A Multi-Objective Load Reduction Method and Related Devices for Wind Turbine Blades Based on Nonlinear Model Predictive Control
By constructing a multi-objective load reduction method for wind turbine blades based on nonlinear model predictive control, the problem of blade fatigue load was solved. This method actively suppressed the out-of-plane bending moment at the blade root and mitigated fatigue damage, thereby extending the service life of the blades and improving the operational safety and economy of the wind turbine.
Patent Information
- Application Number
- CN202611324346.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-28
- Publication Date
- 2026-11-17
AI Technical Summary
The fatigue load problem of existing wind turbine blades has not been effectively suppressed, especially under high turbulent wind conditions. Existing control methods lack coupling processing of blade flexible dynamics and aerodynamics, which leads to increased load fluctuations and the controller cannot actively avoid load over-limit.
A nonlinear model predictive control method is adopted to construct a flexible dynamic model with two degrees of freedom, blade flapping and oscillation. A multi-objective optimization function is designed, and the dynamic balance between out-of-plane bending moment at the blade root, fatigue damage and power generation is achieved through adaptive weight adjustment. The maximum allowable out-of-plane bending moment of the blade is incorporated into the optimization framework as a soft constraint.
While minimizing power generation loss, the out-of-plane bending moment fluctuation at the blade root is actively suppressed to reduce fatigue damage, extend blade service life, and improve operational safety and economy.
Smart Images

Figure FT_1 
Figure FT_2 
Figure SMS_1
Abstract
Description
Technical Field
[0001] This invention belongs to the field of wind turbine control technology, specifically relating to a multi-objective load reduction method and related devices for wind turbine blades based on nonlinear model predictive control. Background Technology
[0002] With the increasing trend towards larger scale in large wind turbines, the continuous increase in blade length makes the fatigue load problem increasingly prominent. The periodic fluctuation of the out-of-plane bending moment at the blade root mainly originates from the combined effects of wind shear, tower shadow effect, turbulence, and yaw error. These cyclic loads accelerate fatigue damage to blade composite materials and shorten the service life of the unit. Therefore, how to effectively suppress blade fatigue load while ensuring power output has become a core technical problem that urgently needs to be solved in this field.
[0003] Existing blade load reduction control methods mainly include traditional PID control, linear model predictive control, and multi-model switching control based on piecewise linearization. Traditional PID pitch controllers are simple in structure, using generator speed deviation as input and outputting pitch angle commands through a single feedback channel. However, the parameters of the PID controller are typically tuned only around the rated operating point. When significant wind speed changes cause the operating point to shift, the fixed gain cannot guarantee optimal load suppression. Especially under high turbulence conditions, PID control is significantly insufficient in suppressing blade root bending moment fluctuations. The fundamental reason is that this control framework lacks feedforward compensation capabilities for aerodynamic loads and cannot explicitly handle physical constraints such as pitch rate and pitch angle limits. This leads to control command saturation under large disturbance conditions, which in turn exacerbates load fluctuations.
[0004] Linear model predictive control (LMDC) has been introduced into the field of wind turbine control in recent years. Its basic principle is to linearize the nonlinear wind turbine model at a certain operating point, establish a linear predictive model, and then solve a quadratic programming problem in a finite time domain online to obtain the optimal control sequence. Some existing technologies employ a piecewise linearization strategy, dividing the entire wind speed range into multiple sub-intervals, each corresponding to a linearized model and a linear MPC controller, achieving full operating condition coverage through a soft switching mechanism. While this approach can improve the performance of PID controllers to some extent, its inherent drawback is that linearization is essentially an approximation of a nonlinear system. When the actual operating point deviates significantly from the linearization point, the accuracy of the predictive model decreases significantly, causing the actual performance of the controller to deviate from the design expectations. Furthermore, the piecewise linearization strategy requires pre-determining multiple operating points and corresponding controller parameters in the offline stage. However, there is an inherent contradiction between the number of operating points and the coverage range—too few divisions result in excessive linearization error accumulation, while too many divisions lead to complex switching logic and control jitter issues during switching transients. More importantly, the objective function of linear MPC usually only considers power tracking error and pitch action amplitude, lacking direct quantification and optimization of blade fatigue load. The load reduction effect is achieved indirectly rather than actively pursued.
