A time series segmented ship model resistance test data evaluation method

Abstract

The application provides a time series segmented ship model resistance test data evaluation method, which is a method for evaluating single ship model resistance test data by using a segmented analysis method of time series segmentation considering the periodicity of the ship model in the running process. The method can be applied to single test of any ship type, eliminates the influence of artificial factors in the ship model resistance test, and makes the determination result of the single ship model resistance test more reliable.

Description

Technical Field

[0001] This invention relates to the field of ship hydrodynamic testing, specifically to a method for evaluating time-series segmented ship model resistance test data. Background Technology

[0002] Hydrodynamic testing technology is an essential technological foundation for the development of the shipping and shipbuilding industries, and it plays a vital role in technological advancements in ship performance and the implementation of energy conservation and emission reduction strategies. Both the International Maritime Organization (IMO) and the International Towing Tank Conference (ITTC) consider hydrodynamic testing technology an indispensable part of ship design and performance evaluation, requiring ocean-going vessels to undergo tank testing during the design phase and recommending that research institutions conduct uncertainty analysis on hydrodynamic technologies.

[0003] The verification results of the pool test are used to determine the performance of the currently designed ship. For the ship model pool test, the ship model resistance test is the first step in evaluating the ship type performance. Its purpose is to evaluate the resistance performance of the currently designed ship type under different loading conditions from the perspective of the ship type. Therefore, the accuracy of the pool test results has a direct impact on the prediction of ship performance. This means that the results of the ship model resistance test should be as accurate as possible, and the repeatability of the results should be as stable as possible.

[0004] According to ITTC regulations (7.5-02-02-01, Resistance Tests), a common method in obtaining ship model resistance test results is to extend the sampling time as much as possible and obtain the results through time series averaging. Although organizations like the ITTC have clear regulations on pool testing methods, these regulations do not specify the exact sampling time. Therefore, current methods involve manually extracting samples over a time series and averaging these samples as the test result. The length of the time series heavily depends on the experience of the testers, leading to a degree of human influence on the results and making data evaluation difficult. Furthermore, this method ignores the periodicity of forces acting on the model ship during its forward movement. The model ship exhibits significant periodicity during testing, and manually averaging the time series masks this periodicity, amplifying the randomness of single results and hindering the quantification of objectivity. While data obtained through repeated tests can improve the reliability of the data and facilitate evaluation, it significantly increases testing costs. Therefore, a method for objectively evaluating data from a single model ship resistance test, while controlling costs, is urgently needed. Summary of the Invention

[0005] To address the problem of objectively evaluating the results of single-test ship model resistance tests, this invention proposes a time-series segmented data evaluation method for ship model resistance tests. This method considers the periodicity of the ship model during its movement and uses a segmented time-series analysis approach to evaluate the data from a single ship model resistance test. This method can eliminate the influence of human factors in ship model resistance tests, making the judgment results of single-test ship model resistance tests more reliable.

[0006] The technical solution of the present invention is as follows:

[0007] A method for evaluating time-series segmented ship model resistance test data, characterized by comprising the following steps:

[0008] S1. Collect the resistance test data of the ship model at each time point in a finite time series during a single ship model resistance test, and calculate the average resistance value R0 of the ship model resistance test data at each time point in the finite time series.

[0009] S2. Establish a coordinate system with the ship model resistance value as the vertical axis and the finite time series as the horizontal axis. Use the average resistance value R0 as the reference line to segment the finite time series. Record the first time point that crosses the R0 reference line, starting from the initial time point of the finite time series, as T0. Sequentially record the second to the last time points that cross the R0 reference line as T1, T2, ... T0. max Then, sequentially assign T0-T1, T1-T2, ..., T max-1 -T max The time series intervals are denoted as periodic segments N1, N2, ..., N max That is, the finite time series is divided into max periodic segments.

