-side converter drive signal is generated by a variable coefficient fast terminal sliding mode controller. This controller dynamically adjusts the control rate according to the sliding mode surface function value, and uses a double power function for segmented adjustment. Among them, the high power segment of 1.5 - 2.0 is used for fast convergence, and the low power segment of 0.5 - 1.0 is used to suppress chattering, and the feedforward compensation algorithm is combined to cancel the torque disturbance;

[0017] Grid-side inverter control: Based on the voltage compensation amount output by the generalized extended state observer, a grid-side converter modulation signal is generated by a variable coefficient fast terminal sliding mode controller. The space vector pulse width modulation technology is used to dynamically adjust the pulse duty cycle according to the grid voltage phase and frequency to achieve the stability of the DC bus voltage and the optimization of the grid-connected power quality;

[0018] S4. Execution and switching:

[0019] Dual-mode operation control: The generator speed is monitored in real time through a speed sensor. When the speed < 520 r / min, the asynchronous mode is triggered: the stator auxiliary contactor is closed to short-circuit the stator winding, and the rotor-side converter independently controls the excitation current, adopting a double-closed-loop structure of power loop - current loop. The input of the power loop is the reference power calculated by the optimal tip speed ratio; when the speed > 1050 r / min, the doubly-fed mode is triggered: the stator main contactor is closed to connect to the grid, and the rotor-side converter adjusts the excitation current based on the stator flux-oriented vector control algorithm, including coordinate transformation, power decoupling, and feed-forward compensation modules.

[0020] Hardware execution unit: IGBT modules are used for the machine-side and grid-side converters. The drive signals are transmitted through optical fibers. The contactor drive circuit is configured with a surge absorption circuit to ensure that the mode switching response time < 20 ms and there is no voltage impact.

[0021] Preferably, the data acquisition stage further includes the following steps:

[0022] A temperature sensor is set to monitor the temperatures of the generator stator and the IGBT modules of the converter; if the temperature exceeds 85 degrees Celsius, the power reduction protection mechanism is started.

[0023] In addition, a vibration sensor is used to monitor the vibration signals of the gearbox of the doubly-fed unit; the spectrum is analyzed through Fourier transform to identify the early signs of mechanical faults.

[0024] Preferably, in the state estimation stage of the high-order extended state observer, the specific implementation is as follows:

[0025] First, a disturbance prediction library including the aerodynamic model of the wind turbine and the electromagnetic model of the generator is constructed; the aerodynamic model pre-stores the tip speed ratio and wind energy utilization coefficient curves corresponding to different pitch angles, while the electromagnetic model calls the stator resistance, stator inductance, and permanent magnet flux linkage parameters in real time. The parameters are obtained through off-line identification, and the identification period is 1 second.

[0026] Second, the Kalman filtering algorithm is used to correct the noise of the observed state; the process noise covariance matrix and the observation noise covariance matrix in the Kalman filter are dynamically updated according to the sensor accuracy and the historical data of the system operation.

[0027] Preferably, the variable coefficient fast terminal sliding mode controller in the control strategy generation stage is composed as follows:

[0028] A real-time sliding mode surface slope calculation module, which dynamically adjusts the parameters of the double-power function according to the current speed error and voltage error; when the error change rate exceeds 10 rad / s 2 the system will automatically increase the weight of the high-power segment to 70%.

[0029] Feedforward compensation calculator. This module adjusts the reference values of the inner current loop and the outer voltage loop according to the disturbance estimation value output by the high-order extended state observer / generalized extended state observer at a ratio of 0.8 - 1.2 times to achieve the compensation purpose. The optimization of the compensation coefficient is carried out in real time through the fuzzy logic algorithm.

[0030] Preferably, the dual-mode operation control also involves the execution and switching phases:

[0031] Synchronization check mechanism before grid connection: Before closing the switch in the doubly-fed mode, the phase difference and frequency difference are detected by the grid voltage sensor. Only when the synchronization error is less than 1° and the frequency difference is less than 0.2 Hz, the closing of the stator main contactor is activated.