[0005] Another similar existing technology is based on nonlinear model predictive control, which directly uses the nonlinear wind turbine dynamics model as the prediction model, theoretically yielding more accurate prediction results. However, this type of approach faces two prominent problems in practical applications: First, most existing nonlinear MPC schemes employ a single objective function, prioritizing power generation maximization, with blade load serving only as a constraint or secondary penalty. They lack a dynamic balance mechanism among multiple objectives, often sacrificing load to maintain power output in high wind speed regions above rated wind speed. Second, nonlinear MPC has a heavy computational burden. Existing schemes typically simplify the flexible dynamics model of the blades, treating them as rigid bodies and neglecting the coupling effect of elastic deformation on aerodynamic loads in the flapping and swaying degrees of freedom. This results in insufficient accuracy in load prediction under heavy turbulence conditions.
[0006] The common defects of the aforementioned existing technologies can be summarized into three aspects, among which: First, regarding model accuracy, neither the linearized model nor the simplified nonlinear model can simultaneously account for the strong nonlinear coupling characteristics of blade flexible dynamics and aerodynamic forces. Wind shearing and tower shadow effects cause the inflow wind speed acting on different blades to vary periodically with the azimuth angle, while the elastic deformation of the blades changes the effective angle of attack, forming aeroelastic coupling. The rigid blade assumption commonly used in existing technologies interrupts this physical process.
[0007] Second, at the objective function level, the cost function of existing methods fails to directly optimize the out-of-plane bending moment at the blade root. Instead, they often use indirect methods, such as reducing the pitch rate or limiting speed fluctuations, to achieve load reduction. This indirect optimization method cannot guarantee the optimality of load suppression.
[0008] Third, at the constraint handling level, although existing linear MPCs can explicitly handle pitch rate and pitch angle limit constraints, blade load constraints (such as maximum allowable out-of-plane bending moment) are difficult to incorporate into the linear framework due to their nonlinear characteristics, which makes it impossible for the controller to actively avoid load over-limit situations.
[0009] The common root cause of the above defects lies in the lack of a unified control framework in existing technologies that can simultaneously handle strong nonlinear dynamics, multi-objective optimization, and complex constraints. Summary of the Invention
[0010] The purpose of this invention is to propose a multi-objective load reduction method and related device for wind turbine blades based on nonlinear model predictive control, which solves the above-mentioned shortcomings in the prior art.
[0011] To achieve the above objectives, the technical solution adopted by the present invention is as follows: In a first aspect, the present invention provides a multi-objective load reduction method for wind turbine blades based on nonlinear model predictive control, comprising the following steps: Based on the acquired system state information at the current moment, the effective inflow wind speed field of the impeller plane is reconstructed; Based on the obtained effective inflow wind speed field, a nonlinear prediction model is constructed; Using a nonlinear prediction model as a constraint, the constructed multi-objective optimization cost function is solved to obtain the optimal pitch sequence; The blade operation is controlled based on the obtained optimal pitch sequence.
[0012] Preferably, based on the acquired system state information at the current moment, the effective inflow velocity field of the impeller plane is reconstructed. Specifically, the method is as follows: Based on the wind speed profile matrix in the system state information, the wind speed values of multiple grid points in the impeller model plane are reconstructed. Based on the obtained wind speed value and the real-time acquired current rotor azimuth angle, combined with wind shear and tower shadow effects, the effective inflow wind speed of each blade at the current position is calculated. Based on the obtained effective inflow velocity, the effective inflow velocity field of the impeller plane is reconstructed.
[0013] Preferably, the nonlinear prediction model includes a blade aerodynamic load calculation submodule, a blade structural dynamics submodule, and a nacelle-tower dynamics submodule, wherein: The nacelle-tower dynamics submodule is used to obtain the tower top displacement based on the vibration describing the front-to-back direction of the tower. The blade structure dynamics submodule is used to obtain the blade flapping speed; The blade aerodynamic load calculation submodule is used to obtain the total aerodynamic flapping moment of each blade based on the effective inflow wind speed, blade flapping speed and the real-time obtained current pitch angle.
[0014] Preferably, the optimal pitch sequence is obtained by solving the constructed multi-objective optimization cost function under the constraint of a nonlinear prediction model. Specifically, the method is as follows: Using a nonlinear prediction model as the equality constraint; The pitch angle boundary and pitch rate limit are used as inequality constraints. The optimal pitch sequence is obtained by solving the multi-objective optimization cost function based on equality and inequality constraints.
