Ultra-long flexible blade flutter control method and equipment based on digital twinning

Abstract

The invention provides an ultra-long flexible blade flutter control method and equipment based on digital twinning, and the method comprises the steps: constructing a virtual-real interaction wind turbine generator ultra-long flexible blade digital twinning model based on an artificial intelligence large model, and inputting data monitored by a sensor into the digital twinning model; the full state quantity of the blade in the operation process of the wind turbine generator can be accurately depicted in real time under the condition of limited measuring points; by building a wind turbine generator set ultra-long flexible blade flutter suppression algorithm based on model prediction, overall collaborative optimization of ultra-long flexible blade flutter suppression among multiple targets and multiple links can be achieved; the real-time state of the blade is monitored and predicted through simulation of a virtual model, and active flutter suppression of the blade is achieved by designing coordination control logic and a multi-target multi-link optimization control algorithm.

Description

Technical Field

[0001] The present invention relates to the field of wind power generation technology, and in particular to a method, device, equipment and storage medium for controlling flutter of ultra-long flexible blades based on digital twins. Background Art

[0002] Blade flutter instability is the most destructive aeroelastic phenomenon. How to develop effective blade flutter control methods is one of the key issues in promoting the development of wind power technology. At present, the active flutter suppression technology of wind turbine blades at home and abroad is basically controlled by a single link of torque or pitch control. There is no report on the coordinated control of torque, pitch and yaw. Ultra-long flexible blades operate in harsh environments for a long time. Their dynamic characteristics are complex and there are many vibration modes. It is difficult to characterize the flutter boundary of the running blades. The existing state monitoring methods cannot effectively realize the comprehensive monitoring of blades. It is difficult to achieve coordinated control only by relying on the theoretical fully coupled aeroelastic dynamics model and flutter characteristics research. Therefore, it is necessary to carry out research on the flutter control method of ultra-long flexible blades based on the digital twin technology of wind turbines. It is of great significance to the safe and efficient operation of large wind turbines, which can further enrich the control theory and methods of wind turbines, and is also an important content of the comprehensive localization of wind turbine control technology.

[0003] In the field of ultra-long flexible blade flutter boundary analysis technology: Wuhan University of Science and Technology established a vibration torque transmission model in the blade-drive shaft-motor rotation plane, but this technology only considered the coupling relationship between the transmission chain torsional vibration and the blade flutter; Southwest Jiaotong University studied the multi-modal coupling vibration of wind turbine blades, focusing on the influence of tension-bending coupling, high-low-order mode coupling, and bending coupling on the aeroelastic stability, vibration characteristics, nonlinear dynamic behavior and stability of the blade; the Ministry of Education Key Laboratory of Power Plant Energy Transfer Conversion and System studied the influencing factors of airfoil flutter under mild stall conditions, focusing on the analysis of the vibration characteristics of the blade itself; North China Electric Power University (Beijing) studied the structural dynamics of large wind turbines, focusing on the coupling between wind turbine towers and blades; Chongqing University achieved wind turbine transmission chain vibration control through active control, and did not further analyze the influence of transmission chain vibration on blade vibration. Based on the above discussion, there is no analysis of the dynamic flutter boundary of ultra-long flexible blades of wind turbines under complex wind conditions.

[0004] In the field of blade flutter suppression technology, domestic and foreign researchers have proposed a variety of methods and control strategies for the flutter suppression of large wind turbine blades. For example, North China Electric Power University suppresses blade flutter by changing the material of the blade; Inner Mongolia University of Technology reduces blade vibration by optimizing the airfoil structure of the blade; China Huaneng Group uses a suppression device composed of multiple pairs of electrically controlled vortex generators in series to suppress blade vibration; Sinoma Science and Technology Wind Power Blade Co., Ltd. provides an active mass distribution control device for wind turbine blades, including a counterweight and a control component that drives the counterweight to move; National Technical University of Athens applies passive load control technology to reduce the vibration caused by stalling during shutdown or no-load operation; Polytechnic University of Valencia analyzes the benefits of composite material laying, which can increase the critical wind speed of blade flutter by 10%. The above studies all focus on passive suppression of blade flutter.

