A floating wind turbine structure model correction method and device

Abstract

The present application relates to the technical field of wind power generation, and particularly relates to a floating wind turbine structure model correction method and device. The present application inputs a plurality of preset initial correction factors into a preset floating wind turbine structure model to obtain a generalized displacement matrix; iterates the plurality of initial correction factors with the preset target function minimum as the target to obtain a plurality of optimized completed correction factors; wherein the preset target function is constructed according to the generalized displacement matrix; and corrects the preset floating wind turbine structure model based on the plurality of optimized completed correction factors. In this way, the present application can accurately correct the floating wind turbine structure model, overcome the problem of insufficient parameter matching accuracy of the traditional model correction, effectively improve the model accuracy, and lay a reliable numerical model foundation for the safety evaluation of the floating wind turbine structure.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of wind power generation technology, in particular to a floating wind turbine structure model correction method and device. BACKGROUND

[0002] With the rapid development of offshore wind power, the problems such as limited wind energy resources in nearshore areas, strong ecological constraints, and weak site clustering are gradually emerging. In order to pursue higher quality development track, the wind power industry gradually develops to deep sea areas. Relevant data shows that more than 80% of the global offshore wind energy resource potential is contained in the sea area with a water depth of more than 60m. In deep sea areas, floating offshore wind turbines are recognized as economic energy development equipment. With the rapid expansion of floating wind power installed capacity, the derived structural safety problem is worth thinking about: there are inevitable errors in the design, processing and production stages of the floating wind turbine structure, and the actual structure and the design model are different due to the problems of marine biological attachment, structural aging and component damage in the service process, so the structural health condition is difficult to evaluate. Model correction can obtain a structural mapping model with good consistency with the actual floating wind turbine, which is an important technical means to evaluate the structural safety performance. At present, the model correction in the field of offshore floating wind turbines usually takes the blades or tower of fixed structures as the research object, and constructs the objective function of model correction based on the measured dynamic fingerprint parameters (such as natural frequency, modal shape, modal strain energy, frequency response function, etc.) of the structure, and iteratively solves the corresponding correction factor of the numerical benchmark model combined with the optimization algorithm. However, this correction method is difficult to directly embed into the floating body structure of the floating wind turbine. The root cause is mainly restricted by two technical difficulties: (1) The floating wind turbine foundation structure evolves into a "floating type", and there is an essential difference between the motion mechanism and the fixed type. The structural dynamic response presents a mixed form of "slow drift motion + reciprocating vibration", and the difficulty of representing the structural coupling behavior increases; (2) The nonlinear characteristics of the floating body structure are significant, and the correction parameter space dimension is high. The traditional correction algorithm faces the dual difficulties of ill-conditioned solution and stable convergence.

[0003] Based on this, the present application provides a floating wind turbine structure model correction method and device to solve the problem of how to accurately correct the floating wind turbine structure model. SUMMARY

[0004] In order to solve the problem of how to accurately correct the floating wind turbine structure model, the present application embodiment provides a floating wind turbine structure model correction method and device.

[0005] In a first aspect, the present application embodiment provides a floating wind turbine structure model correction method, which comprises:

[0006] Input a plurality of preset initial correction factors into a preset floating wind turbine structure model to obtain a generalized displacement matrix;

[0007] iterating the plurality of initial correction factors with a preset objective function minimum as a target to obtain a plurality of optimized completed correction factors; wherein the preset objective function is constructed according to the generalized displacement matrix;

[0008] correcting the preset floating fan structure model based on the plurality of optimized completed correction factors.

[0009] In a second aspect, an embodiment of the present application provides a floating fan structure model correction device, comprising:

[0010] a first data processing module, configured to input a plurality of preset initial correction factors into a preset floating fan structure model to obtain a generalized displacement matrix;

[0011] a second data processing module, configured to iterate the plurality of initial correction factors with a preset objective function minimum as a target to obtain a plurality of optimized completed correction factors; wherein the preset objective function is constructed according to the generalized displacement matrix;

[0012] a third data processing module, configured to correct the preset floating fan structure model based on the plurality of optimized completed correction factors.

[0013] In a third aspect, an embodiment of the present application further provides an electronic device, comprising a memory and a processor, the memory stores a computer program, and the processor executes the computer program to implement the method of any embodiment of the present application.

[0014] In a fourth aspect, an embodiment of the present application further provides a computer readable storage medium, which stores a computer program, and when the computer program is executed in a computer, the computer executes the method of any embodiment of the present application.