[0032] Fault protection logic: Once the DC bus voltage exceeds 110% of the rated value, or the grid frequency fluctuates by more than ±5%, the system will be forced to switch to the safe mode, and the energy is released through the rotor-side chopper circuit. At the same time, a fault signal is sent to the main control system.

[0033] An operation control system for a wind power generation system, including a complete data acquisition - processing - control hardware architecture:

[0034] Data acquisition layer: It includes a speed encoder, voltage sensors, and current sensors. After the output signals of all sensors are processed by the anti-aliasing filter, they are connected to the 16-bit ADC sampling circuit.

[0035] Control operation layer: It includes a machine-side control unit and a grid-side control unit. The machine-side control unit uses a DSP chip, runs the high-order extended state observer algorithm and the variable coefficient sliding mode control algorithm, integrates a hardware floating-point operation unit, and the operation period ≤ 10 μs. The grid-side control unit uses an FPGA chip, executes the generalized extended state observer algorithm and the space vector pulse width modulation algorithm, and uses parallel computing to achieve microsecond-level disturbance response.

[0036] Execution drive layer: It includes an inverter drive module and a contactor control module. The inverter drive module contains an IGBT drive board, supports a drive voltage of ±15V, and is equipped with a dv / dt absorption circuit. The contactor control module uses a solid-state relay to control the stator main / auxiliary contactors, and configures a voltage synchronization detection circuit board for phase matching before grid connection.

[0037] Preferably, the data acquisition layer also includes the following components:

[0038] Sensor calibration module: This module performs calibration operations regularly with a period of 24 hours. It sends calibration signals to each sensor. During the calibration process, it uses a reference resistor or capacitor network to verify the zero offset and gain error of the sensor. If the error exceeds 5%, the system will issue a calibration alarm.

[0039] Data cache unit: This unit uses a dual-port RAM to temporarily store real-time sampled data; the design of this unit supports parallel data reading in the control operation layer, and the update frequency of the cache is consistent with the sampling frequency of the analog-to-digital converter.

[0040] Preferably, a hardware protection circuit is configured between the operation control layer and the execution drive layer:

[0041] Overcurrent protection: When the stator current exceeds 1.5 times the rated current, the IGBT drive signal is immediately blocked through a hardware comparator, and the response time is less than 1 microsecond;

[0042] Undervoltage protection: When the DC bus voltage is lower than 80% of the rated value, the system automatically switches to the asynchronous mode, and reactive power is injected into the power grid through the rotor-side converter to stabilize the voltage.

[0043] (III) Beneficial effects

[0044] Compared with the prior art, the present invention provides an operation control method for a wind power generation system, having the following beneficial effects:

[0045] 1. For the operation control method of this wind power generation system, the advanced extended state observer and the generalized extended state observer are used to accurately estimate the dynamic disturbances caused by wind speed mutations and grid voltage fluctuations; at the generator end, the advanced extended state observer processes the speed and current signals, builds a fifth-order state observation model, and outputs the speed compensation amount and torque compensation amount in real time; at the grid end, the generalized extended state observer processes the DC bus voltage and grid-side current signals, constructs a fourth-order state observation model, and thus outputs the voltage compensation amount; in this way, the speed control error at the generator end is effectively reduced, the DC bus voltage at the grid end is stabilized, and the utilization efficiency of wind energy and the power quality are improved.

[0046] 2. For the operation control method of this wind power generation system, through a dual-mode operation control scheme based on speed, the asynchronous mode is enabled when the generator speed is lower than 520 revolutions per minute, and it switches to the doubly-fed mode when it exceeds 1050 revolutions per minute; this strategy not only reduces the minimum speed threshold for grid connection but also realizes efficient power generation in a low-wind speed environment; specifically, in the asynchronous mode, the system adopts a double closed-loop structure of power loop - current loop to independently control the excitation current; while in the doubly-fed mode, the stator flux-oriented vector control algorithm is used to adjust the excitation current to ensure a smooth transition, avoid voltage fluctuations, make full use of low-wind speed resources, and thus improve the adaptability and stability of the system. Brief description of the drawings

[0047] Figure 1 It is a flowchart of the operation control method for the wind power generation system of the present invention;

[0048] Figure 2 This is the architecture diagram of the operation control system for the wind power generation system of the present invention. Specific embodiments

[0049] Next, in combination with the embodiments and drawings of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present invention.