[0005] The School of Electrical Engineering of Xinjiang University conducted active control research on the flutter of large wind turbine blades, and proposed a wind turbine blade vibration control system with better anti-interference performance by adopting a flexible trailing edge flap method; Harbin Institute of Technology proposed an active vibration control method based on preset performance and terminal sliding mode, and designed a neural network state observer to achieve the regulation of the system's transient performance while ensuring the steady-state performance of the system; Shandong University of Science and Technology designed a variable domain optimal fuzzy proportional integral differential (proportional intergration differentiation, PID) controller and a model predictive control (MPC) algorithm based on offset control for time-varying, lagging, and nonlinear wind turbine blade systems to solve the problem of classic blade flutter fracture failure; The State Key Laboratory of New Energy Power Systems of North China Electric Power University designed a trailing edge flap active controller based on the radial basis function (RBF) neural network adaptive proportional, integral, and differential (PID) method; Yangzhou University proposed a combination of fractional order control and internal model control (IMC) for wind turbine blade flutter systems. A new flutter control method of Inertial Control (IMC) was proposed by Xi'an University of Technology; an active controller was designed using model predictive control (MPC) technology to achieve vibration control of blades; a sliding mode controller based on linear matrix inequality (LMI) was constructed by Shanghai University to suppress the aeroelastic flutter of blades; Shanghai Institute of Electric Power studied the changing laws of vibration characteristics of wind turbines with different blade tip structures under yaw conditions, focusing on the optimization of blade structure; Inner Mongolia University of Technology used computational fluid dynamics (CFD) to study the influence of second-order beat vibration on the average and fluctuation characteristics of torque and axial thrust of wind turbines under yaw conditions, and comprehensively analyzed the distribution patterns of torque, axial load and tangential load, which was mainly aimed at static conditions and could not reflect the changing laws under dynamic control.

[0006] In summary, it is currently difficult to fully monitor the vibration state of the ultra-long flexible blades of wind turbines, and the traditional single bandpass filter damping vibration suppression algorithm cannot achieve multi-link and multi-objective optimization, and the flutter suppression effect needs to be improved. Summary of the invention

[0007] The purpose of the present invention is to provide a flutter control method for ultra-long flexible blades based on digital twins, aiming to accurately depict the full state quantity of the ultra-long flexible blades during the operation of the wind turbine set in real time and actively suppress the flutter of the ultra-long flexible blades.

[0008] The present invention discloses a flutter control method for an ultra-long flexible blade based on digital twin, comprising:

[0009] Collect historical operation data of extra-long flexible blades of wind turbines within a preset time period interval;

[0010] The fluid-solid coupling mechanism model of the wind turbine ultra-long flexible blades was established using ANSYS simulation software. The historical operation data was input into the fluid-solid coupling mechanism model to obtain millisecond-level operation simulation data of the key parts of the tip, middle and root of the wind turbine ultra-long flexible blades.

[0011] A digital twin model of the wind turbine's ultra-long flexible blades is established based on the artificial intelligence big model. The historical operation data and operation simulation data are input into the digital twin model and training is completed. The real-time operation data of the wind turbine is input into the trained digital twin model, and the blade flutter prediction data of the wind turbine's ultra-long flexible blades is output.

[0012] Based on the blade flutter prediction data, a predictive control algorithm is established to control the blade flutter of the extra-long flexible blades of the wind turbine.

[0013] The historical operation data of the ultra-long flexible blades of the wind turbine generator set within the preset time period interval is collected, including:

[0014] By installing an optical fiber load sensor at the root of an ultra-long flexible blade of a wind turbine, load data of the root of the ultra-long flexible blade of the wind turbine is collected;

[0015] A non-contact laser blade vibration sensor is used to collect vibration data of the key parts of the tip, middle and root of the wind turbine's ultra-long flexible blades during the operation of the wind turbine.

[0016] Collect the impeller speed, torque and pitch angle data of wind turbines;

[0017] Ensure that the sampling period and timestamp of impeller speed, torque, pitch angle, blade root load and vibration data are consistent;

[0018] The preset time period interval is at least 1 month.

