Ship body heaving motion prediction analysis method and system based on time sequence analysis

Abstract

The invention discloses a ship body heaving motion prediction analysis method and system based on time sequence analysis. The method comprises the steps of obtaining original data of ship heaving displacement; constructing a periodic autoregression model based on the original data and a time sequence analysis method; obtaining a random error according to the periodic autoregression model; based on the random error, constructing a harmonic steady-state mean regression model; based on the periodic autoregression model and the harmonic steady-state mean value regression model, constructing a periodic-harmonic steady-state mean value regression model; and inputting the original data into a periodic-harmonic steady-state mean value regression model to obtain a predicted value of the ship deep sinking motion. According to the method, the characteristics of randomly distributed peaks, thick tails, biases, fluctuation gathering and the like in the ship heaving motion can be described more accurately.

Description

Technical Field

[0001] The present invention belongs to the technical field of periodic-harmonic steady-state mean regression model analysis, and in particular to a method and system for predicting and analyzing the heave and sink motion of a ship hull based on time series analysis. Background Art

[0002] The prediction and analysis of ship heave motion is not only related to the efficiency of offshore operations, but also affects the operational safety of ship shipping. At present, the ship heave compensation device mainly uses the real-time heave motion signal of the hull as the given signal and the real-time compensation motion signal of the compensation system actuator as the feedback signal, so as to control the actuator to reversely track the hull heave motion in real time. In the actual heave compensation system, the position compensation efficiency can basically reach 85%-90%. The main limiting factor that prevents the compensation accuracy from being further improved is that there is a significant lag between the hull heave motion signal detected by the inertial attitude sensor and the actual movement of the hull, which seriously affects the system performance and stability of the ship heave compensation device. Therefore, by making short-term accurate predictions of the future values ​​of the ship heave motion to compensate for this lag effect, the control performance of the ship heave compensation device can be effectively improved.

[0003] At present, the prediction and analysis of ship heave motion is mostly based on the autoregressive model analysis in the time series analysis method. However, with the continuous expansion of the sea area where ships operate and the increasing distance from the shore, the complexity of ship motion under the influence of wind and wave changes at sea has also increased, making it difficult for traditional autoregressive model analysis to meet the prediction requirements. In order to meet the needs of actual ship operations and ensure the accuracy of ship heave compensation, a real-time and accurate ship motion prediction method is essential. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a method and system for predicting and analyzing the heave and sink motion of a ship hull based on time series analysis to solve the problems existing in the prior art.

[0005] The present invention provides a method for predicting and analyzing the heave and sink motion of a ship hull based on time series analysis, comprising:

[0006] Obtain the original data of ship heave displacement;

[0007] Based on the original data, combined with the time series analysis method, a periodic autoregressive model is constructed;

[0008] According to the periodic autoregressive model, obtaining a random error;

[0009] Based on the random error, a harmonic steady-state mean regression model is constructed;

[0010] Based on the cyclical autoregressive model and the harmonic steady-state mean regressive model, a cyclical-harmonic steady-state mean regressive model is constructed;

[0011] The raw data are input into a periodic-harmonic steady-state mean regression model to obtain the predicted value of the ship's deep motion.

[0012] Optionally, based on the original data and in combination with a time series analysis method, constructing a periodic autoregressive model includes:

[0013] Taking the ship lifting displacement as the dependent variable and combining the time series analysis method, an initial periodic autoregressive model is constructed;

[0014] Based on the original data and in combination with the least square method in the time series analysis method, the model parameters in the initial periodic autoregressive model are estimated to obtain the periodic autoregressive model.

[0015] Alternatively, the method to construct the initial cyclical autoregressive model is:

[0016] Y(t)=Λ(t)+T(t)

[0017] Among them, Y(t) is the data of ship heave displacement, Λ(t) is a periodic or trend function, and T(t) is a random error.

[0018] Optionally, the method to obtain a periodic autoregressive model is:

[0019]

[0020] in, is the estimated value of the ship's heave displacement data, and are the estimated values ​​of the model parameters.

