CPU control method and CPU control board suitable for a variable pitch system

Abstract

The embodiment of the application provides a CPU control method and a CPU control board suitable for a variable pitch system. The CPU control method adopts benchmark PID parameters during normal operation, continuously monitors control performance deviation, automatically triggers online system identification to obtain a current system model when the deviation exceeds a threshold, recompute PID parameters based on the new model by using a pole placement method, and finally updates smoothly under safety constraints to realize closed-loop self-correction of control parameters. In this way, the variable pitch system can adapt to changes in characteristics caused by wear, icing or changes in working conditions in real time without significantly increasing the CPU's regular computing burden, always maintaining optimal control performance, thereby effectively improving power generation efficiency, reducing mechanical load and enhancing the operation reliability of the wind turbine throughout its life cycle.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology, specifically to a CPU control method and CPU control board suitable for pitch control systems. Background Technology

[0002] The pitch control system of a wind turbine generator is one of its core functional units, undertaking the critical task of adjusting the blade pitch angle. By precisely controlling the pitch angle, the aerodynamic forces absorbed by the rotor can be effectively adjusted, thereby optimizing energy capture, maintaining constant power output, and protecting the generator through feathering in high wind speeds or emergencies, while reducing mechanical loads on critical components. Given the complexity of pitch control logic and the high demands for real-time response, the use of a central processing unit (CPU)-based control system to achieve precise and rapid pitch angle adjustment has become an industry standard. Constructing a stable and efficient CPU control scheme is crucial for ensuring the performance of the wind turbine throughout its entire lifecycle.

[0003] However, during the long-term operation of wind turbines, traditional pitch control CPU schemes have gradually revealed their inherent limitations. Currently, pitch systems generally employ a PID (proportional-integral-derivative) control strategy based on fixed parameters. These PID parameters are typically tuned for an idealized system model during the wind turbine design or commissioning phase. However, the physical characteristics of a pitch system are not static. With accumulated operating time, wear and tear on transmission mechanisms such as bearings and gearboxes leads to changes in frictional characteristics; under specific climatic conditions, blade icing significantly alters their mass and moment of inertia; and electronic components such as motor drives experience performance degradation. These factors collectively cause the actual system model parameters to drift compared to the initial model. Furthermore, the system's dynamic response is heavily dependent on external operating conditions; for example, the aerodynamic characteristics differ greatly between high and low wind speeds. In such cases, a fixed set of PID parameters cannot maintain optimal control performance under all operating conditions. This decline in control performance not only affects power generation efficiency but also exacerbates mechanical shock and fatigue loads throughout the transmission chain, posing a challenge to the long-term reliable operation of the wind turbine.

[0004] Therefore, an optimized CPU control method suitable for pitch systems is needed. Summary of the Invention

[0005] The present invention aims to at least solve one of the technical problems existing in the prior art, and provides a CPU control method and CPU control board suitable for pitch systems.

[0006] In a first aspect, embodiments of the present invention provide a CPU control method suitable for a pitch system, comprising the following steps: Obtain the pitch angle setpoint and the actual pitch angle; Based on the baseline PID parameters, baseline PID control and performance deviation monitoring are performed on the pitch angle setpoint and the actual pitch angle to obtain the performance deviation and the data pair of pitch angle setpoint and actual pitch angle for the most recent expected time period. A retuning trigger decision is made based on performance deviation to obtain the retuning trigger flag; In response to the retuning trigger flag being true, online system identification is performed based on the data of the pitch angle setpoint and the actual pitch angle for the most recent expected time period to obtain an updated system model; The pole placement method is used to calculate new PID parameters for the updated system model to obtain candidate PID parameters. Based on the current PID parameters, the candidate PID parameters are smoothly updated under safety constraints to obtain new active PID parameters.

[0007] In some possible embodiments, baseline PID control and performance deviation monitoring are performed on the pitch angle setpoint and actual pitch angle based on baseline PID parameters to obtain performance deviation and data pairs of pitch angle setpoint and actual pitch angle for the most recent expected time period, including: Using reference PID parameters, the setpoint and actual pitch angle of the pitch angle are subjected to reference PID control to obtain the motor control signal; Input the pitch angle setting value into the optimal response model to obtain the reference pitch angle; The absolute value of the difference between the reference pitch angle and the actual pitch angle is calculated as the performance deviation.