[0015] Preferably, the expression for the multi-objective optimization cost function is:
[0016] in, This is the peak out-of-plane bending moment term at the blade root. For fatigue load terms; This is the power tracking error term; This is the cost term for the pitch control maneuver; To constrain violations and penalties; , , , All are weighting coefficients.
[0017] Secondly, the multi-objective load reduction system for wind turbine blades based on nonlinear model predictive control provided by the present invention includes: The wind speed field reconstruction unit is used to reconstruct the effective inflow wind speed field of the impeller plane based on the acquired system state information at the current moment. The model building unit is used to construct a nonlinear prediction model based on the obtained effective inflow wind speed field; The pitch sequence solving unit is used to solve the constructed multi-objective optimization cost function under the constraint of the nonlinear prediction model to obtain the optimal pitch sequence; The blade control unit is used to control the blade operation based on the obtained optimal pitch sequence.
[0018] Thirdly, the present invention provides an electronic device including a processor and a memory, wherein the memory stores computer instructions, and when the computer instructions are executed by the processor, the electronic device performs the method described thereon.
[0019] Fourthly, the present invention provides a computing device cluster, comprising at least one computing device, each computing device including a processor and a memory; The processor of the at least one computing device is used to execute instructions stored in the memory of the at least one computing device to cause the cluster of computing devices to perform the method.
[0020] Fifthly, the present invention provides a computer program product, the computer program product including computer-executable instructions, which, when executed, implement the method described.
[0021] In a sixth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the method described herein.
[0022] Compared with the prior art, the beneficial effects of the present invention are: This invention provides a multi-objective load reduction method for wind turbine blades based on nonlinear model predictive control. This method systematically improves upon the method from three levels: model construction, objective function design, and optimization solution. At the model construction level, this invention establishes a nonlinear predictive model incorporating two degrees of freedom of flexible dynamics: blade flapping and oscillation. An aerodynamic load calculation module based on blade element momentum theory is used to explicitly incorporate the coupling effect between blade elastic deformation and the inflow wind field into the prediction framework. At the objective function level, a weighted multi-objective optimization function is designed, simultaneously balancing three indicators: the direct minimization of the peak out-of-plane bending moment at the blade root, the indirect representation of the fatigue damage accumulation rate, and power generation loss. An adaptive weight adjustment mechanism achieves a dynamic balance between the "power generation-lifespan" objective under different wind speed conditions. At the constraint handling level, the maximum allowable out-of-plane bending moment of the blade is incorporated as a soft constraint in the nonlinear optimization problem in the form of an inequality. Simultaneously, the rate and position limits of the pitch actuator are explicitly treated as hard constraints, forming a complete constraint system.
[0023] In summary, this invention achieves systematic improvements in four dimensions: model construction accuracy, objective function design, constraint handling capability, and online optimization mechanism. Its final engineering effect is to actively suppress the periodic fluctuation of out-of-plane bending moment at the blade root and slow down the accumulation rate of fatigue damage while ensuring that the power generation loss is minimized. This extends the service life of the blade, reduces the risk of unplanned downtime, and significantly improves the operational safety and life-cycle economy of large wind turbines under all wind speed conditions. Attached Figure Description
[0024] Figure 1 This is a schematic diagram of the system structure involved in an embodiment of the present invention; Figure 2 This is a control flowchart related to an embodiment of the present invention. Detailed Implementation
[0025] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.
[0026] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.
[0027] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.
[0028] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."
[0029] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.
[0030] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.
[0031] Example 1 This embodiment provides a multi-objective load reduction system for wind turbine blades based on nonlinear model predictive control, including a sensing and measurement module, a data acquisition and preprocessing unit, a nonlinear model predictive controller, and a pitch actuator, wherein: The sensing and measurement module consists of three types of sensors, specifically: The first type is a four-beam pulse lidar installed on the top of the nacelle, used to measure the wind speed profile within a distance of 50 to 200 meters in front of the wheel hub, outputting radial wind speed distribution data along the beam direction, with a measurement frequency of 10 Hz.
[0032] The second type is fiber Bragg grating strain sensors installed at the root of each blade. Four sensors are installed on each blade, respectively located at the leading edge, trailing edge, suction surface, and pressure surface of the blade root section. They are used to measure the bending moment signal in the flapping and swinging directions, with a sampling frequency of 50 Hz.