[0021] Optionally, constructing a harmonic steady-state mean regression model based on the random error includes:

[0022] Taking random error as dependent variable and combining time series analysis method, an initial harmonic steady-state mean regression model is constructed;

[0023] Based on the random error, the least square method in the time series analysis method is used to estimate the model parameters in the initial harmonic steady-state mean regression model to obtain the harmonic steady-state mean regression model.

[0024] Optionally, the method for obtaining the harmonic steady-state mean reversion model is:

[0025]

[0026] in, is the estimated value of the volatility function, and are the estimated values ​​of the model parameters.

[0027] Optionally, the method for constructing the cyclical-harmonic steady-state mean reversion model is:

[0028]

[0029] The present invention also provides a ship hull heave motion prediction and analysis system based on time series analysis, which is characterized by comprising: a data acquisition module, a model building module and a data prediction module;

[0030] A data acquisition module is used to obtain the original data of the ship's heave and displacement;

[0031] The model building module is used to build a cyclical-harmonic steady-state mean regression model based on the cyclical autoregressive model and the harmonic steady-state mean regression model;

[0032] The data prediction module is used to input the original data into a periodic-harmonic steady-state mean regression model to obtain a predicted value of the ship's deep motion.

[0033] Compared with the prior art, the present invention has the following advantages and technical effects:

[0034] The present invention is based on the periodic mean regression and harmonic steady-state mean regression models, which can more accurately characterize the characteristics of ship heave motion, improve the prediction accuracy of ship heave motion, and optimize the system performance of a ship heave compensation device. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0036] Figure 1 It is a flow chart of a method for predicting and analyzing the heave and sink motion of a ship hull based on time series analysis according to an embodiment of the present invention. DETAILED DESCRIPTION

[0037] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0038] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0039] The present invention proposes a method for predicting and analyzing the heave and sinking motion of a ship based on time series analysis. Figure 1As shown, the specific steps include:

[0040] Obtain the original data of ship heave displacement;

[0041] Based on the original data and combined with the time series analysis method, a periodic autoregressive model is constructed;

[0042] According to the periodic autoregressive model, random errors are obtained;

[0043] Based on random errors, a harmonic steady-state mean regression model is constructed;

[0044] Based on the cyclical autoregressive model and the harmonic steady-state mean reversion model, a cyclical-harmonic steady-state mean reversion model is constructed;

[0045] The raw data are input into the cyclical-harmonic steady-state mean regression model to obtain the predicted value of the ship's deep motion.

[0046] Specifically, it includes: S1, obtaining the original data of the ship's heave displacement, and dividing the original data into training data and verification data;

[0047] S2. Taking the ship heave displacement as the dependent variable, a periodic autoregressive model is constructed based on the time series analysis method;

[0048] S3. Based on the training data, the least square method in the time series analysis method is used to estimate the model parameters in the periodic autoregressive model, determine the form of the periodic autoregressive model, and estimate the random error;

[0049] S4. Taking random error as the dependent variable, a harmonic steady-state mean regression model is constructed;

[0050] S5. Based on the random error results, the least squares estimation method is used to estimate the model parameters in the harmonic steady-state mean regression model and determine the model form of the harmonic steady-state mean regression model;

[0051] S6. Based on the periodic autoregressive form of the above training data and the mean regressive model form of the random error, a periodic-harmonic steady-state mean regressive model is constructed;

[0052] S7. Substitute the verification data into the constructed periodic-harmonic steady-state mean regression model to perform predictive analysis on the ship's heave motion to obtain analysis results.

[0053] Furthermore, based on the original data, a periodic autoregressive model is constructed including:

[0054] Taking the ship lifting displacement as the dependent variable, an initial periodic autoregressive model is constructed based on the time series analysis method.

[0055] Based on the original data and combined with the least square method in the time series analysis method, the model parameters in the initial cyclical autoregressive model are estimated to obtain the cyclical autoregressive model.