[0008] In some possible embodiments, a retuning trigger decision is made based on the performance deviation to obtain a retuning trigger flag, including: setting the retuning trigger flag to true in response to the performance deviation exceeding a preset threshold for a continuous preset time period.

[0009] In some possible embodiments, the preset time period is 500ms.

[0010] In some possible embodiments, in response to the retuning trigger flag being true, online system identification is performed based on the data pair of pitch angle setpoint and actual pitch angle for the most recent expected time period to obtain an updated system model, including: using a recursive least squares algorithm with a forgetting factor to process the data pair of pitch angle setpoint and actual pitch angle for the most recent expected time period to estimate the parameters of the second-order discrete-time system model to obtain the updated system model.

[0011] In some possible embodiments, the pole placement method is used to calculate new PID parameters for the updated system model to obtain candidate PID parameters, including: Based on the updated system model, the desired dominant pole coefficients, and the CPU computation cycle of the control system, mathematical polynomials are defined for the system, the controller, and the desired response. Based on the mathematical polynomials of the system, controller, and desired response, the Diophantine equation is constructed and solved to obtain the coefficients of the PI controller polynomial; The coefficients of the PI controller polynomial are converted into the candidate PID parameters.

[0012] In some possible embodiments, converting the coefficients of the PI controller polynomial into the candidate PID parameters includes: converting the coefficients of the PI controller polynomial into the candidate PID parameters using the following formula:

[0013]

[0014] in, The first coefficient in the PI controller polynomial. The second coefficient in the PI controller polynomial. This refers to the CPU calculation cycle of the control system.

[0015] In some possible embodiments, the candidate PID parameters are smoothly updated under safety constraints based on the current PID parameters to obtain new active PID parameters, including: Check whether the candidate PID parameters are within the predefined safety polygon to obtain the check results; In response to the inspection results meeting the requirements, the candidate PID parameters are smoothly updated based on the current PID parameters to obtain new active PID parameters.

[0016] In some possible embodiments, in response to a check result that meets the requirements, a smooth update is performed on the candidate PID parameters based on the current PID parameters to obtain new active PID parameters. This includes: in response to a check result that meets the requirements, a smooth update is performed on the candidate PID parameters based on the current PID parameters using the following formula to obtain new active PID parameters, wherein the formula is:

[0017] in, Candidate PID parameters, As a smoothing factor, These are the current PID parameters.

[0018] Secondly, embodiments of the present invention provide a CPU control board suitable for a pitch system, comprising: The data acquisition module is used to acquire the pitch angle setpoint and the actual pitch angle; The pitch angle deviation monitoring module is used to perform benchmark PID control and performance deviation monitoring on the pitch angle setpoint and actual pitch angle based on the benchmark PID parameters to obtain the performance deviation and the data pair of the pitch angle setpoint and actual pitch angle for the most recent expected time period. The retuning trigger decision module is used to make retuning trigger decisions based on performance deviations to obtain a retuning trigger flag; The online system identification module is used to perform online system identification based on the data of the pitch angle setpoint and the actual pitch angle during the most recent expected time period in response to the retuning trigger flag being true, so as to obtain an updated system model. The new PID parameter calculation module is used to calculate new PID parameters for the updated system model using the pole placement method to obtain candidate PID parameters. The parameter smoothing update module is used to perform a safe constraint-based smoothing update of the candidate PID parameters based on the current PID parameters to obtain new active PID parameters.

[0019] Compared with existing technologies, this invention provides a CPU control method and CPU control board suitable for pitch control systems. During normal operation, it uses baseline PID parameters and continuously monitors control performance deviations. When the deviation exceeds a threshold, it automatically triggers online system identification to obtain the current system model. Based on the new model, it recalculates the PID parameters using the pole placement method and finally updates them smoothly under safety constraints, achieving closed-loop self-correction of control parameters. In this way, without significantly increasing the CPU's conventional computational burden, the pitch control system can adapt in real time to changes in characteristics caused by wear, icing, or changes in operating conditions, always maintaining optimal control performance. This effectively improves power generation efficiency, reduces mechanical load, and enhances the operational reliability of the wind turbine throughout its entire lifecycle. Attached Figure Description

