A method for judging reliability of ship model test data based on frequency distribution

Abstract

The application provides a ship model test data reliability judgment method based on frequency distribution, and specifically comprises the following steps: S1: collecting time series data of measured forces in a ship model test; S2: obtaining a frequency spectrum diagram of each measured force by using Fourier transform and extracting frequency distribution characteristics; S3: matching the frequency distribution characteristics extracted in S2 with test parameters to form a data set FDs; S4: performing a single ship model test and extracting frequency distribution characteristics; S5: comparing the result of the single ship model test in S4 with the data set FDs in S3 to perform reliability verification; and S6: outputting the result of the reliability verification; the application fully utilizes the frequency characteristics of each measured force in the ship model test, extracts frequency distribution characteristics in the frequency spectrum diagram, and creates a data set to perform reliability verification, thereby avoiding the occurrence of unstable verification data in single ship model test data verification and improving the reliability of the verification result.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of ship model test, in particular to a ship model test data reliability judgment method based on frequency distribution. BACKGROUND

[0002] In the process of ship design and production, the performance of the designed ship type needs to be verified by model test method. Through model test method, the resistance, sailing, motion and other performances of the ship are measured, and the performance of the real ship is predicted through the model-ship conversion relationship. The accuracy of the model test results directly affects the final calculation results, and if the test results are inaccurate, significant deviation results will be produced, so the reliability of the model test results is very important.

[0003] The ship model test includes ship model resistance test, propeller open water test, ship model self-propulsion test, ship model maneuvering test, ship model seakeeping test, etc. Time series are obtained during the test process, and then the test results are output from the time series.

[0004] According to the ITTC (International Towing Tank Conference) regulation, the collection time length should be as long as possible during the model test process, so as to reduce the influence of uncertain factors on the test results and increase the reliability of the test results. The time series of the test results is the source of the result calculation, and the analysis of the test result time series is also a way to evaluate the reliability of the test results from the source.

[0005] Although prolonging the collection time length is a means to reduce the result deviation, the effectiveness of the result cannot be evaluated at present, mainly in three aspects: 1) in a specific case, when the instrument equipment has installation problems or is damaged, the test personnel are difficult to directly judge the reliability of the test results; 2) during the collection of ship model test data, the method used at present is to prolong the collection time length as much as possible and directly average, but for low-frequency time series, the low-frequency time series is more sensitive to the collection time length, so it is also difficult for the test personnel to evaluate the reliability of the directly averaged results; 3) the availability of the test result time series is not fully utilized, and only the general data is archived, and the information and rules contained therein are not explored. SUMMARY

[0006] To solve the problems of time-consuming and laborious, no quantitative conclusion and inaccurate evaluation results caused by only comparing the test results obtained by repeating the same test parameters of the same ship type in the existing reliability evaluation method, the application provides a ship model test data reliability judgment method based on frequency distribution, uses Fourier transform to convert the data of the measured force in the time domain obtained in the ship model test to the frequency domain to obtain a frequency spectrum diagram composed of multiple frequency components, uses the information contained in the frequency spectrum diagram to extract the frequency distribution characteristics of the ship model test, and then forms a frequency distribution data set, compares the test results with the frequency distribution set when a single ship model test is performed, realizes the test of the reliability of the single ship model test data, and uses the frequency distribution data sets of multiple ship types in different states to correct the test results to obtain more reasonable mean results, greatly verifies the efficiency and accuracy of the reliability of the ship model test test results.

[0007] The specific scheme is as follows:

[0008] A ship model test data reliability judgment method based on frequency distribution,

[0009] S1: collecting time series data: collecting the time series signals of the measured force varying with time in the ship model test of different ship types, different block coefficients and different non-dimensional speeds Fr, the measured force being the thrust or resistance or torque received by the ship model;

[0010] S2: extracting the frequency distribution characteristics of the measured force: using Fourier transform to convert the time series signals of the measured force varying with time in S1 to the frequency domain to obtain the frequency spectrum diagram of the measured force in the frequency domain, recording each component of the measured force in the frequency domain as a frequency component, the horizontal axis and the vertical axis of the frequency spectrum diagram being the frequency value and the amplitude value of the frequency component respectively, recording the frequency component corresponding to the peak value of the amplitude value in the frequency spectrum diagram as the main frequency component, and extracting the frequency value of the main frequency component and the frequency distribution interval of all frequency components of the measured force in the frequency domain as the frequency distribution characteristics of the ship model test;