[0033] The third type is the incremental speed encoder installed in the nacelle, used to measure the generator speed and wind turbine speed.
[0034] The data acquisition and preprocessing unit employs a field-programmable gate array (FPGA) combined with a digital signal processor (DSP) architecture. This unit connects to the lidar, strain sensor, speed encoder, and accelerometer via shielded signal cables. Internally, it includes an anti-aliasing filter, a 16-bit analog-to-digital converter (ADC), and a calibration algorithm module. After filtering, denoising, calibrating, and converting the various sensor signals to digital, it generates standardized data frames, which are then transmitted to the controller via the controller's local area network (LAN) protocol.
[0035] The preprocessing unit simultaneously performs temperature compensation on the strain signal to eliminate zero-point drift caused by temperature changes.
[0036] The nonlinear model predictive controller is an embedded computing platform comprising a multi-core processor, a floating-point arithmetic unit, and a memory module. The processor contains at least four computing cores: one dedicated to numerical integration calculations of the predictive model, one to running optimization algorithms, and the other two for data communication and monitoring. The controller's memory pre-stores blade airfoil aerodynamic data tables, structural parameter matrices, and model parameters. The controller's inputs are connected to the outputs of the data acquisition and preprocessing unit via a high-speed data bus, and the outputs are connected to the pitch actuators of the three blades via control buses. The controller executes the control algorithm at predetermined intervals, typically 50 milliseconds, generating independent pitch angle commands for each blade. After receiving the analog commands from the controller, the pitch actuators drive the blades to rotate to the target angle via servo drivers.
[0037] Example 2 Based on Example 1, this example provides a multi-objective load reduction system for wind turbine blades based on nonlinear model predictive control, wherein the controller includes: The wind speed field reconstruction unit is used to reconstruct the effective inflow wind speed field of the impeller plane based on the acquired system state information at the current moment. The model building unit is used to construct a nonlinear prediction model based on the obtained effective inflow wind speed field; The pitch sequence solving unit is used to solve the constructed multi-objective optimization cost function under the constraint of the nonlinear prediction model to obtain the optimal pitch sequence; The blade control unit is used to control the blade operation based on the obtained optimal pitch sequence.
[0038] Example 3 This embodiment provides a multi-objective load reduction method for wind turbine blades based on nonlinear model predictive control, which includes the following steps: Step 1: Obtain the current system status information.
[0039] The wind speed profile matrix, the root flapping moment and yaw moment of the three blades, the generator speed, and the wind turbine speed are obtained. This data is stored in memory as initial conditions for subsequent calculations.
[0040] Step 2: Reconstruct the effective inflow velocity field of the impeller plane.
[0041] Construct multiple mesh points on the plane of the impeller model; Based on the wind speed profile matrix, the wind speed values of multiple grid points in the impeller plane are reconstructed using the gradient descent inversion algorithm.
[0042] Based on the real-time acquired current rotor azimuth angle and combined with the wind speed value, the effective inflow wind speed of each blade at the current position is calculated. The calculation formula takes into account both wind shear and tower shadow effects.
[0043] The effective inflow velocity is calculated as follows:
[0044] in, Let r be the effective inflow velocity at a certain point, and r be the distance from a certain point on the blade to the center of the hub. The wind speed at the hub height is obtained through direct measurement. and These represent wind shear and tower shadow effects, respectively.
[0045] In this embodiment, wind shear refers to the physical phenomenon that wind speed increases with altitude, which can be described as:
[0046] in, is the hub height; r is the distance from a point on the blade to the center of the hub. This represents the current azimuth angle of the blade; 0° indicates the blade is pointing to the horizontal right; 90° indicates the blade is pointing directly upwards; α represents the wind shear index, which is a constant (generally taken as 0.1-0.3). Shearing by wind.
[0047] The tower shadow effect refers to the physical phenomenon where the wind speed decreases locally when the blades pass in front of a tower, due to the tower's obstruction of the airflow. It can be simplified as follows:
[0048]
[0049] in, d is the tower diameter; d is the horizontal distance from a point on the blade to the tower. The wind speed at the wheel hub height. This indicates that the wind speed decreases when the blades move in front of the tower due to the tower shadow effect.