[0050] Please refer to Figure 1-2 , an operation control method for a wind power generation system, comprising the following steps:

[0051] S1. Data acquisition: High-precision sensors are used to obtain the operation parameters of the generator in real time, specifically as follows:

[0052] An incremental encoder is used to collect the generator speed signal, and the sampling frequency ≥ 10 kHz;

[0053] A Hall effect voltage sensor is configured to collect the DC bus voltage and the three-phase grid voltage signals;

[0054] A Rogowski coil current sensor is used to collect the three-phase stator current and the rotor excitation current signals;

[0055] S2. State estimation:

[0056] In the machine-side control module, the speed and current signals are processed by a high-order extended state observer to construct a fifth-order state observation model. This model includes speed observation, speed differential observation, torque disturbance and its first-order / second-order derivative observation. The recursive difference algorithm is used to estimate the dynamic disturbance caused by the sudden change of the wind speed in real time, and the speed compensation amount and torque compensation amount are output;

[0057] In the grid-side control module, the DC bus voltage and the grid-side current signals are processed by a generalized extended state observer to construct a fourth-order state observation model, and the total disturbance caused by the grid voltage fluctuation and the non-linear load is estimated in real time, and the voltage compensation amount is output;

[0058] S3. Control strategy generation:

[0059] Machine-side rectifier control: Based on the disturbance compensation amount output by the high-order extended state observer, a machine-side converter drive signal is generated by a variable coefficient fast terminal sliding mode controller. This controller dynamically adjusts the control rate according to the value of the sliding mode surface function, and uses a double-power function for segmented adjustment. Among them, the high-power segment 1.5 - 2.0 is used for fast convergence, and the low-power segment 0.5 - 1.0 is used to suppress chattering, and the feedforward compensation algorithm is combined to cancel the torque disturbance;

[0060] Grid-side Inverter Control: Based on the voltage compensation amount output by the generalized extended state observer, a modulation signal for the grid-side converter is generated by a variable coefficient fast terminal sliding mode controller. The space vector pulse width modulation technology is adopted to dynamically adjust the pulse duty cycle according to the phase and frequency of the grid voltage, so as to achieve the stability of the DC bus voltage and the optimization of the grid-connected power quality;

[0061] S4. Execution and Switching:

[0062] Dual-mode Operation Control: The generator speed is monitored in real time through a speed sensor. When the speed < 520 r / min, the asynchronous mode is triggered: the stator secondary contactor is closed to short-circuit the stator winding, and the rotor-side converter independently controls the excitation current, adopting a double closed-loop structure of power loop - current loop, and the input of the power loop is the reference power calculated by the optimal tip speed ratio; when the speed > 1050 r / min, the doubly-fed mode is triggered: the stator main contactor is closed to connect to the grid, and the rotor-side converter adjusts the excitation current based on the stator flux-oriented vector control algorithm, including coordinate transformation, power decoupling, and feed-forward compensation modules;

[0063] Hardware Execution Unit: IGBT modules are used for the machine-side and grid-side converters, and the drive signals are transmitted through optical fibers. The contactor drive circuit is configured with a surge absorption circuit to ensure that the mode switching response time < 20 ms and there is no voltage impact.

