A double peak wave spectrum calculation method for offshore wind turbine design

Abstract

This invention relates to a method for calculating bimodal wave spectra for offshore wind turbine design, applicable to the field of offshore wind power generation. Ocean waves typically exist as a mixture of wind waves and swells, manifested in a bimodal wave spectrum. Addressing the issue of poor accuracy in bimodal wave spectrum calculations using conventional wave spectrum calculation methods, this invention proposes screening criteria for bimodal wave spectra and methods for handling secondary peak dispersion, based on measured wave data. It also provides a calculation method for bimodal spectra through combinations of different wave spectra. The calculation process of this invention is as follows: 1. This invention proposes screening criteria for effective measured wave data and screening criteria for bimodal wave spectra based on measured wave data; 2. It provides dimensionless sample spectra and their averaged spectra through spectral estimation methods; 3. It derives the calculation formula for bimodal spectra and the calculation method for characteristic wave elements by fitting combinations of different spectra.

Description

TECHNICAL FIELD

[0001] The present application is suitable for the field of offshore wind power generation, and particularly relates to a calculation method of a bimodal wave spectrum involved in determining wave elements in the design of large offshore wind turbines. BACKGROUND

[0002] Although the offshore wind power industry in China has developed rapidly in recent years, there are still many challenges in the research of offshore wind turbine structures, one of which is how to accurately obtain the dynamic response of offshore wind turbine structures under complex environmental loads, especially under wave environmental loads. At present, the equivalent design wave method is often used to obtain the quasi-static response of offshore wind turbine structures to waves (without considering dynamic response and fatigue problems), while structures with dynamic response need to be modeled in time sequence according to the kinematics principle of the sea surface and its movement, one of the important contents of modeling is to accurately describe the sea state, which is specified by a wave spectrum with given significant wave height, representative frequency, average propagation direction and spreading function. The existing wave spectrum includes JONSWAP spectrum, PM spectrum, TMA spectrum and other theoretical spectrum types. According to the measured wave data of some sea areas in China, it is found that there are differences between the measured wave parameters and the wave parameters calculated by assuming the above wave spectrum, the reason is that sea waves usually exist in the form of mixed waves of wind waves and swell waves, which is reflected in the bimodal spectrum, and the existing wave spectrum is mostly based on the single peak condition, and its direct application to the bimodal spectrum of wind waves and swell waves will result in poor calculation accuracy. SUMMARY

[0003] The present application aims to solve the problem of poor accuracy of conventional wave spectrum calculation methods in bimodal spectrum calculation, and proposes a screening standard for effective measured wave data and a screening standard for bimodal sea wave spectrum through measured wave data, and gives a calculation method of bimodal spectrum through the combination of different wave spectra.

[0004] The technical scheme adopted by the present application is as follows: a bimodal wave spectrum calculation method for offshore wind turbine design, comprising the following steps:

[0005] S1, screening of measured wave data; according to the wave surface height standard, the cross-zero period standard, the wave surface change rate standard and the spectrum characteristic standard, the measured wave data is screened, and the data segment that meets the four standards at the same time is identified as an effective data segment;

[0006] S2, performing spectrum estimation on all effective data segments by Welch method to obtain sample spectrum, and screening bimodal spectrum from the sample spectrum;

[0007] S3, performing non-dimensionalization processing on the bimodal sample spectrum of step S2 and taking average to obtain the non-dimensional average spectrum of the bimodal spectrum of the measuring station;

[0008] S4. Calculate the mixed wave bimodal spectrum suitable for different sea areas by double JONSWAP combination.

[0009] As preferred, in step S1, the wave height standard is that, in each data segment, first, the wave surface data is zero-mean processed, then the root mean square of the processed wave surface data is calculated, if the wave surface value of any time point of this data segment is greater than 4 times the root mean square, then this point is defined as a bad point, otherwise it is an effective point; if the proportion of bad points in a data segment exceeds 10%, then mark the data segment as unusable, otherwise mark the data segment as usable.

[0010] As preferred, in step S1, the cross-zero period standard is that, if there is a wave cross-zero period greater than 25s in a data segment, then mark the data segment as unusable, otherwise mark the data segment as usable.

[0011] As preferred, in step S1, the wave surface change rate standard is that, if the instantaneous change rate of the wave surface is greater than the maximum change rate of the wave surface, then this point is defined as a bad point, otherwise it is an effective point; in each data segment, if the number of bad points is greater than 15% of the number of cross-zero points, then mark the data segment as unusable, otherwise mark the data segment as usable.

[0012] As preferred, in step S1, the spectral feature standard is that, in each data segment, if the spectral value S(0 + ) at zero frequency and the spectral peak value S(f p ) satisfy S(0 + ) / S(f p )≤0.2, it means that the data segment is usable; if the spectral value S(0 + ) at zero frequency and the spectral peak value S(f p ) satisfy S(0 + ) / S(f p )>0.2, it means that the data segment is wrong and should be marked as unusable.