[0020] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0021] Figure 1 This is a flowchart of a CPU control method for a pitch system according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the data flow of a CPU control method for a pitch system according to an embodiment of the present invention; Figure 3 This is a block diagram of a CPU control board for a pitch system according to an embodiment of the present invention. Detailed Implementation

[0022] To enable those skilled in the art to better understand the technical solutions of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0023] Unless otherwise specifically stated, the technical or scientific terms used in the embodiments of this invention should be understood in their ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains. The terms "comprising" or "including," as used in the embodiments of this invention, do not limit the shapes, numbers, steps, actions, operations, components, elements, and / or groups thereof mentioned, nor do they exclude the appearance or addition of one or more other different shapes, numbers, steps, actions, operations, components, elements, and / or groups thereof, or the inclusion of these.

[0024] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention. It should also be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale, and techniques, methods, and apparatus known to those skilled in the art may not be discussed in detail; however, where appropriate, the illustrated techniques, methods, and apparatus should be considered part of the specification. In all the examples shown and discussed herein, any other specific example may have different values. It should be noted that similar symbols and letters in the following figures denote similar items; therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0025] In the description of the embodiments of the present invention, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In the embodiments of the present invention, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in the embodiments of the present invention, as well as the features of different embodiments or examples.

[0026] Hereinafter, exemplary embodiments according to the present invention will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of the present invention, and not all embodiments of the present invention. It should be understood that the present invention is not limited to the exemplary embodiments described herein.

[0027] In the technical solution of this invention, a CPU control method suitable for pitch control systems is proposed. Figure 1 This is a flowchart of a CPU control method for a pitch system according to an embodiment of the present invention. Figure 2 This is a system architecture diagram of a CPU control method for a pitch system according to an embodiment of the present invention. Figure 1 and Figure 2 As shown, the CPU control method for a pitch system according to an embodiment of the present invention includes the following steps: S1, acquiring the pitch angle setpoint and the actual pitch angle; S2, performing benchmark PID control and performance deviation monitoring on the pitch angle setpoint and the actual pitch angle based on benchmark PID parameters to obtain a data pair of performance deviation and pitch angle setpoint-actual pitch angle for the most recent expected time period; S3, making a retuning trigger decision based on the performance deviation to obtain a retuning trigger flag; S4, in response to the retuning trigger flag being true, performing online system identification based on the data pair of pitch angle setpoint-actual pitch angle for the most recent expected time period to obtain an updated system model; S5, using the pole placement method to calculate new PID parameters for the updated system model to obtain candidate PID parameters; S6, performing parameter smoothing update of the candidate PID parameters under safety constraints based on the current PID parameters to obtain new active PID parameters.

[0028] Specifically, in S1, the setpoint and actual pitch angle are acquired. Here, the setpoint represents the system's control objective, while the actual pitch angle represents the system's current state feedback. Acquiring these two variables constitutes the input to the entire feedback control loop. Only by accurately and synchronously acquiring these two key variables can the control deviation be calculated, the controller adjusted, and the system performance accurately evaluated, thereby determining whether an adaptive parameter adjustment mechanism needs to be activated to cope with changes in system characteristics.

[0029] In practice, the first step is to acquire the pitch angle setpoint. This setpoint is a command signal calculated by the wind turbine's main controller based on current operating conditions (such as wind speed, generator speed, and power output) and control strategies (such as maximum wind energy tracking, constant power control, or safe shutdown). The data acquisition module receives the command message from the main controller via a standard industrial communication bus (such as CAN bus or industrial Ethernet) and parses the pitch angle setpoint for that moment. The second step is to acquire the actual pitch angle. This data comes from a high-precision angular position sensor (such as an absolute encoder or resolver) installed on the pitch actuator (such as the blade root or pitch bearing). The data acquisition module collects and decodes the sensor's output signal at a sampling frequency synchronized with the CPU control cycle, thereby obtaining a high-resolution actual pitch angle value that accurately reflects the current physical angle of the blade. It is worth noting that these two acquisition actions are strictly synchronized within each control cycle to ensure the consistency of the setpoint and actual value in time, providing a valid data pair for subsequent calculations.