[0050] When the blades are not in front of the tower =0 Wind shear is corrected using an exponential law, and the tower shadow effect is calculated using a potential flow model. When the blades are in the area in front of the tower, the wind speed decreases.
[0051] Step 3: Establish a nonlinear prediction model that incorporates blade flexibility dynamics.
[0052] The nonlinear prediction model incorporating blade flexibility dynamics is used to predict the trajectory of system state changes within the next 10 seconds, with a discretization step size of 0.1 seconds.
[0053] The nonlinear prediction model comprises three coupled sub-modules: a blade aerodynamic load calculation sub-module, a blade structural dynamics sub-module, and a nacelle-tower dynamics sub-module, wherein: The blade aerodynamic load calculation submodule is constructed based on the blade element momentum theory. It is used to divide each blade into 20 blade element units along the span. Each unit calculates lift and drag based on the effective inflow wind speed, the current pitch angle and the blade flapping speed, and then integrates to obtain the total aerodynamic flapping moment of the entire blade.
[0054] The blade structure dynamics submodule simplifies each blade into an equivalent mass-spring-damped system with two degrees of freedom: flapping and swaying. It then predicts the blade dynamics in the future time domain based on the initial conditions of the current blade displacement and velocity.
[0055] Integrate the obtained blade dynamics to obtain the blade flapping speed.
[0056] The nacelle-tower dynamics submodule uses a single-degree-of-freedom equivalent model to describe the tower's vibration in the forward and backward directions. The tower top displacement indirectly participates in the aerodynamic load calculation by affecting the tower shadow effect correction.
[0057] The three sub-modules alternately iterate within each prediction step, ensuring real-time capture of aeroelastic coupling.
[0058] Step 4: Construct the multi-objective optimization cost function.
[0059] Based on the total aerodynamic flapping moment and generator speed, the cost function is constructed, consisting of four weighted terms:
[0060] in, The peak out-of-plane bending moment term at the blade root is directly minimized by minimizing the maximum absolute value of the flapping bending moment of each blade in the prediction time domain. This is the fatigue load term, obtained using the rainflow calculation method; This is the power tracking error term; The cost of pitch maneuvering is to penalize excessive pitch rate and amplitude. To constrain violations of the penalty clause, the bending moment at the blade root is penalized if it exceeds the limit value; , , , This is a weighting coefficient that is adaptively adjusted based on wind speed conditions. When the wind speed is below the rated wind speed... Take the larger value, which is higher than the rated wind speed and the bending moment exceeds the limit value by 70%. and Increase the load appropriately and switch to the load reduction priority mode.
[0061] Step 5: Solve the nonlinear optimization problem to obtain the optimal pitch sequence.
[0062] The nonlinear optimization problem is solved online using the cost function from step 4 as the optimization objective, the nonlinear prediction model from step 3 as the equality constraint, and the pitch angle boundary and pitch rate limit as the inequality constraint. The optimization variable is the pitch angle sequence of the three blades in the prediction time domain, with a control step size of 0.5 seconds within a 10-second prediction time domain.
[0063] The optimization problem employs an iterative approach, selecting the optimal initial point from multiple sources obtained through translation and extrapolation of the previous cycle's optimal solution. After solving, the optimal pitch angle sequence is obtained, and the controller extracts the command value corresponding to the current moment.
[0064] Step 6: Output the pitch angle command and execute it.
[0065] The controller sends the pitch angle command to the pitch drive, and the servo drive drives the blades to rotate to the target angle through a position closed loop.
[0066] Step 7: Rolling time-domain update.
[0067] At the start of the next control cycle, new sensor data is acquired as initial conditions, the start time of the prediction time domain is updated, and steps 1 to 6 are repeated to achieve real-time optimal control of the blade load.
[0068] This invention constructs a multi-objective cost function that includes peak bending moment and fatigue damage terms, and adopts an adaptive weight adjustment mechanism, making load reduction an active behavior of the controller. By incorporating the maximum allowable bending moment of the blade into the optimization framework in the form of a soft constraint, this invention achieves proactive avoidance of load overload, reducing the number of unplanned downtimes caused by load overload.
[0069] Example 4 This embodiment also provides a computing device. The computing device includes a bus, a processor, a memory, and a communication interface. The processor, memory, and communication interface communicate with each other via the bus. The computing device can be a server or a terminal device. It should be understood that this application does not limit the number of processors and memory in the computing device.