[0015] An embodiment of the present application provides a floating fan structure model correction method and device, first input a plurality of initial correction factors covering initial iteration reference values of structure parameters into a preset floating fan structure model, and solve a generalized displacement matrix containing multi-dimensional motion parameters such as node displacement. Then, guided by minimizing a preset objective function constructed based on the matrix, the initial correction factors are iteratively updated using an intelligent optimization algorithm until the objective function converges to a preset precision, and the optimized completed correction factors are obtained. Finally, based on these optimized correction factors, the preset floating fan structure model is updated and reconstructed, so that the dynamic characteristics of the corrected model are highly consistent with the actual floating fan. The method overcomes the problem of insufficient parameter matching precision of traditional model correction through multi-parameter collaborative optimization and dynamic response coupling analysis, effectively improves the model precision, and lays a reliable numerical model foundation for floating fan structure safety evaluation. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings described below are some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0017] Figure 1 A flow chart of a floating wind turbine structure model correction method according to one embodiment is shown;

[0018] Figure 2 A hardware architecture diagram of an electronic device provided by an embodiment of the present application is shown;

[0019] Figure 3 A structural diagram of a floating wind turbine structure model correction device according to one embodiment is shown;

[0020] Figure 4 A control system of a floating wind turbine model according to one embodiment is shown;

[0021] Figure 5 A target function of a floating body correction according to one embodiment is shown;

[0022] Figure 6 A floating body correction factor iteration process of a DWPSO algorithm according to one embodiment is shown;

[0023] Figure 7 A floating body correction factor iteration process of a WTPSO algorithm according to one embodiment is shown;

[0024] Figure 8 A model correction result of a floating body according to one embodiment is shown. DETAILED DESCRIPTION

[0025] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the following will combine the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0026] Please refer to Figure 1 The embodiments of the present application provide a floating wind turbine structure model correction method, which comprises:

[0027] Step 100: inputting a plurality of preset initial correction factors into a preset floating wind turbine structure model to obtain a generalized displacement matrix;

[0028] Step 102: iteratively correcting the plurality of initial correction factors to obtain a plurality of optimized correction factors, wherein the preset objective function is constructed according to the generalized displacement matrix;

[0029] Step 104: correcting the preset floating wind turbine structure model based on the plurality of optimized correction factors.

[0030] In the embodiment, first, a plurality of initial correction factors covering the initial iteration reference values of the structure parameters are input into the preset floating wind turbine structure model, and a generalized displacement matrix containing a plurality of dimensional motion parameters such as node displacement is obtained. Then, guided by the minimization of the preset objective function constructed based on the matrix, the initial correction factors are iteratively updated using an intelligent optimization algorithm until the objective function converges to a preset precision, and the optimized correction factors are obtained. Finally, based on these optimized correction factors, the parameter of the preset floating wind turbine structure model is updated and reconstructed, so that the dynamic characteristics of the corrected model are highly consistent with the actual floating wind turbine. The method overcomes the problem of insufficient parameter matching precision of the traditional model correction by coupling multi-parameter collaborative optimization and dynamic response analysis, effectively improves the model precision, and lays a reliable numerical model foundation for the safety evaluation of the floating wind turbine structure.

[0031] In an embodiment of the present application, the plurality of initial correction factors include a floating body generalized hydrodynamic correction factor matrix, a floating body generalized mooring force correction factor matrix, and a floating body inertia correction factor matrix.

[0032] In an embodiment of the present application, the preset floating wind turbine structure model is constructed by the following formula:

[0033]

[0034] In the formula, α P is the floating body inertia correction factor matrix, α H is the floating body generalized hydrodynamic correction factor matrix, α M is the floating body generalized mooring force correction factor matrix, q is the generalized displacement matrix, is the velocity matrix, is the acceleration matrix, M is the mass matrix, f is the force matrix, and t is the time of dynamic simulation.

[0035] In the embodiment, the floating body structure model correction dynamics equation is constructed taking the NREL-5MWOC4 semi-submersible floating wind turbine as an example, and based on the Kane method, the nonlinear dynamics equation of the reference model is represented as:

[0036]

[0037] where, and F r are the generalized inertia force and the generalized active force of the reference model, respectively.

[0038] The generalized inertia force F is expanded into the form of generalized partial forces:

[0039]

[0040] where, and F

[0041] The generalized inertia force F contributed by the floating body is expanded into the form of generalized partial forces by Kane's equation:

[0042]

[0043] where, is the inertia correction factor matrix of the floating body; and are the generalized acceleration and the generalized velocity of the i-th degree of freedom of the floating body, respectively; and are the partial linear velocity and the partial angular velocity corresponding to the r-th generalized degree of freedom of the floating body at the position of the center of mass Y of the floating body and the center X of the tower base, respectively. E ω X is the angular velocity of the floating body; m X is the translational inertia (mass) of the floating body, is the corrected rotational inertia.

[0044] When the volume and shape of the real floating body are different from the theoretical model, the displacement volume, the center of buoyancy, the waterplane area or will be changed, and the hydrostatic restoring matrix is expressed as:

[0045]

[0046] where, A0and X f are the waterplane area and the waterplane centroid coordinate of the floating body, respectively; I x and I y are the roll and pitch rotational inertia of the floating body, respectively; Z0is the center of buoyancy coordinate, and V0is the displacement volume. are the correction factors of the waterplane area A0, the center of buoyancy coordinate Z0and the displacement volume V0of the floating body, respectively; p w is the seawater density, g is the gravity acceleration, C S =A0p w g. ​​

[0047] Based on the above equation, the generalized hydrostatic force contributed by the hydrostatic restoring force of the floating body can be expressed as:

[0048]

[0049] where, is the element of the hydrostatic restoring force matrix in the ith row and jth column; q j is the generalized displacement of the jth degree of freedom of the floating body; δ(i,3) is the element of the identity matrix in the ith row and 3th column, δ(i,3) = 1 when i = 3, and δ(i,3) = 0 when i ≠ 3.