[0030] Specifically, S2 involves performing benchmark PID control and performance deviation monitoring on the pitch angle setpoint and actual pitch angle based on benchmark PID parameters to obtain a data pair of performance deviation and the pitch angle setpoint minus the actual pitch angle for the most recent expected time period. It should be understood that during the operation of the pitch system, factors such as mechanical wear and environmental changes can cause the system's dynamic characteristics to drift, requiring continuous monitoring of control performance and recording of operating data. In the technical solution of this invention, benchmark PID control maintains the basic operation of the system, while performance deviation monitoring evaluates whether the current control effect meets the target. When the performance continuously deviates from expectations, a subsequent adaptive adjustment process is triggered.

[0031] In practice, firstly, reference PID parameters are used to perform reference PID control on the setpoint and actual pitch angle to obtain the motor control signal. The reference PID parameters are a fixed set of PID gain values ​​tuned during system design and debugging, used for normal system operation. During this process, the error between the two is first calculated: Then, based on a pre-set set of benchmark PID parameters that are considered optimal under ideal system conditions... The motor control signal for driving the pitch actuator is calculated using a discrete PID control algorithm. The calculation process follows the standard PID control law.

[0032] Next, the pitch angle setpoint is input into the optimal response model to obtain the reference pitch angle. Here, the optimal response model is a virtual mathematical reference model used for simulation. Its parameters are carefully selected during the design phase to represent the most desired system dynamic performance, such as a fast, smooth, and overshoot-free response. The output reference pitch angle is the ideal, smooth response trajectory. It should be understood that simply comparing the actual value and the setpoint ignores the dynamic process of the system. By introducing the optimal response model, we can know how a perfect system should respond under a given step or change in the setpoint, thus providing an accurate, time-varying reference trajectory for performance evaluation. In this process, a mathematical model is predefined and stored in the software of the CPU control board. This model is typically a second-order linear system transfer function with the desired damping ratio and natural frequency, representing the most ideal dynamic response characteristics of the pitch system. In each control cycle, the program uses the latest pitch angle setpoint as the input to this model and calculates the output of the model in real time, i.e., the reference pitch angle.

[0033] Subsequently, the absolute value of the difference between the reference pitch angle and the actual pitch angle is calculated as the performance deviation. That is, the gap between the actual system behavior and the ideal behavior is transformed into a specific, measurable value. Here, the performance deviation value is the absolute error between the actual system output and the ideal model output at the same moment; it is a non-negative scalar and a core indicator for measuring the dynamic tracking performance of the system. In this process, during each control cycle, the CPU obtains the actual pitch angle from the sensor and the reference pitch angle from the optimal response model, and calculates the performance deviation.

[0034] It is worth noting that although performance deviation is used as the basis for triggering PID parameter retuning in the technical solution of this invention, similar deviation signals can also be used as inputs to fault diagnosis modules in a broader range of intelligent control and operation and maintenance systems. In such applications, the fundamental reason for performing confidence assessment based on a fault signature library is the need to effectively distinguish between two distinct system abnormal states: one is the slow drift of system model parameters due to factors such as wear and aging, and this performance degradation should be corrected through adaptive parameter adjustment in this solution; the other is a clear and sudden failure of key components such as sensors and actuators. Confidence assessment based on a fault signature library, by matching real-time deviation characteristics with pre-established signatures associated with specific fault modes, can provide a confidence score, determining the likelihood that the current performance deviation is caused by component failure. This avoids the control system mistakenly attempting to fix the problem by adjusting PID parameters when the equipment has already suffered physical damage, thereby ensuring the correctness of the control strategy and the safety of the equipment.

[0035] Specifically, S3 involves making a retuning trigger decision based on performance deviation to obtain a retuning trigger flag. It should be understood that traditional PID control systems often use fixed parameters or manual periodic adjustments, which are insufficient to address performance degradation issues caused by mechanical wear, environmental changes, and other factors during long-term operation of pitch systems. By monitoring performance deviation in real time and formulating scientific triggering decisions, parameter retuning can be automatically triggered when the control system performance degrades to a certain level. This avoids the computational burden of unnecessary frequent adjustments and promptly corrects control deviations, maintaining the system's optimal operating state. By establishing an intelligent parameter adjustment triggering mechanism, adaptive PID parameter optimization can be achieved while ensuring system stability. This triggering mechanism based on continuous performance monitoring is a key element in realizing closed-loop adaptive control.