[0064] An operating control system for a wind power generation system, including a complete data acquisition - processing - control hardware architecture:

[0065] Data Acquisition Layer: It includes a speed encoder, voltage sensors, and current sensors. After the output signals of all sensors are processed by an anti-aliasing filter, they are connected to a 16-bit ADC sampling circuit;

[0066] Control Operation Layer: It includes a machine-side control unit and a grid-side control unit; the machine-side control unit uses a DSP chip to run the high-order extended state observer algorithm and the variable coefficient sliding mode control algorithm, integrating a hardware floating-point operation unit, and the operation period ≤ 10 μs; the grid-side control unit uses an FPGA chip to execute the generalized extended state observer algorithm and the space vector pulse width modulation algorithm, and uses parallel computing to achieve microsecond-level disturbance response;

[0067] Execution Drive Layer: It includes a converter drive module and a contactor control module; the converter drive module includes an IGBT drive board, supports a ±15V drive voltage, and is equipped with a dv / dt absorption circuit; the contactor control module uses a solid-state relay to control the stator main / secondary contactors, and is configured with a voltage synchronization detection circuit board for phase matching before grid connection.

[0068] Example 1:

[0069] This example details the system architecture of the present invention and the specific implementation of the data acquisition layer, with a focus on sensor configuration, calibration mechanism, and anti-interference design.

[0070] I. Hardware Configuration of the Data Acquisition Layer

[0071] Incremental encoder: An incremental encoder with a resolution of 2500 lines per revolution is selected and installed at the main shaft end of the generator. The signal is transmitted to the control unit on the machine side through optical fiber. The sampling frequency is set to 12 kHz to ensure that the sampling interval of the rotational speed signal ≤ 83.3 μs. Anti-jitter filtering uses a third-order Butterworth filter with a cut-off frequency of 3 kHz.

[0072] Voltage sensor: The DC bus voltage is collected by a Hall effect voltage sensor with a range of ±1500 V, and the sampling rate reaches 100 kHz. The three-phase grid voltage uses a three-phase isolated Hall sensor with an accuracy of ±0.5%, and the sampling frequency is synchronized with the grid frequency, with an accuracy of 50 Hz ± 2 Hz.

[0073] Current sensor: The three-phase stator current is collected by a Rogowski coil sensor with a range of ±500 A and an accuracy of ±1%. The sampling frequency is 20 kHz. The rotor excitation current uses a closed-loop Hall sensor with a range of ±200 A and an accuracy of ±0.2%. The signal is filtered by a 50 Hz notch filter to eliminate grid harmonic interference.

[0074] Temperature and vibration monitoring: The IGBT module temperature sensor uses a PT1000 thermal resistor with an accuracy of ±1°C, and the response delay for triggering the 85°C power reduction protection ≤ 50 ms. The gearbox vibration sensor is a piezoelectric accelerometer with a range of ±50 g. The FFT spectrum analysis uses a 2048-point fast algorithm to identify the fault characteristic frequencies of the gearbox, such as 1× and 2× meshing frequencies.

[0075] II. Sensor Calibration and Data Caching Mechanism

[0076] Calibration module: Automatic calibration is started once every 24 hours, and a calibration signal is generated through a built-in 10 V precision reference source and a 0.1 Ω standard resistor. An alarm is triggered when the zero offset calibration error > 5%, and the calibration history is recorded in the non-volatile memory.

[0077] Data caching unit: A dual-port SRAM with a capacity of 2 MB is used. The real-time sampled data is stored according to the time stamp, and it supports parallel reading by the control layer at a 10 μs cycle. When the cache overflows, the overwrite mode is enabled, and the most recent 100 ms data is preferentially retained for fault traceability analysis.

[0078] III. Anti-Aliasing and Synchronous Sampling

[0079] All analog signals are converted by a 16-bit Σ-Δ ADC with a resolution of 1 mV. The sampling clock is generated by dividing the frequency of a 40 MHz crystal oscillator to ensure that the sampling phase synchronization error between the machine side and the grid side is < ±1 μs. The grid voltage and current signals are synchronously sampled using zero-crossing triggering to eliminate the power calculation error caused by phase shift.

[0080] Embodiment 2:

[0081] This embodiment focuses on the mathematical implementation details of the high-order / generalized extended state observer and the variable coefficient sliding mode control algorithm, including model parameters, filtering strategies, and dynamic adjustment logics.