[0050] The hydrodynamic force load independent of the acceleration of the floating body can be expressed as:

[0051]

[0052] where, and represent the hydrodynamic force and moment, respectively, independent of the acceleration of the floating body. e1, e2, e3 are the basis vectors of the three coordinate axes of the earth coordinate system; F i Wave is the wave force; K ij is the retardation function.

[0053] Similarly, by assigning a correction factor, the added mass matrix of the floating body can be expressed as:

[0054]

[0055] where, C A is the added mass coefficient; S0 represents the total area of the bottom of the pontoon; S, S x , and S xx represent the shape integral parameters corresponding to the three integral dimensions of the floating body. is the correction factor of the total area S0 of the bottom of the pontoon; are the correction factors of the shape integral parameters S, S x , and S xx , respectively.

[0056] The hydrodynamic force and moment dependent on the acceleration of the floating body can be expressed as:

[0057]

[0058] where, A ji represents the element of the added mass matrix in the jth row and ith column.

[0059] Further, the generalized hydrodynamic force load F rHF1 of the floating body can be expressed as:

[0060]

[0061] The generalized loads of the mooring system can be expressed as:

[0062] F i Lines (q) = F i Lines0 K ij q j

[0063] where K ij is the element of the linear stiffness matrix in the i-th row and j-th column; F i Lines0 is the pre-tension caused by the mooring weight.

[0064] Considering the symmetry of the mooring surge and sway stiffness, the symmetry of the pitch and roll stiffness, the overall stiffness matrix of the mooring after the correction can be expressed as:

[0065]

[0066] where K is the correction factor of the mooring stiffness matrix; K ss , K hh , K rr , K yy represent the uncoupled stiffness of the mooring surge, heave, sway and yaw, respectively; K sp and K ps represent the coupled stiffness of the mooring surge and pitch; K wr and K rw represent the coupled stiffness of the mooring sway and roll.

[0067] The correction factor of the mooring pre-tension The generalized loads of the real mooring system can be expressed as:

[0068]

[0069] where K ij is the element of the corrected mooring stiffness matrix in the i-th row and j-th column.

[0070] The generalized forces and generalized moments experienced by the real floating body system are expressed as:

[0071]

[0072] Further, the generalized mooring loads F rHF2 acting on the floating body can be expressed as:

[0073]

[0074] The generalized active load of the semi-submersible floating wind turbine is expanded as:

[0075] F r = F r | AF + F r | GF + F r | EF + F r | DF + F r | HF1 + F r | HF2 + F r | GE

[0076] Wherein, F r | AF , F r | GF , F r | EF , F r | DF and F r | GF are generalized aerodynamic force, generalized gravity, generalized elastic force, generalized damping force and generalized generator driving force of the floating wind turbine respectively.

[0077] The preposed coefficient of the generalized acceleration is extracted, and the dynamic equation of the floating body structure model (preset floating wind turbine structure model) is finally expressed as:

[0078]

[0079] Wherein, and are the correction factor matrices of the generalized hydrodynamic force and the generalized mooring force of the floating body respectively; M is the mass matrix of the semi-submersible floating wind turbine, and f represents the force matrix; q,

[0080] are the generalized displacement, velocity and acceleration matrices of the semi-submersible floating wind turbine respectively; and t is the time of dynamic simulation.

[0081] The innovation is that the correction factor associated with the generalized inertia force of the floating body and the floating body structure, additional water quality and constraint boundary is combined, and the nonlinear dynamic equation suitable for the correction of the floating body structure model is innovatively constructed, thereby breaking the limitation of the traditional model correction method on the floating wind turbine structure, and providing a dynamic model basis for subsequent parameterization correction of the floating wind turbine floating body structure.

[0082] In an embodiment of the present application, a plurality of initial correction factors are iterated with the preset target function minimum as the target, and a plurality of optimized completed correction factors are obtained, including:

[0083] determining whether the target function of the current iteration is less than a preset threshold value;

[0084] if yes, output the correction factor of the current iteration, i.e. the optimization completed correction factor;

[0085] if no, proceed to the next iteration;

[0086] wherein for each iteration, it is determined whether the random inertia weight is replaced by the dynamic inertia weight according to the weight replacement period, and the updating speed is controlled through the weight replacement optimization particle swarm algorithm.