[0036] The retuning trigger decision refers to a complete logical judgment process that determines whether to initiate PID parameter retuning based on the time-series characteristics of performance deviation. The retuning trigger flag is a binary software signal (such as a Boolean variable or a flag bit in a register), and its state (true / false) directly controls whether the subsequent adaptive process is initiated.

[0037] In specific implementation, in response to a performance deviation exceeding a preset threshold for a continuous preset time period, the retuning trigger flag is set to true. The specific execution process can be broken down as follows: First, in each control cycle k, the module will input the performance deviation... Compared to a pre-set threshold that serves as a dividing line between good and bad performance. First, a comparison is made; second, the module maintains a timer or a continuous period counter internally to track performance deviations. consecutive greater than Time. If If the timer increments, then the timer will accumulate; otherwise, if If the timer fails to reach zero, the timer is immediately reset. Finally, the module combines the duration recorded by the timer with a preset time period. Comparison is performed only when the time recorded by the timer reaches or exceeds the specified time. Only when the specified conditions are met will the module set an internal, Boolean-type retuning trigger flag to true. Once this flag is set to True, it will remain in this state until the subsequent parameter update process is completed and the system resets it, thus ensuring that the subsequent online identification and parameter calculation steps are fully executed. In all other cases (i.e., the deviation does not exceed the threshold or the duration is insufficient), the flag remains False, and the system continues to use the baseline PID parameters for control. It is worth mentioning that the preset time period is a predefined time length used to integrate or filter the performance deviation over time. In a specific example of this application, the preset time period is 500ms.

[0038] Specifically, in step S4, in response to the retuning trigger flag being true, online system identification is performed based on the data of the pitch angle setpoint and actual pitch angle over the most recent expected time period to obtain an updated system model. It should be understood that when the system confirms a sustained decline in control performance, it is essential to accurately obtain the current actual dynamic characteristics of the pitch system. However, due to wear, icing, or changes in operating conditions, the original system mathematical model is no longer accurate, and the PID parameters designed based on the old model become invalid. Therefore, in the technical solution of this invention, to design new PID parameters that match the current system characteristics, a data-driven method is used to establish an updated system model that accurately describes the relationship between the current pitch angle setpoint and actual pitch angle. Specifically, data generated during actual system operation is used to create the latest digital profile of the system, providing an accurate and up-to-date model basis for subsequent controller parameter optimization design.

[0039] In specific implementation, a recursive least squares algorithm with a forgetting factor is used to process the data pairs of pitch angle setpoints and actual pitch angles based on the most recent expected time period to estimate the parameters of the second-order discrete-time system model and obtain the updated system model. Specifically, firstly, the module retrieves the stored data pairs of pitch angle setpoints and actual pitch angles for the most recent expected time period from the data buffer. This data pair constitutes the input sequence for system identification. (Pitch angle setting) and output sequence (Actual pitch angle); secondly, the parameters of a pre-defined second-order discrete-time system model are estimated using the recursive least squares (RLS) algorithm with a forgetting factor. This model is typically in the form of autoregressive moving average (ARMA) or autoregressive with external input (ARX).

[0040] Taking the scheme of this invention as an example, when the retuning trigger flag becomes true, the online system identification module is activated. This module first retrieves data pairs of the pitch angle setpoint and actual pitch angle from the buffer over the past 500ms (i.e., 50). Algorithm initialization is then performed, setting the initial parameter estimates. (Can be a zero vector or parameters based on the baseline model), a large initial covariance matrix. (like ,in (as an identity matrix), and set a forgetting factor. The value is 0.98. Then, the algorithm iterates through each of the 50 data pairs, starting with the first one. For each data pair... Each module will execute the four RLS formulas mentioned above, calculating them sequentially. , , and After processing all 50 data pairs, the final parameter vector is obtained. ,Right now The latest estimates are used to obtain an updated system model, which will be passed to the next step to calculate new PID parameters.