[0013] As preferred, in step S2, the discrimination standard of the bimodal spectrum is to satisfy the following conditions at the same time:

[0014] (1) wind wave is dominant, that is, the high-frequency spectral peak value is greater than the low-frequency spectral peak value;

[0015] (2) the interval between the high and low frequencies corresponding to the high and low spectral peak values is greater than 0.05Hz;

[0016] (3) the one with the larger spectral peak value in the above high and low frequencies is the main peak, and the one with the smaller spectral peak value is the secondary peak, then the spectral peak value of the secondary peak needs to reach 30% or more of the spectral peak value of the main peak;

[0017] (4) there is a valley value between the above main and secondary peaks, and the valley value needs to be less than 2 / 3 of the spectral peak value of the secondary peak.

[0018] As preferred, in the step S4, the calculation formula of the mixed wave bimodal spectrum is:

[0019] S(μ) = S1(μ) + S2(μ)

[0020] is the dimensionless frequency, f is the frequency, and the unit is Hz, f p is the spectral peak frequency, and the unit is Hz, S1(μ) and S2(μ) represent the low-frequency spectrum and the high-frequency spectrum, respectively;

[0021] When the bimodal spectrum of the mixed wave is described by the double JONSWAP spectrum, the specific formula is

[0022] 1. Dimensionless form:

[0023]

[0024] wherein f p1 is the spectral peak frequency of the low-frequency spectrum, f p2 is the spectral peak frequency of the high-frequency spectrum;

[0025] m 01 +m 02 = 1, wherein represents the dimensionless average spectrum of the measured bimodal spectrum;

[0026]

[0027] γ i is the spectral peak elevation factor, wherein i = 1, 2;

[0028] is the dimensionless frequency, f is the frequency, and the unit is Hz, f p

[0029] is the spectral peak frequency, and the unit is Hz;

[0030] 2. Dimensional form:

[0031]

[0032]

[0033]

[0034] wherein the value of α represents the relationship between the spectral area m0 of the measured bimodal spectrum screened out in the step S2 and the effective wave height H 1 / 3 of the corresponding data segment, and is obtained by linear regression analysis of m0 and H 1 / 3 ;

[0035] is the spectrum peak period;

[0036] where f p1 represents the spectrum peak frequency of the low-frequency sub-spectrum, f p2 is the spectrum peak frequency of the high-frequency sub-spectrum;

[0037] γ i is the spectrum peak elevation factor, where,

[0038] m 01 +m 02 =1, represents the dimensionless average spectrum of the measured bimodal spectrum;

[0039] where, f is the frequency, the unit is Hz, f p is the spectrum peak frequency, the unit is Hz.

[0040] The present application has the beneficial effects that: the present application proposes the screening standard of the bimodal sea wave spectrum, the method for processing the secondary peak dispersion phenomenon through the measured wave data, and the calculation method of the bimodal spectrum through the combination of different wave spectra. Compared with the single sea wave spectrum such as the JONSWAP spectrum, the PM spectrum, the Wen spectrum and the like, the bimodal spectrum method of the present application can obviously improve the fitting precision when fitting the measured sea wave coexisting with the wind surge, thereby providing more accurate input conditions for the dynamic response modeling of the offshore wind turbine. BRIEF DESCRIPTION OF DRAWINGS

[0041] Figure 1 The measured bimodal spectrum curve and its fitting schematic diagram of Sheyang sea area;

[0042] Figure 2 The measured bimodal spectrum curve and its fitting schematic diagram of Wenling sea area. DETAILED DESCRIPTION

[0043] In order for those skilled in the art to more clearly understand the purpose, technical scheme and advantages of the present application, the present application is further described below in combination with the drawings and examples, but the present application is not limited to the following examples.

[0044] The bimodal wave spectrum calculation method for offshore wind turbine design provided by the present application comprises the following steps: S1, screening of measured wave data; according to the following four standards, the measured wave data is screened, and the data segment meeting the four standards at the same time is identified as an effective data segment: in the existing wave measuring instrument, most of them are divided into working periods according to hours, and one working period measures one data segment;

[0045] Wave height criterion; in each data segment, first, the wave height data is zero-meaned, then the root mean square of the processed wave height data is calculated, if the wave height value of any time point in this data segment is greater than 4 times the root mean square, then this point is defined as a bad point, otherwise it is an effective point; if the proportion of bad points in a data segment exceeds 10%, then mark the data segment as unusable, otherwise mark the data segment as usable;

[0046] Cross-zero period criterion; if there is a wave crossing zero period greater than 25s in a data segment, mark the data segment as unusable, otherwise mark the data segment as usable;