[0087] In the embodiment, when the plurality of initial correction factors are iteratively optimized to obtain the optimization completed correction factor, a rigorous convergence judgment and dynamic optimization strategy are adopted. Specifically, after each iteration, the system compares the target function value of the current iteration with the preset threshold value: if the target function value is less than the preset threshold value, it indicates that the model correction has reached the expected accuracy, and at this time the correction factor of the current iteration is output as the optimization completed correction factor; if the condition is not met, the next iteration is continued. In the iteration process, in order to improve the search efficiency and convergence stability of the algorithm, for each iteration, it is determined whether the random inertia weight is replaced by the dynamic inertia weight according to the weight replacement period, and the updating speed is controlled through the weight replacement optimization particle swarm algorithm. This mechanism can adaptively adjust the global search and local optimization ability of the algorithm, avoid falling into a local optimal solution, so that the iterative optimization process of the correction factor is more efficient and accurate, and ensures that the optimization completed correction factor obtained finally can effectively improve the accuracy of the floating type fan structure model.

[0088] In an embodiment of the present application, the preset target function is constructed by the following formula:

[0089] q i (n)=q[n+(i-1)L]

[0090] i(i=1,2,…L)

[0091]

[0092] SE=RMSE[S i (ω,α P ,α H ,α M ),S j (ω)]

[0093]

[0094] In the formula, ω(n) is a window function, is the average power of the window function ω(n), is an imaginary unit, denotes the energy of the window function ω(n), S j (ω,α P ,α H ,α M ) is the true floating body's response power spectral density, S j (ω) is the generalized displacement power spectral density of the jth degree of freedom of the reference model floating body, x i is the generalized displacement matrix, SF is the preset target function, is the waterline area correction factor of the floating body, is the center of buoyancy coordinate correction factor, is the displacement volume correction factor, is the first shape integral parameter correction factor, is the second shape integral parameter correction factor, is the third shape integral parameter correction factor, is the first mooring stiffness matrix correction factor, is the second mooring stiffness matrix correction factor, is the third mooring stiffness matrix correction factor, is the fourth mooring stiffness matrix correction factor, is the fifth mooring stiffness matrix correction factor, is the sixth mooring stiffness matrix correction factor, is the seventh mooring stiffness matrix correction factor, α H is the floating body generalized hydrodynamic correction factor matrix, α M is the floating body generalized mooring force correction factor matrix, α P is the floating body inertia correction factor matrix, is the first floating body inertia correction factor, is the second floating body inertia correction factor, is the third floating body inertia correction factor, is the fourth floating body inertia correction factor, L is the total number of equal divisions of the discrete floating body displacement signal.

[0095] In the embodiment, the motion of the floating body structure in the marine environment exhibits a dynamic behavior with large amplitude, slow speed and obvious drift characteristics. Under such a motion mode, the motion energy of the floating body structure is dispersed in a wider frequency range, and it is suitable to use power spectral density to characterize it.

[0096] Based on the average periodogram method, the power spectral density of the structure response can be accurately extracted. Specifically, first, the discrete floating body displacement signal with data quantity N is equally divided into L segments, each with data quantity M, and the adjacent segments overlap M / 2 data quantity. Then, the ith(i=1, 2, …L) discrete displacement sequence is:

[0097] q i(n) = q[n + (i - 1)L]

[0098] The smoothing window ω(n) is added to each segment of the discrete displacement sequence and Fourier transform is performed, and the following equation is obtained:

[0099]

[0100] In the equation, Imaginary unit is represented.

[0101] The power spectral density of L segments of discrete displacement is calculated respectively and the average value is taken, and the power spectral density estimated by the average periodogram method is represented as:

[0102]

[0103] In the equation, The average power of the window function ω(n) is; The energy of the window function ω(n) is represented.

[0104] The average value of the power spectral density estimated by the average periodogram method is represented as:

[0105]

[0106] The variance of the estimated power spectral density can be approximated as:

[0107]

[0108] Based on the above theory, it can be found that the average periodogram method is to process the displacement response of the floating body by aliasing, segmenting and windowing, thereby effectively reducing the variance and improving the estimation accuracy of the PSD.

[0109] Let the generalized displacement power spectral density of the jth degree of freedom of the reference model floating body estimated by the average periodogram method be S j (ω), and the response power spectral density of the real floating body be S j (ω, α P , α H , α M ), the dynamic fingerprint function of the model correction is:

[0110] SE = RMSE[S i (ω, α P , α H , α M ), S j (ω)]

[0111] By normalizing to eliminate the magnitude difference of the generalized displacement of the 6 degrees of freedom of the floating body, the objective function of the floating body model correction can be defined as:

[0112]

[0113] The innovation is reflected in the following aspects: It is the first to propose a dynamic fingerprint function for the correction of floating structure models. By estimating the PSD of the floating body's motion response through the average periodogram method, it effectively characterizes the motion characteristics of the floating body's large-scale slow drift. This breaks the limitations of traditional model correction methods for floating wind turbine structures and provides an objective function for the parameterized optimization of subsequent particle swarm optimization algorithms.

[0114] In one embodiment of the present invention, the dynamic inertia weight and the random inertia weight are determined by the following formula:

[0115]

[0116] In the formula, ω d The dynamic inertia weight is N, where N is the number of generations in the population evolution, k is the number of iterations, and ω is the value of k. max The maximum is the maximum inertia weight factor, ω min ω is the minimum inertia weighting factor. r Let σ be the random inertia weight, σ be the standard deviation of the normally distributed random disturbance term, and r be the random inertia weight. d is the first preset parameter, and μ is the second preset parameter.