[0041] Specifically, in step S5, the pole placement method is used to calculate new PID parameters for the updated system model to obtain candidate PID parameters. That is, using the updated system model obtained from the previous online identification step, which accurately reflects the current system's dynamic characteristics, the pole placement method is used to calculate new PID parameters for the updated system model to obtain candidate PID parameters. Unlike traditional trial-and-error methods, the pole placement method is a modern control theory design method that can directly specify the desired dynamic performance (such as response speed, stability, overshoot, etc.) of the closed-loop control system and reverse-engineer the controller parameters that can achieve this performance.

[0042] In practical implementation, firstly, based on the updated system model, the desired dominant pole coefficients, and the CPU computation cycle of the control system, mathematical polynomials for the system, controller, and desired response are defined. Secondly, according to the PI controller structure adopted in this invention, the controller polynomial is defined. Finally, based on the pre-defined expected dominant pole coefficients, a characteristic polynomial representing the response of the ideal closed-loop system is defined. These expected dominant poles determine the dynamic characteristics of the closed-loop system response (such as damping ratio and natural frequency).

[0043] Next, based on the mathematical polynomials of the system, controller, and desired response, the Diophantine equation is constructed and solved to obtain the coefficients of the PI controller polynomial.

[0044] Further, the coefficients of the PI controller polynomial are transformed into the candidate PID parameters. In this process, the coefficients of the PI controller polynomial are transformed into the candidate PID parameters using the following formula:

[0045]

[0046] in, The first coefficient in the PI controller polynomial. The second coefficient in the PI controller polynomial. This refers to the CPU calculation cycle of the control system.

[0047] Specifically, in step S6, the candidate PID parameters are smoothly updated under safety constraints based on the current PID parameters to obtain new active PID parameters. It should be understood that even if the candidate PID parameters calculated in the preceding steps are theoretically optimal, directly using them may pose two major risks. First, there is a safety risk: theoretical calculations may produce a set of parameters that cause system instability due to small errors in model identification or unmodeled dynamics. Second, there is a stability risk: excessive differences between the old and new parameters may cause drastic jumps in the controller output, impacting the mechanical structure of the pitch system and triggering system oscillations. Therefore, in the technical solution of this invention, by introducing two mechanisms—safety constraints and smooth updates—the aim is to ensure that every adjustment of parameters is performed within a known safety boundary, and that the process is gradual and disturbance-free, ultimately achieving seamless switching and improvement of control performance.

[0048] In practice, the first step is to check whether the candidate PID parameters are within a predefined safety polygon to obtain the check result. This is a preliminary safety review step. In software implementation, the vertex coordinates of a safety polygon are pre-stored in the controller's memory. This polygon is defined in the PID parameter space (e.g., using...). The horizontal axis is... In a two-dimensional plane (with the vertical axis as the boundary), its boundary delineates a known parameter region that guarantees the stable operation of the closed-loop system. This region is predetermined through offline simulation, theoretical analysis, or experimental testing. When the module receives the candidate PID parameters calculated in the previous stage (e.g., a...),... After the point is matched, an algorithm (such as ray casting or corner casting) is executed to determine whether the parameter point falls within the safe zone. A Boolean result is output: if the parameter point is within the polygon, the requirement is met; otherwise, it is not. If the result is unsatisfactory, the entire update process is terminated, and the system continues to use the current PID parameters for control, thus avoiding any potential instability risks.

[0049] Subsequently, in response to the check result meeting the requirements, the candidate PID parameters are smoothly updated based on the current PID parameters to obtain the new active PID parameters. That is, after the safety check passes, a gradual update strategy is further adopted to update the candidate PID parameters to obtain the new active PID parameters. Specifically, in response to the check result meeting the requirements, the candidate PID parameters are smoothly updated based on the current PID parameters using the following formula to obtain the new active PID parameters:

[0050] in, Candidate PID parameters, As a smoothing factor, The current PID parameters are used. Through this calculation method, the new PID parameters are actually an interpolation between the current parameters and candidate parameters, thus ensuring the continuity and gradualness of parameter changes and achieving seamless switching.