[0019] Among them, the ANSYS simulation software is used to establish the fluid-solid coupling mechanism model of the ultra-long flexible blades of the wind turbine. The historical operation data is input into the fluid-solid coupling mechanism model to obtain the millisecond-level operation simulation data of the key parts of the tip, middle and root of the ultra-long flexible blades of the wind turbine, including:

[0020] Use the nacelle-mounted laser wind radar to measure the wind speed and direction data at different distances and heights in front of the ultra-long flexible blades of the wind turbine;

[0021] Inverse the flow field pressure distribution within a certain distance in front of the entire impeller face and build a fluid-solid coupling mechanism model for flexible blades;

[0022] The historical operation data is input into the fluid-solid coupling mechanism model to obtain millisecond-level operation simulation data of the key parts of the tip, middle and root of the ultra-long flexible blades of the wind turbine.

[0023] Among them, a digital twin model of the wind turbine's ultra-long flexible blades is established and trained based on the artificial intelligence big model. The digital twin model is trained with historical operation data and operation simulation data. The real-time operation data of the wind turbine is input into the digital twin model, and the blade flutter prediction data of the wind turbine's ultra-long flexible blades is output, including:

[0024] Deploy AI big model technology locally to train high-precision digital twin models of wind turbine ultra-long flexible blades, and deploy the trained big models to local high-performance industrial control computers;

[0025] Input the real-time collected wind speed, wind direction, torque, pitch angle, and impeller speed data into the digital twin model;

[0026] Through the digital twin model, the state monitoring and prediction information of vibration, load and flutter margin of key parts of the blade is output as the blade flutter prediction data of the ultra-long flexible blades of the wind turbine.

[0027] Among them, the predictive control algorithm formula is expressed as:

[0028] X(k+1)=A(k)X(k)+B(k)U(k) (1)

[0029] Wherein, X(k) represents the vibration, load, and flutter margin state of the blade at the kth moment, X(k)=[x1(k)x2(k)x3(k)], X(k+1) represents the vibration, load, and flutter margin state of the blade at the k+1th moment, A(k) represents the state matrix of the system, B(k) represents the input matrix of the system, U(k) represents the input control quantity, U(k)=[ΔT e (k)Δβ(k)],ΔT e(k) represents the control amount of the electromagnetic torque, and Δβ(k) represents the control amount of the pitch angle.

[0030] Among them, the blade flutter of the wind turbine super-long flexible blade is controlled by predictive control algorithm, including:

[0031] The historical operation data and operation simulation data are input into the digital twin model to predict the blade vibration, load, and flutter margin states for the next p control cycles, which are:

[0032]

[0033] Among them, p represents the prediction time domain, look at k+1|k means predicting the blade state at time k+1 at the current time k, and so on;

[0034] The control quantity U(k) in the prediction time domain is expressed as:

[0035]

[0036] The operating status of the next p control cycles is monitored by formula (1), which is expressed as:

[0037]

[0038] Formula (4) is expressed in matrix as:

[0039] X k =Ψ(k)X(k)+Θ(k)U k (5)

[0040] in,

[0041] It also includes the step of establishing a multi-objective optimization function, and the formula of the multi-objective optimization function is expressed as:

[0042] min F(U k )=min(f1(U k ),f2(U k ),—f3(U k ),f4(U k )) (6)

[0043] in,

[0044]

[0045] f4(U k )=U k T WU k , W>0

[0046] The constraint condition is: |U(k+i|k)|≤U max i=0,1,2,…,p-1

[0047] In the multi-objective optimization function, R k It is always 0 during the forecast period. is the mean value of blade vibration within the prediction period, is the critical speed of blade flutter; by solving the established multi-objective optimization function, the optimal control quantity U of the model prediction algorithm is obtained k , so that the blade vibration in the prediction period is close to the reference value, the blade load fluctuation is minimal, the blade flutter margin is maximized, and the control quantity U k , will U k The first row of elements of is extracted as the control quantity of this control cycle; U max It is the maximum value of the control quantity without affecting the stability of the existing control links.

[0048] The present invention discloses a flutter control device for an ultra-long flexible blade based on digital twin, comprising:

[0049] A data acquisition module, used to collect historical operation data of the extra-long flexible blades of the wind turbine generator set within a preset time period interval;

[0050] The fluid-solid coupling mechanism model building module is used to establish the fluid-solid coupling mechanism model of the ultra-long flexible blades of the wind turbine. The historical operation data is input into the fluid-solid coupling mechanism model to obtain millisecond-level operation simulation data of key parts such as the tip, middle and root of the ultra-long flexible blades of the wind turbine.