[0117] In one embodiment of the present invention, the weight permutation optimization particle swarm algorithm is constructed using the following formula:

[0118]

[0119] In the formula, The position component of the generalized inertial force at the k-th iteration. The position component of the generalized hydrodynamic force at the k-th iteration. Let ω be the position component of the generalized mooring force at the k-th iteration. r To randomly change the values, For the rate of particle renewal, The position of particle update, ω d For dynamic inertia weights, c d1 c is the first asynchronous learning factor. d2 r1 is the second asynchronous learning factor, r2 is the first random number, and p is the second random number. best (k) represents the global optimal position found by the population search, and K0 is the weight replacement period.

[0120] In this embodiment, the correction factors for generalized inertial force, generalized hydrodynamic force, and generalized mooring force are determined to be M, D, and P, respectively, based on the derived dynamic equations of the floating body model, forming an M+D+P dimensional objective solution space; the number of individuals in the particle population is defined as m, and the number of generations of the population is N; the position and velocity vectors of the i-th dimension particle in the k-th iteration are represented as:

[0121]

[0122] where, are the position components of the generalized inertia force, the generalized hydrodynamic force and the generalized mooring force at the kth iteration, respectively; are the corresponding velocity components.

[0123] The dynamic inertia weight is introduced to improve the performance of the PSO algorithm. Taking the mass correction factor of the floating body as an example, the velocity and position of the particle at the k+1th iteration are:

[0124]

[0125] where, c d1 , c d2 is the asynchronous learning factor; ω d is the dynamic inertia weight, which can be expanded as:

[0126]

[0127] where, ω max and ω min are the maximum and minimum inertia weight factors, respectively.

[0128] The dynamic inertia weight can autonomously reduce the weight value with the increase of the iteration number. Although this approach can better balance the global search ability at the early stage of iteration and the local improvement performance at the later stage of iteration, if the algorithm does not search the target solution near the early and middle stages of iteration, the inertia weight factor will decrease to a low point at the later stage of iteration, the global search ability will decrease, and the algorithm will fall into a local optimum. In order to make the search solution of the optimization algorithm jump out of the local optimum trap, the random inertia weight ω r is introduced to replace the dynamic inertia weight ω d at the later stage of iteration, which balances the global exploration and local development ability of the algorithm and improves the global search ability of the algorithm at the later stage.

[0129]

[0130] where, σ is the standard deviation of the normal distribution random disturbance term; N(0, 1) is a random number of standard normal distribution; r d is a random number between 0 and 1.

[0131] The velocity transformation of the weight replacement optimization PSO algorithm at the k+1th iteration is:

[0132]

[0133] where, r1 and r2 are random numbers between 0 and 1, respectively; g best (k) is the global optimal position searched by the population; K0 is the weight replacement period. When k≤K0, ωd From large to small, that is, the initial stage of the algorithm has strong global search ability, and gradually increases the local optimization performance with the iteration period; when k > K0, ω r Randomly change the numerical value, and enhance the global search ability of the algorithm again to avoid local optimization.

[0134] According to the correction factor of the k+1 iteration, the motion response of the floating body and the target function are calculated, and the local and global optimal positions of the population particles are updated. When the target function SF of the iteration is smaller than the set threshold SF min or the current evolution generation reaches N, the iteration of the weight replacement optimization particle swarm algorithm ends, and the global optimal position at this time is the final solution of the floating body structure correction factor.

[0135] The innovation lies in: the optimization method of the traditional particle swarm algorithm with fixed inertia weight is converted to the form of random inertia weight replacement dynamic inertia weight, which improves the global search ability and mid-term local search performance of the traditional particle swarm algorithm in the iteration process, and enhances the ability to jump out of the local optimal trap in the later stage of the algorithm iteration, thereby improving the convergence and accuracy of the particle swarm algorithm for optimization and solution of high-dimensional correction parameters.

[0136] In this embodiment, the NREL-5MWOC4 semi-submersible floating wind turbine is taken as the research object, as shown in Figure 4 The floating wind turbine model is equipped with advanced variable pitch and yaw control systems and stable semi-submersible floating body structure, and is typical, especially suitable for the development and utilization of wind energy resources in deep water areas.

[0137] When the generalized inertia force, generalized hydrodynamic force and generalized mooring force of the synchronous correction floating body structure are corrected, the dimension of the target correction factor increases to 18, including 4 generalized inertia force correction factors, 7 generalized hydrodynamic force correction factors and 7 generalized mooring force correction factors. Considering the increase of the dimension of the target correction factor, the population size of the weight replacement particle swarm optimization algorithm is set to 600, and the evolution generation is set to 500; the weight replacement period is 200; c i1 , c i2 is 0.8, c f1 , c f2 is 2.1, ω max and ω min are 0.9 and 0.4 respectively; the position and velocity search interval is set to [0.5, 1] and [-0.2, 0.2] respectively. In order to highlight the advantages of WTPSO algorithm, it is compared with dynamic weight particle swarm (DWPSO) algorithm, and the interference of random term of the algorithm is excluded in the iteration calculation, that is, the initialization position vector a(0) and velocity vector b(0) of the population particles of the two algorithms are unified in the initialization stage, and the random values r1 and r2 of the two algorithms are unified in the loop iteration stage.