[0047] Wave height rate criterion; if the instantaneous change rate of the wave height is greater than the maximum change rate of the wave height, then this point is defined as a bad point, otherwise it is an effective point; in each data segment, if the number of bad points is greater than 15% of the number of zero-crossing points, mark the data segment as unusable, otherwise mark the data segment as usable;

[0048] Spectrum feature criterion; in each data segment, if the spectrum value S(0 + ) at zero frequency and the spectrum peak value S(f p ) satisfy S(0 + ) / S(f p )≤0.2, it means that the data segment is usable; if the spectrum value S(0 + ) at zero frequency and the spectrum peak value S(f p ) satisfy S(0 + ) / S(f p )>0.2, it means that the data segment is erroneous and should be marked as unusable.

[0049] Table 1 gives an example of the application of step S1 criterion;

[0050] Table 1 Screening results of applying step S1 four criteria

[0051] Location Start End Total length Bad points Actual effective Jiangsu Sheyang 2018 / 5 / 1010 2018 / 7 / 2712 1875 27 1848 Zhejiang Wenling 2018 / 5 / 1415 2018 / 7 / 3118 1875 132 1743

[0052] S2, estimate the sample spectrum by Welch method for all effective data segments, and select the double-peak spectrum from the sample spectrum, the specific criteria are as follows:

[0053] (1) wind wave is dominant, i.e. the high-frequency spectrum peak value is greater than the low-frequency spectrum peak value;

[0054] (2) the interval between high and low frequencies corresponding to high and low frequency spectrum peaks is greater than 0.05Hz;

[0055] (3) the larger spectrum peak value in the above high and low frequency spectrum is the main peak, and the smaller spectrum peak value is the secondary peak, then the secondary peak spectrum peak value needs to reach 30% or more of the main peak spectrum peak value;

[0056] (4) There is a valley between the main peak and the secondary peak, and the valley value is less than 2 / 3 of the peak value of the secondary peak.

[0057] Table 2 Screening results of bimodal spectrum

[0058] Location Effective data segment number Bimodal spectrum number Jiangsu Sheyang 1848 324 Zhejiang Wenling 1743 165

[0059] S3, after the dimensionless processing of the bimodal sample spectrum of step S2 and averaging, the dimensionless average spectrum of the bimodal spectrum of the station is obtained;

[0060] S4, through double JONSWAP combination, the mixed wave bimodal spectrum suitable for different sea areas is calculated; the calculation formula of the mixed wave bimodal spectrum is:

[0061] The calculation formula of the mixed wave bimodal spectrum is:

[0062] S(μ)=S1(μ)+S2(μ)

[0063] is the dimensionless frequency, f is the frequency, the unit is Hz, f p is the spectral peak frequency, the unit is Hz, S1(μ), S2(μ) respectively represent the low frequency spectrum and the high frequency spectrum;

[0064] When the double JONSWAP spectrum is used to describe the bimodal spectrum of the mixed wave, the specific formula is:

[0065] 1, dimensionless form:

[0066]

[0067] Where f p1 is the spectral peak frequency of the low frequency spectrum, f p2 is the spectral peak frequency of the high frequency spectrum;

[0068] m 01 +m 02 =1, where is the dimensionless average spectrum of the measured bimodal spectrum;

[0069]

[0070] γ i is the spectral peak rise factor, Where i=1, 2;

[0071] is the dimensionless frequency, f is the frequency, the unit is Hz, f p is the spectral peak frequency, the unit is Hz;

[0072] 2. Dimensional form:

[0073]

[0074]

[0075]

[0076] where α is the value representing the relationship between the measured bimodal spectrum area m0 and the effective wave height H 1 / 3 of the corresponding data segment, which is obtained by linear regression analysis of m0 and H 1 / 3 .

[0077] is the spectral peak period;

[0078] where f p1 represents the spectral peak frequency of the low-frequency sub-spectrum, and f p2 is the spectral peak frequency of the high-frequency sub-spectrum;

[0079] γ i is the spectral peak elevation factor, where, m 01 +m 02 = 1, represents the dimensionless average spectrum of the measured bimodal spectrum;

[0080] where f is the frequency in Hz, and f p is the spectral peak frequency in Hz.

[0081] It is worth noting that the dimensionless form 1 and the dimensional form 2 are equivalent, although they appear to be different, and only α is required to convert each other.

[0082] As shown in Table 1, for the calculation of the mixed wave bimodal spectrum of the coastal waters of Sheyang in Jiangsu: Figure 1

[0083] 1. Dimensionless form

[0084]

[0085]

[0086] 2. Dimensional form

[0087]

[0088] ​

[0089]

[0090] T p = 0.983T 1 / 3

[0091] T p = 1.242T z

[0092] wherein T 1 / 3 is the effective period, T z is the zero-crossing period.