[0011] S3: forming a data set FDs: collecting the test parameters of all ship model tests in S1, the test parameters including: ship type, block coefficient CB, non-dimensional speed, test type, matching the test parameters of each ship model test collected with the frequency distribution characteristics of each ship model test in S2 to form a "ship type-block coefficient-non-dimensional speed-test type-frequency distribution characteristics" data set FDs;

[0012] S4: performing a single ship model test: selecting any one of the different ship types described in S1 to perform a single ship model test, obtaining a time series TS0 of the measured force over time in a period of time, converting the time series TS0 to the frequency domain using Fourier transform to obtain the frequency spectrum described in S2, extracting the frequency value of the main frequency component of the measured force and the frequency distribution interval of all frequency components of the measured force in the frequency domain as the frequency distribution characteristics of the single ship model test, and collecting the test parameters of the single ship model test;

[0013] S5: reliability verification: comparing the test parameters and frequency distribution characteristics of the single ship model test obtained in S4 with the test parameters and frequency distribution characteristics data of each group of ship model tests in the data set FDs described in S3, and if a group of data that is consistent with the test parameters and similar in frequency distribution characteristics of the single ship model test is found in the data set FDs, it is considered that the single ship model test result passes the reliability verification;

[0014] S6: outputting the test results of the single ship model test that passes the reliability verification described in S5, and if the single ship model test does not pass the reliability verification, re-performing the test.

[0015] Preferably, the different ship types described in S1 include oil tankers, bulk carriers, container ships, and special ships.

[0016] Preferably, the calculation method of the square coefficient CB described in S1 is:

[0017]

[0018] where Δ is the displacement volume of the ship, L is the length of the ship, B is the width of the ship, and T is the draft of the ship in the state of the displacement volume.

[0019] Preferably, the calculation process of the dimensionless velocity Fr described in S1 is:

[0020]

[0021] where V is the speed of the ship, L is the length of the ship, and G is the acceleration of gravity.

[0022] Preferably, the process of Fourier transform described in S2 is:

[0023]

[0024] where FD(ω) is the frequency distribution after Fourier transform of the time series data, TS(t) is the value of the measured thrust or resistance or torque at any time, and ω is the angular frequency.

[0025] Preferably, the S3 covers the range of 0.5-0.9 for the block coefficient CB and the range of 0.05-0.4 for the non-dimensional speed Fr in the collected ship model test.

[0026] Preferably, the S4 uses Fourier transform to perform frequency analysis on the time series TS0 in the single ship model test.

[0027]

[0028] wherein fd(ω) is the frequency distribution after Fourier transform of the single ship test time series TS0, ts(t) is the value of the resistance or thrust or torque measured at any time in the single ship test, and ω is the angular frequency.

[0029] Preferably, the S5 uses the condition that the ship type and the measurement type of the ship model test in the data set FDs are consistent with those of the single ship model test, the deviation of the block coefficient CB is less than 10%, and the deviation of the non-dimensional speed Fr is less than 2% to determine whether the test parameters are consistent.

[0030] Preferably, the S5 uses the condition that the deviation of the frequency distribution interval of the measurement force in the frequency domain is less than 1 Hz, and the deviation of the frequency value of the main frequency component is less than 1 Hz to determine whether the frequency distribution characteristics are approximate.

[0031] The present application has the following beneficial effects:

[0032] The application provides a ship model test data reliability judgment method based on frequency distribution, converts time domain data of ship model test measured in a period of time into frequency distribution curves in a frequency domain, evaluates reliability of a single ship model test result by using frequency distribution rules, converts time series data into frequency distribution curves by using FFT fast Fourier transform in S2, provides more information about frequency and amplitude of ship model test data, and is more conducive to analyzing and processing frequency characteristics of ship model test data; meanwhile, the time series data which is only processed by averaging in the past is decomposed into components of different frequencies, improving availability of test data; in S3, test parameters of all ship model tests and frequency distribution characteristics corresponding to the ship model tests are matched to form a data set FDs, existing test data is integrated and classified, the characteristics that the collected test data has many ship types and many experimental conditions are fully utilized, frequency distribution of test results of various ship types in different states is obtained, and then frequency distribution rules of a large amount of time series data are formed, improving efficiency and effectiveness of ship model test reliability evaluation, and realizing more accurate judgment of reliability of single ship model test data; in S5, in addition to comparing ship types and frequency distribution characteristics, square coefficients and non-dimensional speeds are added for comparison, in the field of ship model test, the square coefficient is a main ship type parameter of ship model resistance, is used for representing the relationship between ship displacement and each dimension size of the ship model, and is important data of the degree of fat and thin of the underwater shape of the ship, in the application, the square coefficient is taken as one of parameters, so that the unstable situation of ship model test data caused by ship type coefficient change can be avoided. The non-dimensional speed Fr is a parameter for expressing speed in the field of ship model test, the influence of ship length on test results in the process of ship speed change is removed, and the reliability of ship model test measurement results is improved. BRIEF DESCRIPTION OF DRAWINGS

