A method for stabilization control of wind-wave combined units

Abstract

This application discloses a stabilization and control method based on a wind-wave combined unit, belonging to the field of marine energy utilization. Addressing the problems in improving the stability and wind and wave resistance of large floating offshore wind turbines in existing technologies, this application utilizes a dynamic mode decomposition algorithm to reduce the dimension and decouple the wave field. The dominant wave modes are extracted within the optimal range of the floating body's relative motion and used to construct a reduced-order model. Finally, wave dynamics are predicted based on the development trends of each mode in the reduced-order model. According to the platform's motion response, the floating body is locked or released under appropriate wave surface conditions to timely gain energy or adjust the unit's attitude.

Description

Technical Field

[0001] This invention relates to a stability enhancement control method based on a wind-wave combined unit, belonging to the field of marine energy utilization. Background Technology

[0002] The platform achieves energy conversion by applying a load to the floating body, and the floating body constrains the platform, improving unit stability. Currently, mainstream power take-off (PTO) control mainly includes reactive power control, reactive control, phase control, and interlocking control. Interlocking control (equivalent to adding brake pads between the platform and the floating body to forcibly lock the floating body and prevent relative motion between them) is suitable for situations where the WEC natural period is less than the wave excitation period. It typically locks the floating body under extreme motion conditions and releases the motion under favorable wave conditions to maximize energy take-off.

[0003] Lockout control strategies are primarily based on PTO performance (such as PTO models and parameters), incident conditions (such as wave surface conditions and excitation forces), and control reference variables (such as float displacement, velocity, and acceleration). Considering the randomness of actual sea conditions, non-causal control strategies with predictive behavior (controlling only through prediction without considering the causal relationship of float motion) are employed. However, since lockout control is mainly based on the float's critical velocity or peak excitation force, broadband waves (referring to waves with a very wide frequency range, the comparison between broadband waves and normal waves is...) Figure 1 As shown, this will greatly interfere with the decision to lock and release the float (the effect of the broadband wave on the float is not much different over a period of time, which means that the speed of the float may always hover around 0, and the force of the wave on the float is always very large, but it is impossible to know whether the motion sensitivity of the float is good or bad, and it is difficult to determine when to lock and release the float).

[0004] Essentially, the velocity of the floating body is a consequential variable caused by the excitation force, and is unsuitable as a reference variable for locking and releasing; moreover, the excitation force (wave force) cannot reflect the floating body's motion response to wave conditions. Existing control strategies mainly lock and release the floating body by determining whether its velocity is zero or based on the predicted excitation force (wave force). However, the velocity of the floating body is inherently a consequential variable caused by the excitation force, and is unsuitable as a reference variable for locking and releasing, especially since broadband waves in the waves significantly affect the floating body's velocity; furthermore, the excitation force cannot reflect the floating body's motion response to wave conditions, and a larger excitation force does not necessarily mean a better floating body response. Summary of the Invention

[0005] To address the problems existing in the prior art, this application proposes a stability enhancement control method based on wind-wave combined units, which is mainly used to improve the stability of wind-wave combined units.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is a stability enhancement control method based on a wind-wave combined unit, comprising the following steps:

[0007] 1) Receive the platform's motion posture in real time and determine whether the platform's motion is excessive;

[0008] If the platform moves too much, the relative motion between the float and the platform will be locked.

[0009] 2) While performing step 1), receive the real-time wavefront and decompose the wavefront modes;

[0010] 3) Based on the floating body's response to waves, select a few regular waves to establish a reduced-order model;

[0011] 4) Short-term prediction of wavefront dynamics based on the development trends of each mode in the reduced-order model;

[0012] 5) Based on the short-term prediction results of the reduced-order model, release the floating body in a coordinated manner, and use the floating body to constrain the platform and coordinate the stability of the platform.

[0013] The optimized, wind-wave combined unit stability enhancement control method described above, step 2), specifically includes the following process: at time intervals... Collect snapshot vectors, and by Snapshot vectors Construct a snapshot matrix; perform DMD mode decomposition on the snapshot matrix.

[0014] In the optimized stability enhancement control method based on wind-wave combined units, in step 2), to ensure that the number of snapshot vectors in the snapshot matrix remains constant, only the most recently acquired data is retained. A snapshot vector.

[0015] The optimized, wind-wave combined unit stability enhancement control method described above, specifically step 3), includes retaining a limited optimal range for the floating body's motion response. Establish a reduced-order wave field using wave modes;

[0016] In step 4), for the reduced-order wave field, the wave surface situation in the near future is predicted;

[0017] In step 5), the relative position between the float and the platform is calculated to determine whether the position of the float relative to the platform has reached its peak.

[0018] When the position of the float relative to the platform reaches its peak, but the wave height of the reduced-order wave field does not reach its maximum or minimum value, the float is locked by the brake pads to maintain the relative position of the float and the platform at its peak.