[0082] I. Construction of the state observation model

[0083] HOESO on the machine side: The fifth-order state variables include the rotational speed ω, the derivative of the rotational speed the torque disturbance T d and its first / second derivatives The observation gain matrix is designed as:

[0084]

[0085] The parameters are identified offline, where the stator resistance R s = 0.05 Ω, the inductance L s = 3 mH, and it is updated by the Luenberger observer every second. The identification error threshold is set to 5%.

[0086] GESO on the grid side: The fourth-order model includes the DC bus voltage V dc , the current I ac and the total grid disturbance f grid and its derivative The observation gain matrix is designed as:

[0087]

[0088] II. Design of the variable coefficient fast terminal sliding mode controller

[0089] Machine side control: The sliding mode surface is designed as:

[0090]

[0091] where λ = 0.5, μ = 1.2; the reaching law adopts a double power function:

[0092]

[0093] When the error change rate is the case, the coefficient k1 of the high power segment is automatically increased from the initial value of 5.0 to 8.5, and k2 is fixed at 2.0.

[0094] Feedforward compensation algorithm: The disturbance compensation amount is calculated as follows:

[0095]

[0096] The compensation coefficient is adjusted by fuzzy logic. The input variables are the observed error and the error change rate. The output membership function adopts a triangular distribution. The rule base contains 15 rules, such as "large error and positive error change rate → increase the compensation coefficient".

[0097] III. Implementation of grid-side space vector modulation

[0098] The grid-side controller adopts a three-level SVPWM algorithm. The modulation wave frequency is synchronized with the power grid at 50Hz ± 0.5Hz; the switching frequency is set to 8kHz, and the carrier ratio N c = 16. The output voltage error is controlled within ±1% of the rated value; the dynamic duty cycle adjustment formula is:

[0099]

[0100] where ΔV obs is the voltage compensation amount output by GESO. When the frequency deviation exceeds ±0.2Hz, the duty cycle compensation is triggered.

[0101] Example 3:

[0102] This example details the double-fed / asynchronous mode switching process, the design of the hardware protection circuit, and the fault response logic.

[0103] I. Dual-mode switching control logic

[0104] Asynchronous mode trigger: When the speed < 520r / min, the stator auxiliary contactor is triggered to close. The rated current of the contact is 3000A, and the closing time ≤ 15ms; at this time, the machine-side converter switches to a double closed-loop of power loop - current loop: the power loop reference value P opt (λ = 7.5), and the current loop bandwidth is set to 200Hz.

[0105] Double-fed mode switching: When the speed > 1050r / min, perform a pre-grid connection synchronization check:

[0106] Phase synchronization: Detect the zero-crossing point of the grid voltage through a phase-locked loop. The allowable phase error ≤ 0.8°, and the frequency error ≤ 0.15Hz;

[0107] Contactor operation: The stator main contactor is closed using a "close first and then break" strategy. The switching time difference between the main / auxiliary contactors ≤ 2ms, and an RC absorption circuit is used, where R = 10Ω, C = 1000pF, to suppress transient overvoltage, and its peak value ≤ 1.3Vdc.

[0108] II. Hardware Protection Circuit Design

[0109] Overcurrent protection: The stator current is detected by a Hall sensor with a response time of 1 μs. When the detected I s>1.5 I rated is detected, the IGBT drive signal is blocked; the hardware circuit includes an independent comparator with a response delay ≤ 0.8 μs.

[0110] Undervoltage protection: When the DC bus voltage is lower than 80% of the rated value, the rotor-side converter switches to the reactive power compensation mode and injects reactive power Q inj = 0.3·V rated ·I rated . Through the PI regulator, where K p = 0.8 and T i = 0.05 s, the d-axis current reference value is dynamically adjusted.

[0111] III. Fault Response and Safety Mode

[0112] Overvoltage protection: When V dc > 1.1V rated , the rotor-side chopper circuit is started. The IGBT switching frequency is 20 kHz, and the energy is dissipated to the braking resistor with a resistance value of 1.2 Ω and a rated power of 500 kW.