[0051] A digital twin model building module is used to build and train a digital twin model of the wind turbine's ultra-long flexible blades, input the wind turbine's real-time operating data into the trained digital twin model, and output blade flutter prediction data for the wind turbine's ultra-long flexible blades;

[0052] The control module is used to establish a predictive control algorithm based on blade flutter prediction data to control the blade flutter of the super-long flexible blades of the wind turbine.

[0053] The present invention discloses a computer device, including an input and output unit, a memory and a processor. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor executes the steps in the aforementioned implementation method.

[0054] The present invention discloses a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps in the aforementioned implementation method.

[0055] Different from the prior art, the present invention discloses a flutter control method for ultra-long flexible blades based on digital twins. By constructing a digital twin model of ultra-long flexible blades of a wind turbine with virtual-reality interaction based on a large artificial intelligence model, the method can accurately depict the full state of the blades during the operation of the wind turbine in real time under the condition of limited measuring points; by building a flutter suppression algorithm for wind turbine blades based on model prediction, the overall collaborative optimization of multiple objectives and multiple links of blade flutter suppression can be achieved; the present invention uses the large artificial intelligence model technology to construct a digital twin model of ultra-long flexible blades of a wind turbine, inputs the data monitored by the sensor into the digital twin model, and realizes the monitoring and prediction of the real-time state of the blades through the simulation of the virtual model. By designing coordinated control logic and multi-objective and multi-link optimization control algorithms, the active flutter suppression of the blades can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:

[0057] Figure 1 It is a flow chart of a flutter control method of an ultra-long flexible blade based on digital twinning of the present invention.

[0058] Figure 2 It is a logical schematic diagram of a digital twin-based ultra-long flexible blade flutter control method of the present invention.

[0059] Figure 3 It is a schematic diagram of the overall framework of an ultra-long flexible blade flutter control method based on digital twinning of the present invention.

[0060] Figure 4 It is a structural schematic diagram of an ultra-long flexible blade flutter control device based on digital twinning of the present invention.

[0061] Figure 5 It is a structural schematic diagram of a non-transitory computer-readable storage medium storing computer instructions provided by the present invention. DETAILED DESCRIPTION

[0062] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, specific embodiments of the present invention are now described in detail with reference to the accompanying drawings.

[0063] Please refer to Figure 1 and Figure 2 The present invention discloses a flutter control method for an ultra-long flexible blade based on digital twin, comprising:

[0064] Step S110: collecting historical operation data of the extra-long flexible blades of the wind turbine generator set within a preset time period interval.

[0065] Specifically, step S110 includes the following steps:

[0066] By installing an optical fiber load sensor at the root of an ultra-long flexible blade of a wind turbine, load data of the root of the ultra-long flexible blade of the wind turbine is collected;

[0067] A non-contact laser blade vibration sensor is used to collect vibration data of key parts of the wind turbine's ultra-long flexible blades during the operation of the wind turbine.

[0068] Collect the impeller speed, torque and pitch angle data of wind turbines;

[0069] Ensure that the sampling period and timestamp of impeller speed, torque, pitch angle, blade root load and vibration data are consistent;

[0070] The preset time period interval is at least 1 month.

[0071] S120: Establish a fluid-solid coupling mechanism model for ultra-long flexible blades of wind turbines, input historical operation data into the fluid-solid coupling mechanism model, and obtain millisecond-level operation simulation data for key parts of the tip, middle, and root of the ultra-long flexible blades of the wind turbines.

[0072] Specifically, step S120 includes the following steps:

[0073] Use the nacelle-mounted laser wind radar to measure the wind speed and direction data at different distances and heights in front of the ultra-long flexible blades of the wind turbine;

[0074] ANSYS Fluent is used to invert the flow field pressure distribution within a certain distance in front of the entire impeller surface, and ANSYS FCX is used to build a flexible blade fluid-solid coupling mechanism model;

[0075] The historical operation data is input into the fluid-solid coupling mechanism model to obtain millisecond-level operation simulation data of the key parts of the tip, middle and root of the ultra-long flexible blades of the wind turbine.

[0076] S130: Establish a digital twin model of the ultra-long flexible blades of the wind turbine based on the artificial intelligence big model, input the historical operation data and operation simulation data into the digital twin model and complete the training, input the real-time operation data of the wind turbine into the trained digital twin model, and output the blade flutter prediction data of the ultra-long flexible blades of the wind turbine.