[0010] S3, in T0~T max Extracting T0 to T from the time series min T0~T min+1 …, T0~T max The time series interval is defined, and the average resistance value of the time series interval is calculated, denoted as A. min A min+1 …A max The T min The T is defined as the starting time point used to calculate the average interval resistance of the time series interval. min Let T1, T2...T be the time points. max One of them.

[0011] S4, A min A min+1 …A maxThe standard deviation is calculated and denoted as S0. S0 / R0 is denoted as U0, which is the Type A uncertainty of the ship model resistance data solved in this case.

[0012] S5. Set the reference value for the Type A uncertainty of the ship model resistance data to U. A , if U0<U A Then A max The output is the final ship model resistance test data.

[0013] S6. If U0 > U A Then use A max Replace the average value R0 of the ship model resistance test data with the value of S1-S6, and continue to repeat steps S1-S6 until A that meets the conditions of step S5 is obtained. max The results were used as data for the resistance test of the ship model.

[0014] Preferably, the finite time series in step S1 is a time series in which the ship model resistance test data has reached stability, allowing the ship model resistance test data to first be judged to determine that the ship model resistance test data has reached stability.

[0015] Preferably, the starting time point T used in step S3 to calculate the average interval resistance of the time series interval is... min The value of min is greater than or equal to 10.

[0016] Preferably, the reference value U for the Type A uncertainty of the ship model resistance data in step S5 is characterized by... A =1%.

[0017] The beneficial effects of this invention are as follows:

[0018] This invention proposes a time-series segmented method for evaluating ship model resistance test data. This method takes into account the periodicity of the ship model during its movement and uses time-series segmentation, that is, dividing a finite time series into segments, to analyze the resistance differences in different time periods in order to evaluate the resistance test data of a single ship model. The method of this invention first collects data from a single ship model resistance test in a finite time series and calculates the average resistance value R0 of the finite time series. Then, using the average resistance value R0 of the finite time series in the ship model resistance test as a reference line, the invention divides the finite time series into segments for periodicity. This is mainly due to two reasons: first, the resistance magnitude of the ship model during its movement has obvious periodicity, so this processing makes the output results more reliable; second, the ship model resistance signal sampled contains signals of different frequencies, mainly low frequencies, and the time series has obvious statistical regularity. Therefore, this segmented analysis method based on the low-spectrum characteristics of time-frequency characteristics can effectively utilize the finite time series data of a single test, resulting in higher analysis efficiency. Furthermore, this invention extracts ship model resistance data from different time series intervals and performs uncertainty analysis on them. Here, this invention considers that if the time series interval selected in the ship model resistance test data calculation is too short, or the number of periods is too small, it may lead to large fluctuations in the calculation results, affecting the stability of the calculation results. Therefore, a time point threshold T is selected. min It is defined as the starting time point for calculating the average resistance level of the time series interval, and T0 to T1 are calculated respectively. min T0~T min+1 …, T0~T max The average resistance value of the time series interval is used to avoid large fluctuations in the calculated results of the ship model resistance test, thus improving the reliability of the ship model resistance test data. Finally, the uncertainty of the obtained data is analyzed. Based on the data from the author's previous work (Zhang Li, Chen Jianting, Chen Weimin, et al. Uncertainty Analysis of Composite Voyage Test of Towing Pool Standard Model Resistance [J]. Shipbuilding Engineering, 2020, 42(04):38-43+126), the reference value of the Type A uncertainty of the ship model resistance data is set as U. A If the uncertainty does not meet the requirements of the reference value, an iterative method is used to recalculate and correct the output results, and finally obtain the required ship model resistance data. The iterative method greatly improves the reliability of the output results.

[0019] The method of this invention processes and evaluates data from a single ship model resistance test to obtain a set of ship model resistance data with high confidence. The overall method also avoids the influence of subjective human factors on the final result data in the prior art. It is highly objective and operable, and theoretically it can be applied to single test data of any ship type. It comprehensively solves the problems of poor reliability of the current method of taking the average value of single ship model resistance test data and high cost of repeated ship model resistance tests. Attached Figure Description

[0020] Figure 1 This is an overall flowchart of the method of the present invention.