[0051] Taking the solution of this invention as an example, assume that the candidate PID parameters received by the module are: The parameters currently being used by the system are... First, the module performs a safety check, comparing the point (0.5, 200) with a predefined safe polygon. If the point is within the specified area, the check is considered satisfactory. Next, the module performs a smooth update, setting the smoothing factor α to 0.1, and calculates the new... and :

[0052]

[0053] Therefore, the new active PID parameters are obtained as follows: In the next control cycle, the pitch CPU control board will use these new parameters for PID control. Simultaneously, the entire adaptive process completes a closed loop, and the system will reset the retuning trigger flag to false, returning to normal baseline PID control and performance monitoring.

[0054] In summary, the CPU control method for pitch systems according to embodiments of the present invention is explained. It employs baseline PID parameters during normal operation and continuously monitors control performance deviations. When the deviation exceeds a threshold, online system identification is automatically triggered to obtain the current system model. Based on the new model, the PID parameters are recalculated using the pole placement method. Finally, the parameters are smoothly updated under safety constraints, achieving closed-loop self-correction of the control parameters. In this way, without significantly increasing the CPU's conventional computational burden, the pitch system can adapt in real time to changes in characteristics caused by wear, icing, or changes in operating conditions, always maintaining optimal control performance. This effectively improves power generation efficiency, reduces mechanical load, and enhances the operational reliability of the wind turbine throughout its entire lifecycle.

[0055] Furthermore, a CPU control board suitable for pitch systems is also provided.

[0056] Figure 3 This is a block diagram of a CPU control board suitable for a pitch system according to an embodiment of the present invention. Figure 3As shown, a CPU control board 300 for a pitch system according to an embodiment of the present invention includes: a data acquisition module 310, used to acquire a pitch angle setpoint and an actual pitch angle; a pitch angle deviation monitoring module 320, used to perform benchmark PID control and performance deviation monitoring on the pitch angle setpoint and actual pitch angle based on benchmark PID parameters to obtain a data pair of performance deviation and pitch angle setpoint-actual pitch angle for the most recent expected time period; a retuning trigger decision module 330, used to make a retuning trigger decision based on the performance deviation to obtain a retuning trigger flag; an online system identification module 340, used to perform online system identification based on the data pair of pitch angle setpoint-actual pitch angle for the most recent expected time period in response to the retuning trigger flag being true, to obtain an updated system model; a new PID parameter calculation module 350, used to calculate new PID parameters for the updated system model using the pole placement method to obtain candidate PID parameters; and a parameter smoothing update module 360, used to perform parameter smoothing update of the candidate PID parameters under safety constraints based on the current PID parameters to obtain new active PID parameters.

[0057] As described above, the CPU control board 300 for a pitch system according to embodiments of the present invention can be implemented in various wireless terminals, such as servers with CPU control algorithms suitable for pitch systems. In one possible implementation, the CPU control board 300 for a pitch system according to embodiments of the present invention can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the CPU control board 300 for a pitch system can be a software module in the operating system of the wireless terminal, or it can be an application developed for the wireless terminal; of course, the CPU control board 300 for a pitch system can also be one of many hardware modules of the wireless terminal.

[0058] Alternatively, in another example, the CPU control board 300 for the pitch system and the wireless terminal can also be separate devices, and the CPU control board 300 for the pitch system can be connected to the wireless terminal via wired and / or wireless networks, and transmit interactive information in accordance with an agreed data format.

[0059] It is understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present invention, and the present invention is not limited thereto. For those skilled in the art, various modifications and improvements can be made without departing from the spirit and essence of the present invention, and these modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. A CPU control method suitable for pitch control systems, characterized in that, Including the following steps: Obtain the pitch angle setpoint and the actual pitch angle; Based on the baseline PID parameters, baseline PID control and performance deviation monitoring are performed on the pitch angle setpoint and the actual pitch angle to obtain the performance deviation and the data pair of pitch angle setpoint and actual pitch angle for the most recent expected time period. A retuning trigger decision is made based on performance deviation to obtain the retuning trigger flag; In response to the retuning trigger flag being true, online system identification is performed based on the data of the pitch angle setpoint and the actual pitch angle for the most recent expected time period to obtain an updated system model; The pole placement method is used to calculate new PID parameters for the updated system model to obtain candidate PID parameters. Based on the current PID parameters, the candidate PID parameters are smoothly updated under safety constraints to obtain new active PID parameters.