[0019] When the predicted wave height reaches its maximum or minimum value, the float is released to control the float to gain maximum energy or adjust the unit's attitude.

[0020] In the optimized stability enhancement control method based on wind-wave combined units, the input to the DMD during DMD mode decomposition is a set of snapshot matrices.

[0021] , , where this pair of snapshot vectors in the snapshot matrix ( () between constant time intervals ( The data is separated, and the snapshot vector at each moment contains wave data at different locations.

[0022] In the optimized, wind-wave combined unit stability enhancement control method described above, in step 2), the snapshot vector at a certain moment is represented as: Snapshot vector and Between , Wave field snapshot matrix and Between Snapshot matrix Represented as , This represents finding the inverse of a matrix.

[0023] The DMD algorithm is applied to the floating body locking control, and the matrix is ​​solved using the DMD algorithm. .

[0024] The optimized stability enhancement control method based on wind-wave combined units described above uses the DMD algorithm to solve the matrix. The specific process includes,

[0025] Snapshot matrix of wave field Perform simplified SVD: ,

[0026] Among them, matrix It is a left singular matrix. It is a singular value diagonal matrix. It is a right singular matrix; the evolution operator is represented as .

[0027] The optimized stability enhancement control method based on wind-wave combined units described above uses the DMD algorithm to solve the matrix. The specific process also includes,

[0028] Evolution operator In the matrix Projecting onto the orthogonal subspace yields the approximate evolution operator. :

[0029] ;in, Combining expressions and ,get ;

[0030] Approximate evolution operator Perform eigenvalue decomposition to obtain its eigenvalues ​​and their corresponding eigenvectors:

[0031] , The matrix form is , ;

[0032] Among them, matrix eigenvalues These are the DMD eigenvalues, and the corresponding DMD wave modes. Represented as

[0033] The matrix form is ;

[0034] in, The DMD wave mode matrix, Approximate evolution operator eigenvector matrix;

[0035] DMD wave mode growth / decline rate and oscillation frequency It can be indirectly obtained through eigenvalues The real and imaginary parts are obtained as follows:

[0036] , ;

[0037] Among them, oscillation frequency The relationship between the wave frequency and the wave frequency is satisfied ;

[0038] Wave snapshot vector ( )Depend on Approximate recombination of DMD wave modes:

[0039] ,

[0040] in, Indicated by Amplitude of each DMD wave mode A diagonal matrix with diagonal elements. For matrix The former eigenvalues Power Form column vectors;

[0041] DMD wave mode amplitude The matrix formed Through the pseudo-inverse matrix of the DMD wave mode matrix and before Snapshot vectors calculate:

[0042] ;

[0043] DMD wave mode ( Time coefficient Represented as

[0044] ;

[0045] In step 4), the frequency of each wave mode is obtained. Then, based on the response of the floating body to the wave frequency, select The wave field is reconstructed using wave modes to obtain a reduced-order model of the wave field; the reduced-order model has excluded wave modes that are insensitive to the floating body; the reduced-order flow field at any time in the short term is obtained through... Modal prediction:

[0046] .

[0047] In step 5), according to the above formula, the float is released when the predicted wave height reaches its maximum / minimum value, and the float is controlled to maximize energy gain or the unit attitude is adjusted.

[0048] The beneficial effects of this application are as follows:

[0049] The technical solution of this application mainly addresses the problem of improving the stability and wind and wave resistance of large floating offshore wind turbines. Unlike the proposed wind-wave combined unit concept device, this invention proposes a control strategy to improve unit stability based on the concept of combined energy harvesting. That is, it improves unit stability by controlling the floating bodies deployed around the floating platform. At the same time, the control strategy can also be used to improve the energy harvesting efficiency of the floating bodies.

[0050] The technical solution of this application can effectively avoid erroneous judgments in the control of floating bodies by broadband waves. Furthermore, it fully considers the response of waves to the floating body, which is something that excitation forces cannot achieve.

[0051] In the technical solution of this application, to avoid erroneous judgments caused by broadband waves, the patent proposes a lockout control strategy based on wave spectrum peak elements. This strategy utilizes a dynamic mode decomposition algorithm to reduce the dimensionality and decouple the wave field (decomposing the random wave field into a finite number of regular waves), extracts the dominant wave modes within the optimal range of the floating body's relative motion (selecting waves suitable for the floating body's motion, i.e., waves that are sensitive to the floating body's motion), and uses them to construct a reduced-order model (superimposing a few waves), and finally predicts wave dynamics based on the development trends of each mode in the reduced-order model. According to the platform's motion response, the floating body is locked and released under appropriate wave surface conditions to timely gain energy or adjust the unit's attitude (as mentioned above, the floating body has two objectives: stabilization and energy gain). Attached Figure Description

[0052] Figure 1 A comparison diagram of broadband waves and normal wavefronts;

[0053] Figure 2 This is a flowchart of the technical solution of this application;

[0054] Figure 3 This is a schematic diagram of the wave direction and the platform and floating body in this application;

[0055] Figure 4 This is a modal analysis diagram of the technical solution in this application. Detailed Implementation

[0056] The technical features of the present invention will be further illustrated below with reference to specific embodiments.