[0021] Figure 2 This is a flowchart of the traditional ship model resistance test method.

[0022] Figure 3 This is a graph showing the raw sampling results and mean calculation of resistance data from a ship model resistance test.

[0023] Figure 4 This is a fast Fourier transform graph of the resistance test results of a certain ship model.

[0024] Figure 5 This is a schematic diagram of the segmented resistance data from a ship model resistance test.

[0025] Figure 6 This is a graph showing the average resistance data of a model ship during different periods of the model resistance test. Detailed Implementation

[0026] The present invention will now be described with reference to the accompanying drawings.

[0027] This invention relates to a method for evaluating time-series segmented ship model resistance test data, the overall flowchart of which is shown below. Figure 1 As shown, it includes the following steps:

[0028] A method for evaluating time-series segmented ship model resistance test data, characterized by comprising the following steps:

[0029] S1. Collect the ship model resistance test data at each time point of a finite time series during a single test, and calculate the average resistance value R0 of the ship model resistance test data at each time point of the finite time series.

[0030] S2. Establish a coordinate system with the ship model resistance value as the vertical axis and the finite time series as the horizontal axis, using the average resistance value R0 as the reference line. Figure 3 The finite time series is segmented as follows: starting from the initial point of the ship model's motion, the first time point that crosses the R0 reference line is denoted as T0, and the second to the last time point that crosses the R0 reference line are denoted as T1, T2, ... T1. maxThen, sequentially assign T0-T1, T1-T2, ..., T max-1 -T max The time series intervals are denoted as periodic segments N1, N2, ..., N max That is, the finite time series is divided into max periodic segments.

[0031] S3, in T0~T max Extracting T0 to T from the time series min T0~T min+1 …, T0~T max The time series intervals are defined as time intervals starting from T0, and the average resistance value of each time series interval is calculated and denoted as A. min A min+1 …A max The T min The T is defined as the starting time point used to calculate the average interval resistance of the time series interval. min Let T1, T2...T be the time points. max One of them.

[0032] S4, A min A min+1 …A max The standard deviation is calculated and denoted as S0. S0 / R0 is denoted as U0, which is the Type A uncertainty of the ship model resistance data solved in this case.

[0033] S5. Set the reference value for the Type A uncertainty of the ship model resistance data to U. A , if U0<U A Then A max The output is the final ship model resistance test data.

[0034] S6. If U0 > U A Then use A max Replace the average value R0 of the ship model resistance test data with the value, and continue to repeat steps S1-S8 until A that meets the conditions of step S5 is obtained. max The results were used as data for the resistance test of the ship model.

[0035] Preferably, the finite time series in step S1 is a time series in which the ship model resistance test data has reached stability, allowing the ship model resistance test data to first be judged to determine that the ship model resistance test data has reached stability.

[0036] Preferably, the starting time point T used in step S3 to calculate the average interval resistance of the time series interval is... min The value of min is greater than or equal to 10.

[0037] Preferably, the reference value U for the Type A uncertainty of the ship model resistance data in step S5 is characterized by... A =1%.

[0038] According to ITTC regulations (7.5-02-02-01, Resistance Tests), common methods for obtaining ship model resistance test results include... Figure 2 As shown, the method of obtaining the ship model resistance test results by extending the sampling time as much as possible and averaging the time series is used. That is, after the ship model test enters a stable operating state, a time series over a period of time is selected as the test acquisition results, and the average of these acquisition results is taken as the test output result.

[0039] However, this method has three main drawbacks:

[0040] 1) The length of the time series is determined by human judgment. The average of the sampled values ​​of a time series is taken manually and the test result is output. This method will result in the test result being greatly affected by human factors and will depend heavily on the experience of the testers.