[0033] Figure 1 A flowchart of a ship model test data reliability judgment method based on frequency distribution.

[0034] Figure 2 is a curve graph of the change of the thrust of the oil tanker A with time in the ship model test collected by the embodiment S1.

[0035] Figure 3 is a frequency spectrum graph of the thrust of the bulk carrier A in the frequency domain in the embodiment S2.

[0036] Figure 4 is a curve graph of the change of the thrust of the bulk carrier C with time in the single test in the embodiment S4.

[0037] Figure 5 is a frequency spectrum graph of the thrust of the bulk carrier C in the frequency domain in the embodiment S4. DETAILED DESCRIPTION

[0038] The application will be further described below in conjunction with the drawings and examples.

[0039] As shown in Figure 1 a frequency distribution based ship model test data reliability judgment method,

[0040] S1: Collect time series data: collect the time series signal of the measured force varying with time in the ship model test of different ship types, different block coefficients and different non-dimensional speeds Fr, the measured force being the thrust or resistance or torque received by the ship model; as shown in Figure 2 the thrust varying with time curve diagram of oil tanker A with a block coefficient of 0.76 at a non-dimensional speed of 0.1 in the ship model test.

[0041] S2: Extract the frequency distribution characteristics of the measured force: convert the time series signal of the measured force varying with time in S1 to the frequency domain by Fourier transform to obtain the frequency spectrum diagram of the measured force in the frequency domain, record each component in the frequency domain as a frequency component, the horizontal and vertical axes of the frequency spectrum diagram being the frequency value and amplitude value of the frequency component respectively, record the frequency component corresponding to the amplitude value peak in the frequency spectrum diagram as the main frequency component, and extract the frequency value of the main frequency component and the frequency distribution interval of all frequency components in the frequency domain of the measured force as the frequency distribution characteristics of the ship model test; as shown in Figure 3 the frequency spectrum diagram of the thrust in the frequency domain obtained after Fourier transform of the time series signal of the thrust varying with time of oil tanker A collected in S1, wherein points M, N and O are amplitude value peaks, the frequencies corresponding to points M, N and O being 0.025 Hz, 3.8 Hz and 7.6 Hz respectively, and the frequency distribution interval of all frequency components of the thrust of oil tanker A in the frequency domain being 0-9 Hz.

[0042] S3: Form a data set FDs: collect the test parameters of all ship model tests in S1, the test parameters including: ship type, block coefficient, non-dimensional speed, test type, match the test parameters of each ship model test collected with the frequency distribution characteristics of each ship model test in S2 to form a "ship type-block coefficient-non-dimensional speed-test type-frequency distribution characteristics" data set FDs, the "ship type-block coefficient-non-dimensional speed-test type-frequency distribution characteristics" data set FDs in part of the ship model tests collected being as shown in Table 1:

[0043] Table 1

[0044]

[0045] S4: Perform a single ship model test: select any ship type in S1 to perform a single ship model test, obtain a time series TS0 of the measured force varying with time in a period of time, as shown inFigure 4 Fig. 1 shows a plot of the measured thrust of a bulk carrier C with a block coefficient of 0.8 and a non-dimensional speed of 0.13 over time, and Fig. 2 shows the frequency spectrum S2 of the time series TS0 obtained by converting the time series TS0 into the frequency domain using Fourier transform, wherein the points X, Y, Z are the peak values of the amplitudes, and the frequency values corresponding to the points X, Y, Z are the main frequency components of the thrust of the bulk carrier C. Figure 5 Fig. 1 shows a plot of the measured thrust of a bulk carrier C with a block coefficient of 0.8 and a non-dimensional speed of 0.13 over time, and Fig. 2 shows the frequency spectrum S2 of the time series TS0 obtained by converting the time series TS0 into the frequency domain using Fourier transform, wherein the points X, Y, Z are the peak values of the amplitudes, and the frequency values corresponding to the points X, Y, Z are the main frequency components of the thrust of the bulk carrier C.