[0138] The calculation results are as follows:

[0139] The objective function of the optimization algorithm for solving the overall correction factor of the floating body structure is as shown in the formula (1). Figure 5 The results show that the objective function of the DWPSO algorithm is stabilized after 156 iterations. In contrast, the objective function values of the WTPSO algorithm are significantly reduced at the 207th, 301st and 468th iterations and finally tend to 0, indicating that the overall convergence performance of the WTPSO algorithm is better.

[0140] As shown in the formulas (2) and (3), Figure 5 , Figure 6 , Figure 7 In this embodiment, the results show that the correction factors No. 12, 13, 15, 16 and 17 solved by the DWPSO algorithm fall into local optimum after the 140th, 156th, 59th, 143rd and 92nd iterations, respectively, Figure 6 as shown in the formula (4). This phenomenon continues until the end of the optimization process. Due to the introduction of the random inertia weight, the WTPSO algorithm makes the correction factors falling into local optimum at the middle stage jump out of the trap, for example, the correction factors No. 12, 15, 16 and 17 no longer limit to the steady state during the 200th to 400th iterations and exhibit strong volatility, that is, the diversity of particle iteration movement is enhanced.

[0141] As shown in the formulas (5) and (6), Figure 8 In this embodiment, with the increase of the dimension of the correction factor, the accuracy of the DWPSO algorithm for solving is reduced. For example, the mooring correction factors No. 13, 14 and 15 have great difference with the theoretical values, and the relative errors are 6.67%, 5.76% and 21.44%, respectively. The reason is that the dynamic inertia weight focuses on local development at the later iteration stage and loses the ability of external exploration of the correction factors not iterated to the vicinity of the target solution. The 18 correction factors solved by the WTPSO algorithm have higher degree of agreement with the theoretical values, and the identification accuracy of all factors is more than 96%, and the relative errors of the identified are 3.65%, 3.9% and 0.18%, respectively, which reflects the accuracy of the WTPSO for solving high-dimensional problems. It is generally considered that the WTPSO algorithm combining the dynamic inertia weight and the random inertia weight enhances the global exploration ability at the early iteration stage, and then gradually improves the local search performance, prevents the algorithm from falling into the local optimal trap at the later iteration stage, increases the opportunity to jump out of the local solution, and effectively improves the optimization accuracy of the algorithm.

[0142] As shown in the formulas (7) and (8), Figure 2 , Figure 3As shown, this embodiment of the invention provides a floating wind turbine structural model correction device. The device embodiment can be implemented through software, hardware, or a combination of both. From a hardware perspective, such as... Figure 2 The diagram shown is a hardware architecture diagram of an electronic device for correcting a floating wind turbine structure model, provided in an embodiment of the present invention. (Except for...) Figure 2 In addition to the processor, memory, network interface, and non-volatile memory shown, the electronic device in the embodiment may also include other hardware, such as a forwarding chip responsible for processing packets. Taking software implementation as an example, such as... Figure 3 As shown, a device in a logical sense is formed by the CPU of the electronic device in which it is located reading the corresponding computer program from the non-volatile memory into the memory for execution.

[0143] like Figure 3 As shown, this embodiment provides a floating wind turbine structure model correction device, comprising:

[0144] The first data processing module 300 is used to input multiple preset initial correction factors into a preset floating wind turbine structure model to obtain a generalized displacement matrix.

[0145] The second data processing module 302 is used to iterate over multiple initial correction factors with the goal of minimizing a preset objective function to obtain multiple optimized correction factors; wherein, the preset objective function is constructed based on the generalized displacement matrix;

[0146] The third data processing module 304 is used to correct the preset floating wind turbine structure model based on multiple optimization correction factors.

[0147] In one embodiment of the present invention, the plurality of initial correction factors include a floating body generalized hydrodynamic correction factor matrix, a floating body generalized mooring force correction factor matrix, and a floating body inertia correction factor matrix.

[0148] In one embodiment of the present invention, the preset floating wind turbine structure model is constructed using the following formula:

[0149]

[0150] In the formula, α P Let α be the matrix of correction factors for the inertia of the floating body. H Let α be the generalized hydrodynamic correction factor matrix of the floating body. M Let q be the generalized mooring force correction factor matrix for the floating body, and q be the generalized displacement matrix. For the velocity matrix, Let f be the acceleration matrix, M be the mass matrix, f be the force matrix, and t be the time of the dynamic simulation.

[0151] In one embodiment of the present application, the second data processing module 302 is configured to perform the following operations:

[0152] determining whether the target function of the current iteration is less than a preset threshold value;

[0153] if yes, outputting the correction factor of the current iteration, i.e., the optimization completion correction factor;

[0154] if no, performing the next iteration;

[0155] wherein for each iteration, it is determined whether the random inertia weight is replaced by the dynamic inertia weight according to the weight replacement period, and the weight replacement optimization particle swarm algorithm is used to control the update speed.