[0057] To avoid misjudgments caused by broadband waves, this application provides a wind-wave combined unit stabilization control method, proposing a lockout control strategy based on wave spectrum peak elements. This strategy utilizes a dynamic mode decomposition algorithm to reduce the dimensionality of the wave field (decomposing the random wave field into a finite number of regular waves), extracts the dominant wave modes within the optimal range of the floating body's relative motion (selecting waves suitable for the floating body's motion, i.e., waves sensitive to the floating body's motion), and uses these modes to construct a reduced-order model (superimposing a few waves). Finally, wave dynamics are predicted based on the development trends of each mode in the reduced-order model. According to the platform's motion response, the floating body is locked or released under appropriate wave surface conditions to timely gain energy or adjust the unit's attitude.

[0058] The technical solution of this application is a floating body lockout control strategy based on wave spectrum peak elements, mainly used to improve the stability of wind-wave combined units. Under the premise of platform stability, energy harvesting efficiency can be improved through control. The control strategy flow of this application is as follows: Figure 2 As shown, it mainly includes:

[0059] 1) Receive real-time wavefronts and decompose the wavefront modes;

[0060] 2) Based on the floating body's response to waves, select a few regular waves to establish a reduced-order model;

[0061] 3) Short-term prediction of wavefront dynamics based on the development trends of each mode in the reduced-order model;

[0062] 4) Receive the platform's motion attitude in real time and determine if the platform's motion is excessive. If the platform's motion is excessive, lock the relative motion between the float and the platform;

[0063] 5) Based on the short-term prediction results of the reduced-order model, release the floating body in a coordinated manner, and improve the stability of the platform by constraining the platform through the floating body.

[0064] The steps for coupling DMD theory with control strategies:

[0065] DMD requires a set of snapshot matrices from physical experiments or numerical simulation data. and , where this pair of snapshot vectors in the snapshot matrix ( () between constant time intervals ( The data is separated, and the snapshot vector at each moment contains wave data at different locations.

[0066] Taking a numerical model / real sea state as an example, if six numerical probes / buoys are placed side-by-side on the platform to detect wave duration data, then the snapshot vector at a certain moment can be represented as follows: .

[0067] Snapshot Vector and Between , Thus, the wave field snapshot matrix and Between .matrix It can be represented as

[0068] (1),

[0069] in, , Representing a pseudo-inverse is equivalent to finding the matrix in the least squares sense. The best-fit solution.

[0070] Since matrix A contains wave surface information of the wave field, a crucial step in applying the DMD algorithm to floating body locking control is calculating the matrix. The specific solution steps are as follows:

[0071] Snapshot matrix of wave field Perform simplified SVD:

[0072] (2),

[0073] Among them, matrix It is a left singular matrix. It is a singular value diagonal matrix. It is a right singular matrix. The evolution operator can be represented as:

[0074] (3).

[0075] Evolution operator In the matrix Projecting onto the orthogonal subspace yields the approximate evolution operator. :

[0076] (4).

[0077] in, Combining expressions (3) and (4), we can obtain

[0078] (5)

[0079] Approximate evolution operator Perform eigenvalue decomposition (EVD) to obtain its eigenvalues ​​and their corresponding eigenvectors:

[0080] , (in matrix form) , (6)

[0081] Among them, matrix eigenvalues These are the DMD eigenvalues, and the corresponding DMD wave modes (evolution operators). (eigenvectors) It can be represented as

[0082] (in matrix form) (7)

[0083] in, The DMD wave mode matrix, Approximate evolution operator The eigenvector matrix.

[0084] DMD wave mode growth / decline rate and oscillation frequency It can be indirectly obtained through eigenvalues The real and imaginary parts are obtained as follows:

[0085] (8a)

[0086] (8b)

[0087] Among them, oscillation frequency The relationship between the wave frequency and the wave frequency is satisfied .