[0041] 2) The periodicity of the forces acting on the model ship during its forward movement was ignored, affecting the reliability of the data:

[0042] like Figure 3 As shown, the ship model exhibits obvious periodicity during the testing process. The method of manually extracting time series data for averaging will mask the periodicity of the ship model resistance test sampling signal.

[0043] like Figure 4 As shown, the Fast Fourier Transform of a certain resistance test result reveals that the time series exhibits significant low-frequency characteristics, primarily concentrated below 5 Hz. This indicates that the sampled resistance signal from the ship model contains signals of varying frequencies, predominantly low frequencies.

[0044] 3) The randomness of a single result is amplified, making it impossible to quantify the objectivity of the output result.

[0045] To address the above problems, the present invention first collects data from a single ship model resistance test within a finite time series and calculates the average resistance value R0 of the finite time series. Then, using the average resistance value R0 from the finite time series in the ship model resistance test as a reference line, the present invention divides the finite time series into segments for periodicity. This is mainly due to two reasons: first, the resistance magnitude of the ship model during its movement exhibits obvious periodicity, thus this processing makes the output results more reliable; second, the sampled ship model resistance signal contains signals of different frequencies, predominantly low frequencies, and the time series has obvious statistical regularity characteristics. Therefore, this segmented analysis method based on the low-spectrum characteristics of time-frequency characteristics can effectively utilize the finite time series data from a single test, resulting in higher analysis efficiency. Furthermore, the present invention extracts ship model resistance data from different time series intervals and performs uncertainty analysis on them. Here, the present invention considers that selecting a time series interval that is too short during the calculation of ship model resistance test data may lead to large fluctuations in the calculation results, affecting the stability of the calculation results. Therefore, a time point threshold T is selected. min It is defined as the starting time point for calculating the average resistance level of the time series interval, and T0 to T1 are calculated respectively. min T0~T min+1 …, T0~T max The average resistance value of the time series interval is used to avoid large fluctuations in the calculated results of the ship model resistance test, thus improving the reliability of the ship model resistance test data. Finally, the uncertainty of the obtained data is analyzed. Based on the data from the author's previous work (Zhang Li, Chen Jianting, Chen Weimin, et al. Uncertainty Analysis of Composite Voyage Test of Towing Pool Standard Model Resistance [J]. Shipbuilding Engineering, 2020, 42(04):38-43+126), the reference value of the Type A uncertainty of the ship model resistance data is set as U. A If the uncertainty does not meet the requirements of the reference value, an iterative method is used to recalculate and correct the output results, and finally obtain the required ship model resistance data. The iterative method greatly improves the reliability of the output results.

[0046] Taking a model resistance test as an example, the method of the present invention is illustrated by obtaining the time series of test results for the stable section of the ship model resistance test through model testing.

[0047] 1) The calculated mean is R0 = 3.577 kg. Figure 3 As shown, the horizontal axis represents time, and the vertical axis represents the resistance value of the ship model.

[0048] 2) After obtaining the mean R0, the time series curve is segmented using R0, with each two zero crossings marked as a period T. The original time series is then divided into N segments. max There are 10 cycles. Preferably, the minimum value of the number of cycles N is N0.min The value is 10. This number is based on the following: approximately 10 cycles are sufficient for stable model ship motion, resulting in minimal fluctuations in calculations, and the motion time is generally long enough. This number is chosen empirically and can be adjusted appropriately. Therefore, the time series starts from T0 to T... max The resistance value fluctuates within this period. (T0 to T) max The values ​​are as follows Figure 5 As shown, the horizontal axis represents time and the vertical axis represents the resistance value. The time when the average value R0 is crossed is taken as the value of T.

[0049] 3) The time series starts counting from T0. Every two zero crossings generate a period, with a corresponding mean of A1. Therefore, the time series from T0 to T... min The mean between them is A min And so on, from T0 to T max The mean is A max .

[0050] 4) The valid result obtained is from A min To A max The value. The numerical value is as follows: Figure 6 As shown, this represents the model ship resistance calculated for each time series interval.