[0113] Fault signal transmission: The main control PLC sends the fault code to the SCADA system through the CAN bus at a baud rate of 1 Mbps. For example, 0x03 represents overcurrent protection. At the same time, the fault waveform is recorded in the black box memory with a storage capacity of 1 GB and a sampling rate of 100 kHz.

[0114] Example 4:

[0115] This example illustrates the extended application and parameter optimization method of the system under extreme working conditions, including low wind speed optimization and grid fault response.

[0116] I. Low Wind Speed Operation Optimization

[0117] In the asynchronous mode with a rotational speed < 520 r / min, by adjusting the PI parameters of the power loop, where K p = 0.3 and T i = 2 s, the optimal tip speed ratio tracking is achieved. The specific process is as follows:

[0118] Calculate the optimal λ opt = 7.5 ± 0.3 corresponding to the current wind speed per second. By adjusting the pitch angle, λ is maintained within the range of the target value ± 5%, where the response time of the stepper motor ≤ 500 ms.

[0119] When the wind speed suddenly changes resulting in When enabled, the recursive difference algorithm of HOESO expands the disturbance estimation error bandwidth to 50 Hz.

[0120] Grid voltage dip protection:

[0121] When the grid voltage amplitude < 85% of the rated value and lasts > 200 ms, trigger the following actions:

[0122] The torque reference value on the machine side is reduced to 50% of the rated value, and the mechanical kinetic energy is quickly released through a sliding mode controller, with the release time ≤ 300 ms.

[0123] The grid side switches to the constant power mode, and the d-axis current is regulated to maintain V dc in the 90% - 110% rated range.

[0124] In summary, the operation control method of this wind power generation system uses an advanced extended state observer and a generalized extended state observer to accurately estimate the dynamic disturbances caused by sudden changes in wind speed and grid voltage fluctuations; at the generator end, the advanced extended state observer processes the speed and current signals, constructs a fifth-order state observation model, and outputs the speed compensation amount and torque compensation amount in real time; at the grid end, the generalized extended state observer processes the DC bus voltage and grid side current signals, constructs a fourth-order state observation model, and thus outputs the voltage compensation amount; in this way, the speed control error at the generator end is effectively reduced, the DC bus voltage at the grid end is stabilized, and the utilization efficiency of wind energy and power quality are improved.

[0125] Moreover, the operation control method of this wind power generation system, through a dual-mode operation control scheme based on speed, enables the asynchronous mode when the generator speed is lower than 520 revolutions per minute, and switches to the doubly-fed mode when it exceeds 1050 revolutions per minute; this strategy not only reduces the minimum grid connection speed threshold but also realizes efficient power generation in low wind speed environments; specifically, in the asynchronous mode, the system adopts a double closed-loop structure of power loop - current loop to independently control the excitation current; while in the doubly-fed mode, the stator flux orientation vector control algorithm is used to adjust the excitation current to ensure a smooth transition, avoid voltage fluctuations, make full use of low wind speed resources, thereby enhancing the adaptability and stability of the system, and solving the problem that the traditional observer has insufficient estimation accuracy for dynamic disturbances such as sudden changes in wind speed and grid voltage fluctuations, which leads to large speed control errors on the generator side and unstable DC bus voltage on the grid side, thus affecting the utilization efficiency of wind energy and the quality of electric energy; secondly, under low wind speed conditions, the doubly-fed wind turbine generator set is restricted by the rotor side voltage, resulting in a relatively high minimum grid connection speed, usually greater than or equal to 600 revolutions per minute; moreover, the existing control strategies lack an efficient mode switching mechanism, which makes it impossible to make full use of low wind speed resources.

[0126] All relevant modules involved in this system are hardware system modules or functional modules that combine computer software programs or protocols in the prior art with hardware. The computer software programs or protocols themselves involved in this functional module are all well-known technologies to those skilled in the art and are not the improvements of this system. The improvement of this system lies in the interaction relationship or connection relationship between each module, that is, the overall structure of the system is improved to solve the corresponding technical problems to be solved by this system.