[0056] Specifically, in S2, the periodic mean regression model with ship heave displacement as the dependent variable Y(t) is:

[0057] Y(t)=Λ(t)+T(t) (1)

[0058] Where T(t) is the random error term of the model, and Λ(t) is a periodic or trend function and has the following form:

[0059]

[0060] Here, a0, b0, a i 、b i , T1, I1 and I2 are unknown parameters describing the trend of propagation heave displacement, and T(t) represents the random error result.

[0061] Furthermore, based on the training data {Y(t k )}, k = 1, ..., n, use the least squares method to estimate the periodic autoregressive model formula (1) and obtain the parameters that minimize the random error term Estimated value of Right now:

[0062]

[0063] The parameter estimates Substituting into formula (2), we get the estimated result of Y(t) Furthermore, using the relation Obtain the estimated result of the random error term T(t)

[0064] Specifically, the estimated result of Y(t) is for:

[0065]

[0066] in, is the estimated value of the ship's heave displacement data, and are the estimated values ​​of the model parameters.

[0067] Furthermore, based on random errors, the construction of a harmonic steady-state mean regression model includes:

[0068] Taking random error as the dependent variable, an initial harmonic steady-state mean regression model is constructed;

[0069] Based on random errors, the least squares method is used to estimate the model parameters in the initial harmonic steady-state mean regression model to obtain the harmonic steady-state mean regression model.

[0070] Specifically, the random error results are used as the dependent variable to construct a mean reversion model driven by harmonic steady-state noise;

[0071] In S4, the mean reversion process driven by the harmonic steady-state noise to which the random error result T(t) obeys is:

[0072]

[0073] Here, σ(t) has the following form:

[0074]

[0075] Among them, c0, c j d j , T2, J1 and J2 are model parameters. v(t) = dL(t) = L(t) - L(t-1) obeys the harmonic steady-state distribution TS α (a, b, m), its distribution function is:

[0076]

[0077] Among them, a∈(0,1), b>0, and m∈R are unknown parameters in the distribution function.

[0078] Furthermore, the method to obtain the harmonic steady-state mean reversion model is:

[0079]

[0080] in, is the estimated value of the volatility function, and are the estimated values ​​of the model parameters.

[0081] Calculate the estimate of v(t):

[0082]

[0083] Estimation results based on v(t) Calculate its empirical characteristic function value:

[0084]

[0085] Among them, 1{v(t j )≤x} is the indicator function, that is, when v(tj)≤x, it takes a value of 1, otherwise it takes a value of 0. Using the least squares method, the parameter θ=(α,a,b,m) that minimizes the difference between the empirical distribution function value and the theoretical distribution function value is estimated, that is:

[0086]

[0087] Get the estimated value of θ = (α, a, b, m) Determine the approximate distribution of v(t)

[0088] Furthermore, the method of constructing a cyclical-harmonic steady-state mean reversion model is:

[0089]

[0090] Moreover, v(t) = dL(t) obeys the harmonic steady-state distribution

[0091] S7. Substitute the verification data into the constructed periodic-harmonic steady-state mean regression model to perform predictive analysis on the ship's heave motion to obtain analysis results.

[0092] In S8, based on the analysis results, as shown in Table 1:

[0093] Table 1

[0094] Cyclic-Harmonic Steady-State Mean Reversion Model Analysis Classical autoregressive model analysis <![CDATA[Goodness of fit R 2 0.95]]> <![CDATA[Goodness-of-fit R 2 0.75]]> Error 0.2 Error 2.2

[0095] The conclusion is drawn from the analysis data. The prediction results of the hull heave motion using the cyclical-harmonic steady-state mean regression model are better than those in the original analysis. Therefore, it can be considered that the prediction analysis method based on the cyclical-harmonic steady-state mean regression model is more effective.