[0077] Specifically, step S130 includes the following steps:

[0078] Deploy artificial intelligence big model technology locally, train high-precision digital twin models of ultra-long flexible blades of wind turbines, and deploy the trained big models to local high-performance industrial control computers; the digital twin model can choose a variant of the Transformer neural network architecture.

[0079] Input historical operation data and operation simulation data into the digital twin model, including wind speed, wind direction data, impeller speed, torque, pitch angle, blade root load and vibration data collected by real-time laser wind radar;

[0080] Through the digital twin model, the state monitoring and prediction information of vibration, load and flutter margin of key parts of the blade is output as the blade flutter prediction data of the ultra-long flexible blades of the wind turbine.

[0081] S140: Based on the blade flutter prediction data, a predictive control algorithm is established to control the blade flutter of the wind turbine's ultra-long flexible blades

[0082] The predictive control algorithm formula is expressed as:

[0083] X(k+1)=A(k)X(k)+B(k)U(k) (1)

[0084] Wherein, X(k) represents the vibration, load, and flutter margin state of the blade at the kth moment, X(k)=[x1(k)x2(k)x3(k)], X(k+1) represents the vibration, load, and flutter margin state of the blade at the k+1th moment, A(k) represents the state matrix of the system, B(k) represents the input matrix of the system, U(k) represents the input control quantity, U(k)=[ΔT e (k)Δβ(k)],ΔT e (k) represents the control amount of the electromagnetic torque, and Δβ(k) represents the control amount of the pitch angle.

[0085] Among them, the blade flutter of the wind turbine super-long flexible blade is controlled by predictive control algorithm, including:

[0086] The historical operation data and operation simulation data are input into the digital twin model to predict the blade vibration, load, and flutter margin states for the next p control cycles, which are:

[0087]

[0088] Among them, p represents the prediction time domain, look at k+1|k means predicting the blade state at time k+1 at the current time k, and so on;

[0089] The control quantity U(k) in the prediction time domain is expressed as:

[0090]

[0091] The operating status of the next p control cycles is monitored by formula (1), which is expressed as:

[0092]

[0093] Formula (4) is expressed in matrix as:

[0094] X k =Ψ(k)X(k)+Θ(k)U k (5)

[0095] in, It also includes the step of establishing a multi-objective optimization function, and the formula of the multi-objective optimization function is expressed as:

[0096] min F(U k )=min(f1(U k ),f2(U k ),—f3(U k ),f4(U k )) (6)

[0097] in,

[0098] The constraint condition is: |U(k+i|k)|≤U max i=0,1,2,…,p-1

[0099] In the multi-objective optimization function, R k It is always 0 during the forecast period. is the mean value of blade vibration within the prediction period, is the critical speed of blade flutter; by solving the established multi-objective optimization function, the optimal control quantity U of the model prediction algorithm is obtained k , so that the blade vibration in the prediction period is close to the reference value, the blade load fluctuation is minimal, the blade flutter margin is maximized, and the control quantity U k , will U k The first row of elements of is extracted as the control quantity of this control cycle; U max It is the maximum value of the control quantity without affecting the stability of the existing control links.

[0100] like Figure 3 As shown, Figure 3 This is the overall framework diagram of the system of the present invention. To complete the control of the entire system, a high-performance industrial control computer and a nacelle-type laser wind measurement radar need to be added. The main control PLC of the wind turbine needs to add ΔT e The two control variables Δβ(k) and Δβ(k) are superimposed on the original torque and pitch control loop.

[0101] like Figure 4 As shown, the present invention proposes an ultra-long flexible blade flutter control device 300 based on digital twin, comprising:

[0102] The data collection module 310 is used to collect historical operation data of the super-long flexible blades of the wind turbine generator set within a preset time period interval;

[0103] The fluid-solid coupling mechanism model building module 320 is used to establish a fluid-solid coupling mechanism model of the ultra-long flexible blade of the wind turbine using ANSYS simulation software, input the historical operation data into the fluid-solid coupling mechanism model, and obtain millisecond-level operation simulation data of the key parts of the tip, middle and root of the ultra-long flexible blade of the wind turbine;

[0104] A digital twin model building module 330 is used to establish and train a digital twin model of an ultra-long flexible blade of a wind turbine based on an artificial intelligence large model, input historical operation data and operation simulation data into the digital twin model and complete the training, input the real-time operation data of the wind turbine into the trained digital twin model, and output blade flutter prediction data of the ultra-long flexible blade of the wind turbine;

[0105] The control module 340 is used to establish a prediction control algorithm based on the blade flutter prediction data to control the blade flutter of the super-long flexible blades of the wind turbine.