[0127] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for operating and controlling a wind power generation system, characterized in that, Including the following steps: S1. Data acquisition: High-precision sensors are used to obtain the operating parameters of the generator in real time, specifically as follows: An incremental encoder is used to collect the generator speed signal, and the sampling frequency is ≥10 kHz; A Hall effect voltage sensor is configured to collect the DC bus voltage and the three-phase grid voltage signals; A Rogowski coil current sensor is used to collect the three-phase stator current and the rotor excitation current signals; S2. State estimation: In the machine-side control module, the speed and current signals are processed by a high-order extended state observer to construct a fifth-order state observation model. This model includes speed observation, speed differential observation, torque disturbance and its first / second derivative observations. The recursive difference algorithm is used to estimate the dynamic disturbance caused by sudden wind speed changes in real time, and the speed compensation amount and torque compensation amount are output; In the grid-side control module, the DC bus voltage and grid-side current signals are processed by a generalized extended state observer to construct a fourth-order state observation model, and the total disturbance caused by grid voltage fluctuations and nonlinear loads is estimated in real time, and the voltage compensation amount is output; S3. Control strategy generation: Machine-side rectifier control: Based on the disturbance compensation amount output by the high-order extended state observer, a variable coefficient fast terminal sliding mode controller is used to generate the driving signal of the machine-side converter. This controller dynamically adjusts the control rate according to the value of the sliding mode surface function, and uses a double-power function for segmented adjustment. Among them, the high-power segment of 1.5 - 2.0 is used for fast convergence, and the low-power segment of 0.5 - 1.0 is used to suppress chattering, and the feedforward compensation algorithm is combined to cancel the torque disturbance; Grid-side inverter control: Based on the voltage compensation amount output by the generalized extended state observer, a variable coefficient fast terminal sliding mode controller is used to generate the modulation signal of the grid-side converter. The space vector pulse width modulation technology is adopted to dynamically adjust the pulse duty ratio according to the grid voltage phase and frequency to achieve the stability of the DC bus voltage and the optimization of the grid-connected power quality; S4. Execution and switching: Dual-mode operation control: The generator speed is monitored in real time through a speed sensor. When the speed < 520 r / min, the asynchronous mode is triggered: the stator auxiliary contactor is closed to short-circuit the stator winding, and the rotor-side converter independently controls the excitation current, adopting a double closed-loop structure of power loop - current loop. The input of the power loop is the reference power calculated by the optimal tip speed ratio; when the speed > 1050 r / min, the double-fed mode is triggered: the stator main contactor is closed to connect to the grid, and the rotor-side converter adjusts the excitation current based on the stator flux-oriented vector control algorithm, including coordinate transformation, power decoupling and feedforward compensation modules; Hardware execution unit: IGBT modules are used for the machine-side and grid-side converters. The driving signals are transmitted through optical fibers. The contactor drive circuit is configured with a surge absorption circuit to ensure that the mode switching response time < 20 ms and there is no voltage impact.

2. The operating control method of a wind power generation system according to claim 1, characterized in that The data acquisition stage further includes the following steps: A temperature sensor is set to monitor the temperatures of the generator stator and the IGBT module of the converter; if the temperature exceeds 85 degrees Celsius, the power reduction protection mechanism is started; In addition, a vibration sensor is used to monitor the vibration signal of the gearbox of the double-fed unit; the spectrum is analyzed through Fourier transform to identify the early signs of mechanical faults.

3. The operating control method of a wind power generation system according to claim 1, characterized in that, In the state estimation stage of the high-order extended state observer, the specific implementation is as follows: First, construct a disturbance prediction library that includes the aerodynamic model of the wind turbine and the electromagnetic model of the generator; the aerodynamic model pre-stores the tip speed ratio and wind energy utilization coefficient curves corresponding to different pitch angles, while the electromagnetic model calls the stator resistance, stator inductance, and permanent magnet flux linkage parameters in real time. The parameters are obtained through offline identification, and the identification period is 1 second; Second, use the Kalman filter algorithm to correct the noise of the observed state; the process noise covariance matrix and the observation noise covariance matrix in the Kalman filter will be dynamically updated according to the sensor accuracy and the historical data of the system operation.