[0046] S5: Reliability verification: compare the test parameters and the frequency distribution characteristics of the single ship model test obtained in S4 with the test parameters and the frequency distribution characteristics data of each group of ship model tests in the data set FDs in S3, and if a group of data that is consistent with the test parameters and similar in frequency distribution characteristics of the single ship model test is found in the data set FDs, it is considered that the single ship model test result passes the reliability verification.

[0047] S6: output the test results of the single ship model test that passes the reliability verification in S5, and if the single ship model test does not pass the reliability verification, retest.

[0048] Preferably, the different ship types in S1 include oil tankers, bulk carriers, container ships, and special ships.

[0049] Preferably, the calculation method of the block coefficient CB in S1 is:

[0050]

[0051] where Δ is the displacement of the ship, L is the length of the ship, B is the width of the ship, and T is the draft of the ship in the state of the displacement.

[0052] Preferably, the calculation process of the non-dimensional speed Fr in S1 is:

[0053]

[0054] where V is the speed of the ship, L is the length of the ship, and G is the acceleration of gravity.

[0055] Preferably, the process of Fourier transform in S2 is:

[0056]

[0057] Wherein, FD(ω) is the frequency distribution of the time series data after Fourier transform, TS(t) is the value of the measured thrust or resistance or torque at any time, and ω is the angular frequency.

[0058] Preferably, the square coefficient CB of the collected ship model test in S3 can cover the range of 0.5-0.9, and the non-dimensional speed Fr can cover the range of 0.05-0.4.

[0059] Preferably, the process of frequency analysis of the time series TS0 in the single ship model test in S4 is:

[0060]

[0061] Wherein, fd(ω) is the frequency distribution of the single ship test time series TS0 after Fourier transform, ts(t) is the value of the measured resistance or thrust or torque at any time, and ω is the angular frequency.

[0062] Preferably, the judgment condition of the consistent test parameters in S5 is that the ship type of the certain ship model test in the data set FDs is consistent with the single ship model test, the measurement type is consistent, the deviation of the square coefficient CB is less than 10%, and the deviation of the non-dimensional speed Fr is less than 2%. In S4, the square coefficient of the bulk carrier C is 0.8, and the non-dimensional speed is 0.13. Compared with the square coefficient and the non-dimensional speed of the bulk carrier A collected in the data set FDs in S2: 1) the ship type is consistent, i.e. both are bulk carriers, 2) the geometric parameters are approximate, i.e. the deviation range of the square coefficient CB is less than 10%, 3) the navigation parameters are approximate, i.e. the deviation range of the non-dimensional speed Fr is less than 0.02, and 4) the measurement parameters are consistent, i.e. both are thrust. Therefore, the bulk carrier C satisfies the condition of consistent test parameters in the data set FDs, and the comparison of the frequency distribution characteristics can be carried out.

[0063] Preferably, the requirement of the frequency distribution characteristic approximation in S5 is that the frequency distribution interval deviation of the measured force of the certain ship model test in the data set FDs and the single ship model test is less than 1 Hz in all frequency components in the frequency domain, and the frequency value deviation of the main frequency component is less than 1 Hz. The comparison of the frequency distribution characteristics of the bulk carrier C and the bulk carrier A collected in the data set FDs in S2 is obtained: 1) the frequency distribution interval of the thrust in all frequency components in the frequency domain is approximately the main frequency interval, and the deviation is not more than 1 Hz, and 2) the frequency value of the main frequency component is approximately, and the deviation is not more than 1 Hz; the requirement of the frequency distribution characteristic approximation is satisfied, so the test result reliability of the single test of the bulk carrier C satisfies the basic requirement, and the test result of the single test of the bulk carrier C is output.

[0064] It should be noted that the above detailed description of the specific embodiments of the present application is not intended to limit the present application in any way. Thus, while the present application has been described in detail with reference to specific embodiments thereof, it will be apparent to those skilled in the art that various modifications and changes can be made thereto without departing from the spirit and scope of the present application.