[0051] 5) Regarding A min To A max Find the value of A, and solve for the standard deviation and uncertainty. min To A max There are M values ​​in total, and the uncertainty of these M values ​​is:

[0052] Based on this case, the calculated value is 0.30%.

[0053] 6) Set the baseline reference value U for uncertainty. A U is preferred A =1%, the reason for this value can be obtained from the prior work of the patent applicant (Zhang Li, Chen Jianting, Chen Weimin, et al. Uncertainty analysis of composite voyage test of towing tank model resistance[J]. Shipbuilding Engineering, 2020, 42(04):38-43+126.) and experience.

[0054] 7) Compare U0 and U A In this case, the convergence condition is met, so the result can be output. If the convergence condition is not met, then A... max As the mean, the output of the time series is recalculated until the condition is met.

[0055] The method of this invention processes and evaluates data from a single ship model resistance test to obtain a set of ship model resistance data with high confidence. The overall method also avoids the influence of subjective human factors on the final result data in the prior art. It is highly objective and highly operable. Theoretically, it can be applied to single test data of any ship type. It comprehensively solves the problems of poor reliability of the current method of taking the average value of single ship model resistance test data and high cost of multiple ship model resistance tests.

[0056] It should be noted that the specific embodiments described above enable those skilled in the art to more fully understand the present invention, but do not limit the present invention in any way. Therefore, although the present invention has been described in detail with reference to the accompanying drawings and embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the present invention. In short, all technical solutions and improvements that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the present invention patent.

Claims

1. A method for evaluating time-series segmented ship model resistance test data, characterized in that, Includes the following steps: S1. Collect the resistance test data of the ship model at each time point in a finite time series during a single ship model resistance test, and calculate the average resistance value R0 of the ship model resistance test data at each time point in the finite time series. S2. Establish a coordinate system with the ship model resistance value as the vertical axis and the finite time series as the horizontal axis. Use the average resistance value R0 as the reference line to segment the finite time series. Record the first time point that crosses the R0 reference line, starting from the initial time point of the finite time series, as T0. Sequentially record the second to the last time points that cross the R0 reference line as T1, T2, ... T0. max Then, sequentially assign T0-T1, T1-T2, ..., T max-1 -T max The time series intervals are denoted as periodic segments N1, N2, ..., N max That is, the finite time series is divided into max periodic segments; S3, in T0~T max Extracting T0 to T from the time series min T0~T min+1 …, T0~T max The time series interval is defined, and the average resistance value of the time series interval is calculated, denoted as A. min A min+1 …A max The T min Defined as the starting time point for calculating the average interval resistance of a time series interval, T min Let T1, T2...T be the time points. max One of them; S4, A min A min+1 …A max The standard deviation is calculated and denoted as S0. S0 / R0 is denoted as U0, which is the Type A uncertainty of the ship model resistance data solved in this case. S5. Set the reference value for the Type A uncertainty of the ship model resistance data to U. A , if U0<U A Then A max The output is the final ship model resistance test data; S6. If U0 > U A Then use A max Replace the average value R0 of the ship model resistance test data with the value, and continue to repeat steps S1-S6 until A that meets the conditions of step S5 is obtained. max The results were used as data for the ship model resistance test.

2. The method for evaluating time-series segmented ship model resistance test data according to claim 1, characterized in that, The finite time series in step S1 is a time series in which the ship model resistance test data has reached stability, allowing the ship model resistance test data to first be judged to determine that the ship model resistance test data has reached stability.

3. The method for evaluating time-series segmented ship model resistance test data according to claim 1 or 2, characterized in that, The starting time point T used in step S3 to calculate the average interval resistance of the time series interval is... min The value of min is greater than or equal to 10.

4. The method for evaluating time-series segmented ship model resistance test data according to claim 1 or 2, characterized in that, The reference value U for the Type A uncertainty of the ship model resistance data in step S5. A =1%.

Images