[0088] Wave snapshot vector ( ) can be Approximate recombination of DMD wave modes:

[0089] (9)

[0090] in, Indicated by Amplitude of each DMD wave mode A diagonal matrix with diagonal elements. For matrix The former eigenvalues Power Composed of column vectors. DMD wave mode amplitude The matrix formed The pseudo-inverse matrix of the DMD wave mode matrix can be used. and before Snapshot vectors calculate:

[0091] (10)

[0092] Correspondingly, DMD wave mode ( Time coefficient It can be represented as

[0093] (11)

[0094] In acquiring the frequency of each wave mode Then, selection can be made based on the floating body's response to wave frequency. The wave field is reconstructed using wave modes, resulting in a reduced-order model of the wave field. This reduced-order model excludes wave modes that are insensitive to the floating body. The reduced-order flow field at any given time in the short term can be completely obtained through... Modal prediction:

[0095] (12)

[0096] The locking control of the floating body is mainly based on the prediction results of the reduced-order model. The specific process is as follows:

[0097] (501) by time interval Collect snapshot data (vector), and by Snapshot vectors Construct a snapshot matrix. To ensure that the number of snapshot vectors in the snapshot matrix remains constant, only the most recently acquired snapshots are retained. A snapshot vector.

[0098] (502) Perform DMD mode decomposition on the snapshot matrix, and retain a finite range based on the optimal range of the floating body motion response. A reduced-order wave field is established using wave modes.

[0099] (503) For the reduced-order wave field, the wave surface situation in the short term is predicted based on Equation (12).

[0100] (504) Calculate the relative position between the float and the platform, and determine whether the position of the float relative to the platform has reached the peak value.

[0101] (505) When the position of the float relative to the platform reaches the peak value, but the wave height of the reduced wave field does not reach the maximum / minimum value, the float is locked by the brake pads to keep the relative position of the float and the platform at the peak value.

[0102] (506) According to (503), the float is released when the predicted wave height reaches the maximum / minimum value, so as to achieve the purpose of maximizing the energy gain of the float or adjusting the attitude of the unit.

[0103] This application focuses on algorithms and strategies, replacing the reference variable from the floating body velocity / excitation force with a reduced-order wave field model, without paying attention to the type of control method, such as reactive control, reactive control, phase control, and lockout control, all of which can be based on this strategy.

[0104] The patent employs a dynamic mode decomposition algorithm, and while other mode decomposition algorithms exist, such as Fourier decomposition and intrinsic orthogonal decomposition, the reduced-order models are all obtained by directly superimposing wave results from different decomposition algorithms. This patent focuses more on the selection of reference variables, replacing traditional control reference quantities based on buoy velocity / excitation force with a wave field reduced-order model.

[0105] Of course, the above description is not intended to limit the present invention, and the present invention is not limited to the examples given above. Any changes, modifications, additions or substitutions made by those skilled in the art within the scope of the present invention should be protected by the present invention.

Claims

1. A wind-wave combined unit stability enhancement control method, characterized in that: The method comprises the following steps: 1) Real-time receive the platform motion posture, and judge whether the platform is too large; If the platform is too large, lock the relative motion of the float and the platform; 2) At the same time of step 1), receive real-time wave surface, and decompose the wave surface mode; 3) According to the motion response of the float to the wave, select a few regular waves to establish a reduced order model; 4) Based on the development trend of each mode in the reduced order model, short-term predict the wave surface dynamic; 5) According to the short-term prediction result of the reduced order model, release the float cooperatively, and through the constraint of the float to the platform, stabilize the platform; The specific process of step 2) includes, at time intervals The snapshot vectors are collected and combined by The snapshot vectors are collected and combined by The snapshot vectors are collected and combined by Perform DMD modal decomposition on the snapshot matrix; The specific process of step 3) includes reserving a limited number of wave modes based on the floating body motion response optimal interval a reduced order wave field is established from the individual wave modes; In step 4), predict the wave surface in the future short term for the reduced wave field; In step 5), measure the relative position between the float and the platform, and judge whether the position of the float relative to the platform reaches the peak value; When the position of the float relative to the platform reaches the peak value, but the wave height of the reduced wave field does not reach the maximum or minimum value, lock the float through the brake pad, and keep the relative position of the float and the platform as the peak value; Release the float when the predicted wave height reaches the maximum or minimum value, control the maximum energy gain of the float or adjust the unit posture.

2. The method for increasing stability control based on a wind-wave combined unit according to claim 1, characterized in that: In step 2), to ensure that the number of snapshot vectors in the snapshot matrix remains constant, only the most recently acquired snapshot vectors are retained.

3. The method of claim 1, wherein the method is characterized by: When performing DMD modal decomposition, the input of DMD is a group of snapshot matrices 、 where the snapshot vectors in this pair of snapshot matrices are separated by a constant time interval ( ) and each snapshot vector at a time instant contains wave data at different locations. ( ) and each snapshot vector at a time instant contains wave data at different locations.

4. The method for increasing stability control based on a wind-wave combined unit according to claim 1, characterized in that: In step 2), the snapshot vector at a certain time is denoted as , the snapshot vector satisfies , ; the wave field snapshot matrix satisfies , the snapshot matrix is denoted as , denotes matrix inversion;​​ Applying the DMD algorithm to the closed-loop control of a floating body, the matrix is solved using the DMD algorithm.

Images