2. The CPU control method for a pitch system according to claim 1, characterized in that, Based on the baseline PID parameters, baseline PID control and performance deviation monitoring are performed on the pitch angle setpoint and actual pitch angle to obtain performance deviation and data pairs of pitch angle setpoint - actual pitch angle for the most recent expected time period, including: Using reference PID parameters, the setpoint and actual pitch angle of the pitch angle are subjected to reference PID control to obtain the motor control signal; Input the pitch angle setting value into the optimal response model to obtain the reference pitch angle; The absolute value of the difference between the reference pitch angle and the actual pitch angle is calculated as the performance deviation.

3. The CPU control method for a pitch system according to claim 1, characterized in that, The retuning trigger decision based on performance deviation to obtain a retuning trigger flag includes: setting the retuning trigger flag to true in response to the performance deviation exceeding a preset threshold for a continuous preset time period.

4. The CPU control method for a pitch system according to claim 1, characterized in that, The preset time period is 500ms.

5. The CPU control method for a pitch system according to claim 1, characterized in that, In response to the retuning trigger flag being true, online system identification is performed based on the data pair of pitch angle setpoint and actual pitch angle for the most recent expected time period to obtain an updated system model. This includes: using a recursive least squares algorithm with a forgetting factor to process the data pair of pitch angle setpoint and actual pitch angle for the most recent expected time period to estimate the parameters of the second-order discrete-time system model to obtain the updated system model.

6. The CPU control method for a pitch system according to claim 1, characterized in that, The pole placement method is used to calculate new PID parameters for the updated system model to obtain candidate PID parameters, including: Based on the updated system model, the desired dominant pole coefficients, and the CPU computation cycle of the control system, mathematical polynomials are defined for the system, the controller, and the desired response. Based on the mathematical polynomials of the system, controller, and desired response, the Diophantine equation is constructed and solved to obtain the coefficients of the PI controller polynomial; The coefficients of the PI controller polynomial are converted into the candidate PID parameters.

7. The CPU control method for a pitch system according to claim 6, characterized in that, Converting the coefficients of the PI controller polynomial into the candidate PID parameters includes: converting the coefficients of the PI controller polynomial into the candidate PID parameters using the following formula: in, The first coefficient in the PI controller polynomial. The second coefficient in the PI controller polynomial. This refers to the CPU calculation cycle of the control system.

8. The CPU control method for a pitch system according to claim 1, characterized in that, Based on the current PID parameters, the candidate PID parameters are smoothly updated under safety constraints to obtain new active PID parameters, including: Check whether the candidate PID parameters are within the predefined safety polygon to obtain the check results; In response to the inspection results meeting the requirements, the candidate PID parameters are smoothly updated based on the current PID parameters to obtain new active PID parameters.

9. The CPU control method for a pitch system according to claim 8, characterized in that, In response to the check result meeting the requirements, the candidate PID parameters are smoothly updated based on the current PID parameters to obtain new active PID parameters. This includes: in response to the check result meeting the requirements, the candidate PID parameters are smoothly updated based on the current PID parameters using the following formula to obtain new active PID parameters, wherein the formula is: in, Candidate PID parameters, As a smoothing factor, These are the current PID parameters.

10. A CPU control board suitable for a pitch system, characterized in that, include: The data acquisition module is used to acquire the pitch angle setpoint and the actual pitch angle; The pitch angle deviation monitoring module is used to perform benchmark PID control and performance deviation monitoring on the pitch angle setpoint and actual pitch angle based on the benchmark PID parameters to obtain the performance deviation and the data pair of the pitch angle setpoint and actual pitch angle for the most recent expected time period. The retuning trigger decision module is used to make retuning trigger decisions based on performance deviations to obtain a retuning trigger flag; The online system identification module is used to perform online system identification based on the data of the pitch angle setpoint and the actual pitch angle during the most recent expected time period in response to the retuning trigger flag being true, so as to obtain an updated system model. The new PID parameter calculation module is used to calculate new PID parameters for the updated system model using the pole placement method to obtain candidate PID parameters. The parameter smoothing update module is used to perform a safe constraint-based smoothing update of the candidate PID parameters based on the current PID parameters to obtain new active PID parameters.

Images