4. The operating control method of a wind power generation system according to claim 1, characterized in that, The variable coefficient fast terminal sliding mode controller in the control strategy generation stage is composed as follows: Real-time sliding mode surface slope calculation module, which dynamically adjusts the parameters of the double power function according to the current rotational speed error and voltage error; when the error change rate exceeds 10 rad / s 2 , the system will automatically increase the weight of the high power segment to 70%; A feed-forward compensation calculator. This module adjusts the reference values of the current inner loop and the voltage outer loop according to the disturbance estimation value output by the high-order extended state observer / generalized extended state observer at a ratio of 0.8 - 1.2 times to achieve the compensation purpose; the optimization of the compensation coefficient is carried out in real time through the fuzzy logic algorithm.

5. A method for operating and controlling a wind power generation system according to claim 1, characterized in that, The dual-mode operation control also involves the execution and switching stages: Synchronization check mechanism before grid connection: Before closing the knife switch in the double-fed mode, use the grid voltage sensor to detect the phase difference and frequency difference; only when the synchronization error is less than 1° and the frequency difference is less than 0.2 Hz, activate the closing of the stator main contactor; Fault protection logic: Once the DC bus voltage exceeds 110% of the rated value, or the grid frequency fluctuates by more than ±5%, the system will be forced to switch to the safe mode, and the energy will be released through the rotor side chopper circuit; at the same time, a fault signal will be sent to the main control system.

6. An operating control system for a wind power generation system, characterized in that, Include a complete data acquisition - processing - control hardware architecture: Data acquisition layer: It includes a speed encoder, voltage sensors, and current sensors. After the output signals of all sensors are processed by the anti-aliasing filter, they are connected to the 16-bit ADC sampling circuit; Control operation layer: It includes a machine-side control unit and a grid-side control unit; the machine-side control unit uses a DSP chip, runs the high-order extended state observer algorithm and the variable coefficient sliding mode control algorithm, integrates a hardware floating-point operation unit, and the operation period ≤ 10 μs; the grid-side control unit uses an FPGA chip, executes the generalized extended state observer algorithm and the space vector pulse width modulation algorithm, and uses parallel computing to achieve microsecond-level disturbance response; Execution drive layer: It includes an inverter drive module and a contactor control module; the inverter drive module includes an IGBT drive board, supports a ±15V drive voltage, and is equipped with a dv / dt absorption circuit; the contactor control module uses a solid-state relay to control the stator main / auxiliary contactors, and configures a voltage synchronization detection circuit board for phase matching before grid connection.

7. The operating control system of a wind power generation system according to claim 1, characterized in that, The data acquisition layer also includes the following components: Sensor calibration module: This module performs calibration operations regularly, with a period of 24 hours. It sends calibration signals to each sensor; during the calibration process, it uses a reference resistor or capacitor network to verify the zero offset and gain error of the sensor; if the error exceeds 5%, the system will issue a calibration alarm; Data cache unit: This unit uses dual-port RAM to temporarily store real-time sampled data; the design of this unit supports parallel data reading in the control operation layer, and the update frequency of the cache is consistent with the sampling frequency of the analog-to-digital converter.

8. The operating control system of a wind power generation system according to claim 1, characterized in that, Configure a hardware protection circuit between the operation control layer and the execution drive layer: Overcurrent protection: When the stator current exceeds 1.5 times the rated current, the IGBT drive signal is immediately blocked through a hardware comparator, and the response time is less than 1 microsecond; Undervoltage protection: When the DC bus voltage is lower than 80% of the rated value, the system automatically switches to the asynchronous mode, and reactive power is injected into the power grid through the rotor-side converter to stabilize the voltage.

Images