[0156] In one embodiment of the present application, the dynamic inertia weight and the random inertia weight are determined by the following formula:

[0157]

[0158] wherein ω d is the dynamic inertia weight, N is the population evolution algebra, k is the iteration number, ω max max is the maximum inertia weight factor, ω min min is the minimum inertia weight factor, ω r is the random inertia weight, σ is the standard deviation of the normal distribution random disturbance term, r d is the first preset parameter, and μ is the second preset parameter.

[0159] In one embodiment of the present application, the weight replacement optimization particle swarm algorithm is constructed by the following formula:

[0160]

[0161] wherein, is the position component of the generalized inertia force at the kth iteration, is the position component of the generalized hydrodynamic force at the kth iteration, is the position component of the generalized mooring force at the kth iteration, ω r is the random transformation value, is the speed of particle update, is the position of particle update, ω d is the dynamic inertia weight, c d1 is the first asynchronous learning factor, c d2 is the second asynchronous learning factor, r1 is the first random number, r2 is the second random number, p best (k) is the global optimal position searched by the population, and K0 is the weight replacement period.

[0162] In one embodiment of the present application, the preset objective function is constructed by the following formula:

[0163] q i (n) = q [n + (i - 1) L]

[0164] i (i = 1, 2, … L)

[0165]

[0166]

[0167] SE = RMSE [S i (ω, α P , α H , α M ), S j (ω)]

[0168]

[0169] In the formula, ω(n) is a window function, is the average power of the window function ω(n), is an imaginary unit, denotes the energy of the window function ω(n), S j (ω, α P , α H , α M ) is the response power spectral density of the real floating body, S j (ω) is the generalized displacement power spectral density of the jth degree of freedom of the reference model floating body, x i is the generalized displacement matrix, and SF is the preset objective function, is a floating body waterline area correction factor, is a center of buoyancy coordinate correction factor, is a displacement volume correction factor, is a first shape integral parameter correction factor, is a second shape integral parameter correction factor, is a third shape integral parameter correction factor, is a first mooring stiffness matrix correction factor, is a second mooring stiffness matrix correction factor, is a third mooring stiffness matrix correction factor, is a fourth mooring stiffness matrix correction factor, is a fifth mooring stiffness matrix correction factor, is a sixth mooring stiffness matrix correction factor, is a seventh mooring stiffness matrix correction factor, and α His a generalized hydrodynamic correction factor matrix for the floating body, a M is a generalized mooring force correction factor matrix for the floating body, a P is a generalized inertia correction factor matrix for the floating body, is a first floating body inertia correction factor, is a second floating body inertia correction factor, is a third floating body inertia correction factor, is a fourth floating body inertia correction factor, L is the total number of equal divisions of the discrete floating body displacement signal.

[0170] It can be understood that the structure of the embodiment of the present application does not constitute a specific limitation on the floating wind turbine structure model correction device. In other embodiments of the present application, a floating wind turbine structure model correction device can include more or fewer components than the illustration, or combine certain components, or split certain components, or different component arrangement. The components shown can be implemented in hardware, software, or a combination of software and hardware.

[0171] The information interaction, execution process and the like between the modules in the device are based on the same concept as the method embodiments of the present application, and the specific content can be referred to the description in the method embodiments of the present application, which will not be repeated here.

[0172] The embodiment of the present application also provides an electronic device, including a memory and a processor, the memory has stored a computer program, when the processor executes the computer program, a floating wind turbine structure model correction method in any embodiment of the present application is realized.

[0173] The embodiment of the present application also provides a computer readable storage medium, the computer readable storage medium has stored a computer program, when the computer program is executed by a processor, the processor executes a floating wind turbine structure model correction method in any embodiment of the present application.

[0174] Specifically, a system or device provided with a storage medium can be provided, and the storage medium stores software program codes for realizing the functions of any embodiment of the above embodiments, and the computer (or CPU or MPU) of the system or device reads and executes the program codes stored in the storage medium.

[0175] In this case, the program codes read from the storage medium can realize the functions of any one of the above embodiments, and therefore the program codes and the storage medium storing the program codes constitute a part of the present application.

[0176] The storage medium for providing the program code includes floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), magnetic tapes, nonvolatile memory cards and ROMs. Alternatively, the program code can be downloaded from a server computer via a communication network.

[0177] In addition, it should be understood that not only the program code read by the computer can be executed, but also the operating system or the like operating on the computer can be caused to perform part or all of the actual operations based on the instructions of the program code, thereby realizing the functions of any one of the above-described embodiments.

[0178] In addition, it can be understood that the program code read by the storage medium is written into the memory provided in the expansion board inserted into the computer or the memory provided in the expansion module connected to the computer, and then part or all of the actual operations are performed based on the instructions of the program code by the CPU or the like installed on the expansion board or the expansion module, thereby realizing the functions of any one of the above-described embodiments.