[0106] In order to implement an embodiment, the present invention also proposes an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute each step of the method for blade flutter suppression of the aforementioned technical solution.

[0107] like Figure 5 As shown, the non-transitory computer-readable storage medium includes a memory 810 of instructions, an interface 830, and the instructions can be executed by a processor 820 for blade flutter suppression to complete the method. Alternatively, the storage medium can be a non-transitory computer-readable storage medium, for example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0108] In order to implement the embodiment, the present invention further proposes a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the operation optimization for blade flutter suppression as in the embodiment of the present invention is implemented.

[0109] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0110] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0111] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present invention belong.

[0112] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0113] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the described embodiment, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0114] A person skilled in the art may understand that all or part of the steps of implementing the method of the embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0115] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0116] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present invention have been shown and described above, it can be understood that the embodiments are exemplary and cannot be construed as limiting the present invention. A person of ordinary skill in the art may change, modify, replace and modify the embodiments within the scope of the present invention.

Claims

1. A flutter control method for ultra-long flexible blades based on digital twins, characterized in that: include: Collect historical operation data of extra-long flexible blades of wind turbines within a preset time period interval; The fluid-solid coupling mechanism model of the ultra-long flexible blade of the wind turbine is established by using ANSYS simulation software, and the historical operation data is input into the fluid-solid coupling mechanism model to obtain millisecond-level operation simulation data of the key parts of the tip, middle and root of the ultra-long flexible blade of the wind turbine; Establishing a digital twin model of the ultra-long flexible blades of the wind turbine generator set based on the artificial intelligence big model, inputting the historical operation data and the operation simulation data into the digital twin model and completing the training, inputting the real-time operation data of the wind turbine generator set into the trained digital twin model, and outputting blade flutter prediction data of the ultra-long flexible blades of the wind turbine generator set; Based on the blade flutter prediction data, a prediction control algorithm is established to control the blade flutter of the super-long flexible blades of the wind turbine generator set.

2. The flutter control method of ultra-long flexible blades based on digital twinning according to claim 1 is characterized in that: Collect historical operation data of wind turbine ultra-long flexible blades within a preset time period, including: By installing an optical fiber load sensor at the root of the super-long flexible blade of the wind turbine generator set, load data of the root of the super-long flexible blade of the wind turbine generator set is collected; A non-contact laser blade vibration sensor is used to collect vibration data of key parts of the tip, middle and root of the wind turbine's ultra-long flexible blades during the operation of the wind turbine; Collecting the impeller speed, torque and pitch angle data of the wind turbine; Ensure that the sampling period and timestamp of impeller speed, torque, pitch angle, blade root load and vibration data are consistent; The preset time period interval is at least 1 month.

3. The flutter control method of ultra-long flexible blades based on digital twinning according to claim 1 is characterized in that: The fluid-solid coupling mechanism model of the ultra-long flexible blade of the wind turbine is established by using ANSYS simulation software, and the historical operation data is input into the fluid-solid coupling mechanism model to obtain millisecond-level operation simulation data of the key parts of the tip, middle and root of the ultra-long flexible blade of the wind turbine, including: Using a nacelle-type laser wind measuring radar, the wind speed and wind direction data at different distances and heights in front of the ultra-long flexible blades of the wind turbine are measured; Inverse the flow field pressure distribution within a certain distance in front of the entire impeller face and build a fluid-solid coupling mechanism model for flexible blades; The historical operation data is input into the fluid-solid coupling mechanism model to obtain millisecond-level operation simulation data of key parts such as the tip, middle and root of the ultra-long flexible blade of the wind turbine generator set.