[0093] As Figure 2 shown, for the mixed wave bimodal spectrum of the coastal waters of Wenling, Zhejiang:

[0094] 1. Dimensionless form

[0095]

[0096]

[0097] 2. Dimensional form

[0098]

[0099]

[0100]

[0101] T p = 1.003T 1 / 3

[0102] T p = 1.331T z

[0103] wherein T 1 / 3 is the effective period, T z is the zero-crossing period.

[0104] The above only describes the preferred embodiments of the present application, and for those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A double peaked wave spectrum calculation method for offshore wind turbine design, characterized in that, It comprises the following steps: S1, screening of measured wave data; according to the wave height standard, the cross-zero period standard, the wave surface change rate standard, the spectrum characteristic standard, the measured wave data is screened, and the data segment meeting the four standards is identified as an effective data segment; S2, spectrum estimation of all effective data segments by Welch method to obtain sample spectrum, and screening of double-peak spectrum from the sample spectrum; S3, non-dimensionalization processing of the double-peak sample spectrum of step S2 and taking average to obtain the non-dimensional average spectrum of the double-peak spectrum of the measuring station; S4, calculation of mixed wave double-peak spectrum suitable for different sea areas by double JONSWAP combination; In the step S1, the wave height standard is: in each data segment, first, the wave surface data is zero-mean processed, then the root mean square of the processed wave surface data is calculated, if the wave surface value of any time point of this data segment is greater than 4 times the root mean square, the point is defined as a bad point, otherwise it is an effective point; if the proportion of bad points in a data segment exceeds 10%, the data segment is marked as unusable, otherwise the data segment is marked as usable; In the step S1, the cross-zero period standard is: if there is a wave cross-zero period greater than 25s in a data segment, the data segment is marked as unusable, otherwise the data segment is marked as usable; In the step S1, the wave surface change rate standard is: if the instantaneous change rate of the wave surface is greater than the maximum change rate of the wave surface, the point is defined as a bad point, otherwise it is an effective point; in each data segment, if the number of bad points is greater than 15% of the number of cross-zero points, the data segment is marked as unusable, otherwise the data segment is marked as usable; The spectrum feature criterion in step S1 is: in each data segment, if the spectrum value S(0 + ) of zero frequency and the spectrum peak value S(f p ) satisfy S(0 + ) / S(f p )≤0.2, it means that the data segment is available; if the spectrum value S(0 + ) of zero frequency and the spectrum peak value S(f p ) satisfy S(0 + ) / S(f p )>0.2, it means that the data segment is erroneous and should be marked as unavailable; In the step S2, the discrimination standard of double-peak spectrum is to meet the following conditions at the same time: (1) wind wave is dominant, that is, high frequency spectrum peak value is greater than low frequency spectrum peak value; (2) the interval between high and low frequency corresponding to high and low frequency spectrum peak values is greater than 0.05 Hz; (3) the one with larger spectrum peak value in the above high and low frequency spectrum is the main peak, and the one with smaller spectrum peak value is the secondary peak, then the spectrum peak value of the secondary peak needs to reach 30% or more of the spectrum peak value of the main peak; (4) there is a valley value between the above main and secondary peaks, which needs to be less than 2 / 3 of the spectrum peak value of the secondary peak; In the step S4, the calculation formula of mixed wave double-peak spectrum is: S(μ)=S1(μ)+S2(μ) is the dimensionless frequency, f is the frequency in Hz, f p is the spectral peak frequency in Hz, S1(μ), S2(μ) represent the low and high frequency sub-spectra, respectively; When double JONSWAP spectrum is used to describe the double-peak spectrum of mixed wave, the specific formula is 1, non-dimensional form: where f p1 denotes the spectral peak frequency of the low-frequency sub-spectrum, f p2 is the spectral peak frequency of the high-frequency sub-spectrum; m 01 +m 02 = 1, where indicates the dimensionless average spectrum of the measured bimodal spectrum; gamma i is a spectral peak raising factor, where i = 1, 2; is the dimensionless frequency, f is the frequency in Hz, and f p is the spectrum peak frequency, unit is Hz; 2, dimensional form: Wherein, the value of a represents the relationship between the spectrum area m0 of the measured bimodal spectrum screened in step S2 and the effective wave height H of the corresponding data segment, which is obtained by linear regression analysis of m0 and H 1 / 3 . 1 / 3 . is the peak period; where f p1 denotes the spectral peak frequency of the low-frequency sub-spectrum, f p2 is the spectral peak frequency of the high-frequency sub-spectrum; gamma i is a spectral peak raising factor, wherein, m 01 +m 02 = 1, represents the dimensionless average spectrum of the measured bimodal spectrum; where f is the frequency in Hz, f p is the spectral peak frequency in Hz.

Images