[0070] A bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, a bus can include a path for transmitting information between various components of a computing device (e.g., memory, processor, communication interfaces).
[0071] The processor may include any one or more of the following: central processing unit (CPU), graphics processing unit (GPU), tensor processing unit (TPU), application specific integrated circuit (ASIC), field-programmable gate array (FPGA), microprocessor (MP), or digital signal processor (DSP).
[0072] Memory can include volatile memory, such as random access memory (RAM). Processors can also include non-volatile memory. volatile memory, such as read-only memory (ROM). ROM (memory only), flash memory, hard disk drive (HDD), or solid state drive (SSD).
[0073] The memory stores executable program code, which the processor executes to implement the functions of the aforementioned units, thereby achieving, for example, the method described in Embodiment 1. That is, the memory may store instructions for the methods and functions relating to the computing device in any of the above embodiments.
[0074] The communication interface uses transceiver modules such as, but not limited to, network interface cards and transceivers to enable communication between computing devices and other devices or communication networks.
[0075] Example 5 This embodiment also provides a computing device cluster. The computing device cluster includes at least one computing device. The computing device can be a server, such as a central server, an edge server, or a local server in a local data center. In some embodiments, the computing device can also be a terminal device such as a desktop computer, a laptop computer, or a smartphone.
[0076] The computing device cluster includes at least one computing device. The memory of one or more computing devices in the computing device cluster may store the same instructions for performing the methods and functions related to the computing devices in any of the above embodiments.
[0077] In some possible implementations, the memory of one or more computing devices in the computing device cluster may also store partial instructions for performing the methods and functions of the computing devices involved in any of the above embodiments. In other words, a combination of one or more computing devices can jointly execute the instructions for performing the methods and functions of the computing devices.
[0078] It should be noted that the memory in different computing devices within a computing device cluster can store different instructions, which are used to execute parts of the device's functions.
[0079] In some possible implementations, one or more computing devices in a computing device cluster can be connected via a network. This network can be a wide area network (WAN) or a local area network (LAN), etc. Two computing devices are connected via the network. Specifically, they connect to the network through communication interfaces on each computing device.
[0080] Embodiments of this disclosure also provide a computer program product containing instructions that, when run on a computer, cause the computer to perform the methods and functions related to a computing device in any of the above embodiments.
[0081] Example 6 This embodiment also provides a computer-readable storage medium storing computer instructions that, when executed by a processor, cause the processor to perform the methods and functions of the computing device involved in any of the above embodiments.
[0082] Generally, the various embodiments of this disclosure can be implemented in hardware or dedicated circuitry, software, logic, or any combination thereof. Some aspects can be implemented in hardware, while others can be implemented in firmware or software, which can be executed by a controller, microprocessor, or other computing device. Although various aspects of the embodiments of this disclosure are shown and described as block diagrams, flowcharts, or represented using some other illustration, it should be understood that the blocks, apparatuses, systems, techniques, or methods described herein can be implemented as, as non-limiting examples, in hardware, software, firmware, dedicated circuitry or logic, general-purpose hardware or controllers or other computing devices, or some combination thereof.
[0083] Example 7 This embodiment provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer-executable instructions, such as instructions included in program modules, which execute in a device on a target real or virtual processor to perform the processes / methods as described above with reference to the accompanying drawings. Typically, program modules include routines, programs, libraries, objects, classes, components, data structures, etc., that perform specific tasks or implement specific abstract data types. In various embodiments, the functionality of program modules can be combined or divided among program modules as needed. The machine-executable instructions for the program modules can execute within a local or distributed device. In a distributed device, the program modules can reside in both local and remote storage media.
[0084] Computer program code used to implement the methods of this disclosure may be written in one or more programming languages. This computer program code may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus, such that when executed by the computer or other programmable data processing apparatus, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be performed. The program code may be executed entirely on a computer, partially on a computer, as a stand-alone software package, partially on a computer and partially on a remote computer, or entirely on a remote computer or server.
[0085] In the context of this disclosure, computer program code or related data may be carried on any suitable carrier to enable a device, apparatus, or processor to perform the various processes and operations described above. Examples of carriers include signals, computer-readable media, and so on. Examples of signals may include electrical, optical, radio, sound, or other forms of propagation signals, such as carrier waves, infrared signals, etc.