[0179] It should be noted that, in this document, the terms such as first and second are merely used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such a process, method, article or device.

[0180] It can be understood by those skilled in the art that all or part of the steps of the above-mentioned method embodiments can be completed by program instruction related hardware, and the aforementioned program can be stored in a computer readable storage medium, which, when executed, performs steps including the above-mentioned method embodiments; and the aforementioned storage medium includes ROM, RAM, magnetic disk or optical disk and various program code storage media.

[0181] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the above embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A floating wind turbine structure model correction method, characterized by, The method comprises: inputting a plurality of preset initial correction factors into a preset floating wind turbine structure model to obtain a generalized displacement matrix; iterating the plurality of initial correction factors with a preset target function minimum as a target to obtain a plurality of optimized completed correction factors; wherein the preset target function is constructed according to the generalized displacement matrix; correcting the preset floating wind turbine structure model based on the plurality of optimized completed correction factors; the preset floating wind turbine structure model is constructed by the following formula: wherein is a matrix of body inertia correction factors, is a matrix of body generalized hydrodynamic correction factors, is a matrix of body generalized mooring force correction factors, is a matrix of generalized displacements, is a matrix of velocities, is a matrix of accelerations, is a matrix of masses, is a matrix of forces, t is a time of the dynamic simulation; the iterating the plurality of initial correction factors with a preset target function minimum as a target to obtain a plurality of optimized completed correction factors comprises: determining whether the target function of the current iteration is less than a preset threshold value; if yes, outputting the correction factor of the current iteration, that is, the optimized completed correction factor; if no, performing the next iteration; wherein for each iteration, whether the random inertia weight is replaced by the dynamic inertia weight is determined according to the weight replacement period, and the updating speed is controlled by the weight replacement optimization particle swarm algorithm; the dynamic inertia weight and the random inertia weight are determined by the following formula: wherein, is the dynamic inertia weight, is the population evolutionary generation, is the iteration number, is the maximum inertia weight factor, is the minimum inertia weight factor, is the random inertia weight, is the standard deviation of the normal distribution random disturbance term, is the first preset parameter, is the second preset parameter, is a random number of a standard normal distribution; the weight replacement optimization particle swarm algorithm is constructed by the following formula: wherein is the velocity of the particle for the k+1th iteration particle update, is a random transform value, is the position of the particle update, is the velocity of the particle update, is the dynamic inertia weight, is a first asynchronous learning factor, is a second asynchronous learning factor, is a first random number, is a second random number, is the global optimal position found by the swarm, is a weight replacement period; the preset target function is constructed by the following formula: [ ] [ ] [ ] wherein is a window function, is a window function is the average power, is the imaginary unit, is a window function is the energy, M is the data amount of each segment of the discrete buoy displacement signal, is the power spectral density estimated by the average periodogram method, is the response power spectral density of the real buoy, is the generalized displacement power spectral density of the reference model buoy in the degree of freedom, is the preset objective function, is the waterplane area correction factor of the buoy, is the center of buoyancy coordinate correction factor, is the displacement volume correction factor, is the first shape integral parameter correction factor, is the correction factor of the total area of the buoy bottom, is the second shape integral parameter correction factor, is the third shape integral parameter correction factor, is the first mooring stiffness matrix correction factor, is the second mooring stiffness matrix correction factor, is the third mooring stiffness matrix correction factor, is the fourth mooring stiffness matrix correction factor, is the fifth mooring stiffness matrix correction factor, is the sixth mooring stiffness matrix correction factor, is the seventh mooring stiffness matrix correction factor, is the generalized hydrodynamic correction factor matrix of the buoy, is the generalized mooring force correction factor matrix of the buoy, is the inertia correction factor matrix of the buoy, is the first buoy inertia correction factor, is the second buoy inertia correction factor, is the third buoy inertia correction factor, is the fourth buoy inertia correction factor, is the total number of equal divisions of the discrete buoy displacement signal.

2. The method of claim 1, wherein, the plurality of preset initial correction factors comprise a floating body generalized hydrodynamic correction factor matrix, a floating body generalized mooring force correction factor matrix, and a floating body inertia correction factor matrix.

3. A floating wind turbine structure model correction device, characterized by, for executing the method as claimed in any one of claims 1-2, comprising: a first data processing module for inputting a plurality of preset initial correction factors into a preset floating wind turbine structure model to obtain a generalized displacement matrix; a second data processing module for iterating the plurality of initial correction factors with a preset target function minimum as a target to obtain a plurality of optimized completed correction factors; wherein the preset target function is constructed according to the generalized displacement matrix; a third data processing module for correcting the preset floating wind turbine structure model based on the plurality of optimized completed correction factors.

4. An electronic device, comprising: comprising a memory and a processor, the memory has stored therein a computer program, and the processor, when executing the computer program, implements the method as claimed in any one of claims 1-2.

5. A computer readable storage medium, characterized in that, having stored thereon a computer program, which, when executed in a computer, causes the computer to execute the method as claimed in any one of claims 1-2.

Images