Claims

1. A frequency distribution-based ship model test data reliability judgment method, characterized in that, S1: Collecting time series data: collecting the time series signal of the measured force varying with time in the ship model test of different ship types, different block coefficients and different non-dimensional speeds Fr, the measured force being the thrust or resistance or torque received by the ship model; S2: Extracting the frequency distribution characteristics of the measured force: converting the time series signal of the measured force varying with time in S1 into the frequency domain by Fourier transform to obtain the frequency spectrum diagram of the measured force in the frequency domain, each component in the frequency domain being recorded as a frequency component, the horizontal and vertical axes of the frequency spectrum diagram being the frequency value and amplitude value of the frequency component respectively, the frequency component corresponding to the peak amplitude value in the frequency spectrum diagram being recorded as the main frequency component, and the frequency value of the main frequency component and the frequency distribution interval of all frequency components of the measured force in the frequency domain being extracted as the frequency distribution characteristics of the ship model test; S3: Forming a data set FDs: collecting the test parameters of all ship model tests in S1, the test parameters including ship type, block coefficient CB, non-dimensional speed, test type, matching the test parameters of each ship model test collected with the frequency distribution characteristics of each ship model test in S2 to form a "ship type-block coefficient-non-dimensional speed-test type-frequency distribution characteristics" data set FDs; S4: Performing a single ship model test: selecting any ship type in S1 to perform a single ship model test, obtaining a time series TS0 of the measured force varying with time within a period of time, converting the time series TS0 into the frequency domain by Fourier transform to obtain the frequency spectrum diagram in S2, extracting the frequency value of the main frequency component of the measured force and the frequency distribution interval of all frequency components of the measured force in the frequency domain as the frequency distribution characteristics of the single ship model test, and collecting the test parameters of the single ship model test; S5: Reliability verification: comparing the test parameters and frequency distribution characteristics of the single ship model test obtained in S4 with the test parameters and frequency distribution characteristics data of each group of ship model tests in the data set FDs in S3, and if a group of data consistent with the test parameters and similar in frequency distribution characteristics of the single ship model test is found in the data set FDs, it is considered that the single ship model test result passes the reliability verification; S6: Outputting the test results of the single ship model test that passes the reliability verification in S5, and if the single ship model test does not pass the reliability verification, re-performing the test.

2. The method for judging the reliability of ship model test data based on frequency distribution according to claim 1, characterized in that, The different ship types in S1 include oil tankers, bulk carriers, container ships and special ships.

3. The method for judging the reliability of ship model test data based on frequency distribution according to claim 1, characterized in that, The calculation method of the block coefficient CB in S1 is: where Δ is the displacement of the ship, L is the length of the ship, B is the beam of the ship, and T is the draft of the ship in the state of the displacement.

4. The method for judging the reliability of ship model test data based on frequency distribution according to claim 1, characterized in that, The calculation process of the non-dimensional speed Fr in S1 is: where V is the ship speed, L is the length of the ship, and G is the acceleration of gravity.

5. The method for judging the reliability of ship model test data based on frequency distribution according to claim 1, characterized in that, The process of Fourier transform in S2 is: Wherein, FD(ω) is the frequency distribution of the time series data after Fourier transform, TS(t) is the value of the measured thrust or resistance or torque at any time, and ω is the angular frequency.

6. The method for judging the reliability of ship model test data based on frequency distribution according to claim 1, characterized in that, S3, the coverage range of the collected ship model test square coefficient CB is 0.5-0.9, and the coverage range of the dimensionless velocity Fr is 0.05-0.

4.

7. The method for judging the reliability of ship model test data based on frequency distribution according to claim 1, characterized in that, S4, the process of frequency analysis of time series TS0 in the single ship model test by Fourier transform is: Wherein, fd(ω) is the frequency distribution of the single ship test time series TS0 after Fourier transform, ts(t) is the value of the measured resistance or thrust or torque at any time, and ω is the angular frequency.

8. The method for judging the reliability of ship model test data based on frequency distribution according to claim 1, characterized in that, S5, the determination condition of the consistent test parameters is that the ship type and the measurement type of the certain ship model test in the data set FDs and the single ship model test are consistent, the deviation of the square coefficient CB is less than 10%, and the deviation of the dimensionless velocity Fr is less than 2%.

9. The method for judging the reliability of ship model test data based on frequency distribution according to claim 1, characterized in that, S5, the requirement for the approximate frequency distribution characteristics is that the frequency distribution interval deviation of the measured force in the frequency domain of the certain ship model test in the data set FDs and the single ship model test is less than 1Hz, and the frequency value deviation of the main frequency component is less than 1Hz.

Images