[0086] Computer-readable media can be any tangible medium that contains or stores programs for or relating to an instruction execution system, apparatus, or device, or a data storage device such as a data center containing one or more available media. Computer-readable media can be computer-readable signal media or computer-readable storage media. Computer-readable media can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. More detailed examples of computer-readable storage media include electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical storage devices, magnetic storage devices, or any suitable combination thereof.
[0087] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.
Claims
1. A multi-objective load reduction method for wind turbine blades based on nonlinear model predictive control, characterized in that, Includes the following steps: Based on the acquired system state information at the current moment, the effective inflow wind speed field of the impeller plane is reconstructed; Based on the obtained effective inflow wind speed field, a nonlinear prediction model is constructed; Using a nonlinear prediction model as a constraint, the constructed multi-objective optimization cost function is solved to obtain the optimal pitch sequence; The blade operation is controlled based on the obtained optimal pitch sequence.
2. The multi-objective load reduction method for wind turbine blades based on nonlinear model predictive control according to claim 1, characterized in that, Based on the acquired system state information at the current moment, the effective inflow velocity field of the impeller plane is reconstructed. The specific method is as follows: Based on the wind speed profile matrix in the system state information, the wind speed values of multiple grid points in the impeller model plane are reconstructed. Based on the obtained wind speed value and the real-time acquired current rotor azimuth angle, combined with wind shear and tower shadow effects, the effective inflow wind speed of each blade at the current position is calculated. Based on the obtained effective inflow velocity, the effective inflow velocity field of the impeller plane is reconstructed.
3. The multi-objective load reduction method for wind turbine blades based on nonlinear model predictive control according to claim 1, characterized in that, The nonlinear prediction model includes a blade aerodynamic load calculation submodule, a blade structural dynamics submodule, and a nacelle-tower dynamics submodule, wherein: The nacelle-tower dynamics submodule is used to obtain the tower top displacement based on the vibration describing the front-to-back direction of the tower. The blade structure dynamics submodule is used to obtain the blade flapping speed; The blade aerodynamic load calculation submodule is used to obtain the total aerodynamic flapping moment of each blade based on the effective inflow wind speed, blade flapping speed and the real-time obtained current pitch angle.
4. The multi-objective load reduction method for wind turbine blades based on nonlinear model predictive control according to claim 1, characterized in that, Using a nonlinear prediction model as a constraint, the constructed multi-objective optimization cost function is solved to obtain the optimal pitch sequence. The specific method is as follows: Using a nonlinear prediction model as the equality constraint; The pitch angle boundary and pitch rate limit are used as inequality constraints. The optimal pitch sequence is obtained by solving the multi-objective optimization cost function based on equality and inequality constraints.
5. The multi-objective load reduction method for wind turbine blades based on nonlinear model predictive control according to claim 1, characterized in that, The expression for the cost function in multi-objective optimization is: in, This is the peak out-of-plane bending moment term at the blade root. For fatigue load terms; This is the power tracking error term; This is the cost term for the pitch control maneuver; To constrain violations and penalties; , , , All are weighting coefficients.
6. A multi-objective load reduction system for wind turbine blades based on nonlinear model predictive control, characterized in that, include: The wind speed field reconstruction unit is used to reconstruct the effective inflow wind speed field of the impeller plane based on the acquired system state information at the current moment. The model building unit is used to construct a nonlinear prediction model based on the obtained effective inflow wind speed field; The pitch sequence solving unit is used to solve the constructed multi-objective optimization cost function under the constraint of the nonlinear prediction model to obtain the optimal pitch sequence; The blade control unit is used to control the blade operation based on the obtained optimal pitch sequence.
7. An electronic device, characterized in that, It includes a processor and a memory, the memory storing computer instructions that, when executed by the processor, cause the electronic device to perform the method of any one of claims 1 to 5.
8. A computing device cluster, characterized in that, It includes at least one computing device, each computing device including a processor and memory; The processor of the at least one computing device is configured to execute instructions stored in the memory of the at least one computing device to cause the cluster of computing devices to perform the method according to any one of claims 1 to 5.
9. A computer program product, characterized in that, The computer program product includes computer-executable instructions that, when executed, implement the method of any one of claims 1 to 5.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions that, when executed by a processor, implement the method of any one of claims 1 to 5.