[0096] The present invention also provides a ship hull heave motion prediction and analysis system based on time series analysis, which is characterized by comprising: a data acquisition module, a model building module and a data prediction module;

[0097] A data acquisition module is used to obtain the original data of the ship's heave and displacement;

[0098] A model building module, used for building a cyclical-harmonic steady-state mean reversion model based on a cyclical autoreversion model and a harmonic steady-state mean reversion model;

[0099] The data prediction module is used to input the original data into the periodic-harmonic steady-state mean regression model to obtain the predicted value of the ship's deep motion.

[0100] The present invention is a prediction and analysis method that integrates the periodic mean regression and the harmonic steady-state mean regression model. It improves the accuracy of prediction results for variables with certain randomness and periodicity. It supports professional test personnel to obtain more accurate prediction results. Taking the prediction and analysis problem of ship heave motion as an example, the periodic mean regression and the harmonic steady-state mean regression model are integrated to accurately characterize the periodic characteristics of the ship heave motion under the interference of random factors, thereby obtaining a more accurate prediction result of the ship heave motion and improving the system performance of the ship compensation device. At the same time, it lays a solid foundation for supporting the development of future task-oriented complex system / system simulation tests.

[0101] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A ship hull heave motion prediction and analysis method based on time series analysis, characterized in that: include: Obtain the original data of ship heave displacement; Based on the original data, combined with the time series analysis method, a periodic autoregressive model is constructed; According to the periodic autoregressive model, obtaining a random error; Based on the random error, a harmonic steady-state mean regression model is constructed; Based on the cyclical autoregressive model and the harmonic steady-state mean regressive model, a cyclical-harmonic steady-state mean regressive model is constructed; The raw data are input into a periodic-harmonic steady-state mean regression model to obtain the predicted value of the ship's deep motion.

2. The method for predicting and analyzing the heave and sinking motion of a ship based on time series analysis according to claim 1, characterized in that: Based on the original data and combined with the time series analysis method, a periodic autoregressive model is constructed, including: Taking the ship lifting displacement as the dependent variable and combining the time series analysis method, an initial periodic autoregressive model is constructed; Based on the original data and in combination with the least square method in the time series analysis method, the model parameters in the initial periodic autoregressive model are estimated to obtain the periodic autoregressive model.

3. The method for predicting and analyzing the heave motion of a ship hull based on time series analysis according to claim 2 is characterized in that: The method to construct the initial cyclical autoregressive model is: Y(t)=Λ(t)+T(t) Among them, Y(t) is the data of ship heave displacement, Λ(t) is a periodic or trend function, and T(t) is a random error.

4. The method for predicting and analyzing the heave motion of a ship based on time series analysis according to claim 2 is characterized in that: The method to obtain the periodic autoregressive model is: in, is the estimated value of the ship's heave displacement data, and are the estimated values ​​of the model parameters.

5. The method for predicting and analyzing the heave motion of a ship based on time series analysis according to claim 4 is characterized in that: Based on the random error, constructing a harmonic steady-state mean regression model includes: Taking random error as dependent variable and combining time series analysis method, an initial harmonic steady-state mean regression model is constructed; Based on the random error, the least square method in the time series analysis method is used to estimate the model parameters in the initial harmonic steady-state mean regression model to obtain the harmonic steady-state mean regression model.

6. The method for predicting and analyzing the heave motion of a ship based on time series analysis according to claim 5, characterized in that: The method for obtaining the harmonic steady-state mean reversion model is: in, is the estimated value of the volatility function, and are the estimated values ​​of the model parameters.

7. The method for predicting and analyzing the heave motion of a ship based on time series analysis according to claim 6, characterized in that: The method for constructing the cyclical-harmonic steady-state mean reversion model is:

8. The ship hull heave motion prediction and analysis system based on time series analysis is characterized by: include: Data acquisition module, model building module and data prediction module; A data acquisition module is used to obtain the original data of the ship's heave and displacement; The model building module is used to build a cyclical-harmonic steady-state mean regression model based on the cyclical autoregressive model and the harmonic steady-state mean regression model; The data prediction module is used to input the original data into a periodic-harmonic steady-state mean regression model to obtain a predicted value of the ship's deep motion.

Images