4. The flutter control method of ultra-long flexible blades based on digital twinning according to claim 1 is characterized in that: A digital twin model of the ultra-long flexible blades of the wind turbine is established based on the artificial intelligence big model, the historical operation data and the operation simulation data are input into the digital twin model for training, the real-time operation data of the wind turbine is input into the trained digital twin model, and blade flutter prediction data of the ultra-long flexible blades of the wind turbine is output, including: Deploy AI big model technology locally to train high-precision digital twin models of wind turbine ultra-long flexible blades, and deploy the trained big models to local high-performance industrial control computers; Inputting the real-time collected wind speed, wind direction, torque, pitch angle, and impeller speed data into the digital twin model; The digital twin model is used to output status monitoring and prediction information of vibration, load, and flutter margin of key parts of the blade as blade flutter prediction data of the ultra-long flexible blades of the wind turbine.

5. The flutter control method of ultra-long flexible blades based on digital twinning according to claim 4 is characterized in that: The predictive control algorithm formula is expressed as: X(k+1)=A(k)X(k)+B(k)U(k) (1) Wherein, X(k) represents the vibration, load, and flutter margin state of the blade at the kth moment, X(k)=[x1(k)x2(k)x3(k)], X(k+1) represents the vibration, load, and flutter margin state of the blade at the k+1th moment, A(k) represents the state matrix of the system, B(k) represents the input matrix of the system, U(k) represents the input control quantity, U(k)=[ΔT e (k)Δβ(k)],ΔT e (k) represents the control amount of the electromagnetic torque, and Δβ(k) represents the control amount of the pitch angle.

6. The flutter control method of ultra-long flexible blades based on digital twinning according to claim 5 is characterized in that: The blade flutter of the super-long flexible blade of the wind turbine is controlled by the predictive control algorithm, including: The historical operation data and the operation simulation data are input into the digital twin model to predict the blade vibration, load, and flutter margin states for the next p control cycles, which are: Among them, p represents the prediction time domain, look at k+1|k means predicting the blade state at time k+1 at the current time k, and so on; The control quantity U(k) in the prediction time domain is expressed as: The operating status of the next p control cycles is monitored by formula (1), which is expressed as: Formula (4) is expressed in matrix as: X k =Ψ(k)X(k)+Θ(k)U k (5) in, 7. The flutter control method of ultra-long flexible blades based on digital twinning according to claim 6 is characterized in that: The step of establishing a multi-objective optimization function is also included. The multi-objective optimization function formula is expressed as: min F(U k ) = min (f1(U k ), f2(U k ), —f3(U k ), f4(U k )) (6) in, The constraint condition is: |U(k+i|k)|≤U max i=0,1,2,…,p-1 In the multi-objective optimization function, R k It is always 0 during the forecast period. is the mean value of blade vibration within the prediction period, is the critical speed of blade flutter; by solving the established multi-objective optimization function, the optimal control quantity U of the model prediction algorithm is obtained k , so that the blade vibration in the prediction period is close to the reference value, the blade load fluctuation is minimal, the blade flutter margin is maximized, and the control quantity U k , will U k The first row of elements of is extracted as the control quantity of this control cycle; U max It is the maximum value of the control quantity without affecting the stability of the existing control links.

8. A flutter control device for ultra-long flexible blades based on digital twins, characterized in that: include: A data acquisition module, used to collect historical operation data of the extra-long flexible blades of the wind turbine within a preset time interval; A fluid-solid coupling mechanism model building module is used to establish a fluid-solid coupling mechanism model of the ultra-long flexible blade of the wind turbine using ANSYS simulation software, input the historical operation data into the fluid-solid coupling mechanism model, and obtain millisecond-level operation simulation data of key parts such as the tip, middle, and root of the ultra-long flexible blade of the wind turbine; A digital twin model building module, used to establish and train a digital twin model of the ultra-long flexible blades of the wind turbine based on the artificial intelligence big model, input the historical operation data and the operation simulation data into the digital twin model and complete the training, input the real-time operation data of the wind turbine into the trained digital twin model, and output blade flutter prediction data of the ultra-long flexible blades of the wind turbine; A control module is used to establish a prediction control algorithm based on the blade flutter prediction data to control the blade flutter of the super-long flexible blades of the wind turbine.

9. A computer device, characterized in that: The method comprises an input-output unit, a memory and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor executes the steps in any one of the methods as claimed in claims 1 to 7.

10. A storage medium storing computer-readable instructions, characterized in that: When the computer readable instructions are executed by one or more processors, the one or more processors are caused to perform the steps in the method according to any one of claims 1 to 7.

Images