CN117332510A - Ocean vessel navigational speed loss prediction method based on hybrid prediction model - Google Patents
Ocean vessel navigational speed loss prediction method based on hybrid prediction model Download PDFInfo
- Publication number
- CN117332510A CN117332510A CN202311452226.0A CN202311452226A CN117332510A CN 117332510 A CN117332510 A CN 117332510A CN 202311452226 A CN202311452226 A CN 202311452226A CN 117332510 A CN117332510 A CN 117332510A
- Authority
- CN
- China
- Prior art keywords
- speed loss
- prediction model
- vessel
- navigational
- loss
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000012549 training Methods 0.000 claims abstract description 16
- 238000013528 artificial neural network Methods 0.000 claims abstract description 15
- 239000011159 matrix material Substances 0.000 claims abstract description 14
- 238000007781 pre-processing Methods 0.000 claims abstract description 10
- 238000011425 standardization method Methods 0.000 claims abstract description 4
- 230000005540 biological transmission Effects 0.000 claims description 6
- 230000002706 hydrostatic effect Effects 0.000 claims description 6
- 238000002759 z-score normalization Methods 0.000 claims description 4
- 238000005457 optimization Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 description 9
- 238000010801 machine learning Methods 0.000 description 6
- 210000002569 neuron Anatomy 0.000 description 5
- 230000004913 activation Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000010219 correlation analysis Methods 0.000 description 3
- 230000000875 corresponding effect Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 230000008439 repair process Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 241000501754 Astronotus ocellatus Species 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000005206 flow analysis Methods 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012384 transportation and delivery Methods 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B63—SHIPS OR OTHER WATERBORNE VESSELS; RELATED EQUIPMENT
- B63B—SHIPS OR OTHER WATERBORNE VESSELS; EQUIPMENT FOR SHIPPING
- B63B71/00—Designing vessels; Predicting their performance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/15—Vehicle, aircraft or watercraft design
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Geometry (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Computer Hardware Design (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mechanical Engineering (AREA)
- Aviation & Aerospace Engineering (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Automation & Control Theory (AREA)
- Computational Mathematics (AREA)
- Combustion & Propulsion (AREA)
- Chemical & Material Sciences (AREA)
- Ocean & Marine Engineering (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a marine vessel navigational speed loss prediction method based on a hybrid prediction model, which comprises the steps of constructing a semi-empirical model, estimating the speed loss of a vessel according to the semi-empirical model, vessel power parameters and a marine vessel navigational data set, adding the estimated speed loss into the marine vessel navigational data set, preprocessing the marine vessel navigational data set based on a Z score standardization method, taking the parameters in the marine vessel navigational data set as characteristics, calculating correlation coefficients among the characteristics and constructing a correlation coefficient matrix, acquiring a characteristic set related to the speed loss according to the correlation coefficient matrix, constructing a navigational speed loss hybrid prediction model based on a multi-layer artificial neural network, acquiring a training set, training the model, optimizing the super parameters of the model based on a grid search method, and predicting the navigational speed loss of the marine vessel according to the optimized navigational speed loss hybrid prediction model. The prediction accuracy of the ocean vessel speed loss is improved based on the mixed prediction model.
Description
Technical Field
The invention relates to the technical field of ship navigation, in particular to a method for predicting the navigation speed loss of ocean vessels based on a mixed prediction model.
Background
The involuntary speed loss that occurs when the ship is sailing reduces energy efficiency. In order to increase the energy efficiency of a ship, a reliable method is often required to describe the speed loss of the ship in order to perform the ship operation. The prediction result can be used for enterprise operation management, such as component purchasing, finished product assembling and product shipping, and can effectively reduce delivery delay and stock backlog. Therefore, in the aspects of planning the voyage period, calculating the ship position, commanding the safe production, reducing the sea loss accident and the like, the correct application of stall has great significance for improving the ship operation efficiency and increasing the financial income of enterprises.
In the past, the prediction model mainly includes a Semi-empirical model (SEM) for estimating the loss of the speed and a machine learning model for predicting the loss of the speed. However, SEM has the disadvantage of poor prediction performance, while machine learning models have better prediction performance, but poor applicability, and require a large amount of historical data for modeling, and thus are not suitable for modeling new ships and vessels after dock repair.
Disclosure of Invention
The invention provides a marine vessel navigational speed loss prediction method based on a hybrid prediction model, which aims to overcome the technical problems.
A marine vessel navigational speed loss prediction method based on a mixed prediction model comprises the following steps of,
step 1, constructing a semi-empirical model for estimating the speed loss of the ocean vessel during navigation,
step 2, acquiring ocean vessel navigation data set and vessel power parameters, estimating the speed loss of the vessel according to the semi-empirical model, the vessel power parameters and the ocean vessel navigation data set, adding the estimated speed loss into the ocean vessel navigation data set, preprocessing the ocean vessel navigation data set based on a Z score standardization method,
step 3, taking parameters in ocean vessel navigation data set as characteristics, calculating correlation coefficients among the characteristics based on a pearson correlation coefficient method, constructing a correlation coefficient matrix according to the correlation coefficients among the characteristics, acquiring a characteristic set related to speed loss according to the correlation coefficient matrix,
step 4, constructing a navigational speed loss mixed prediction model based on a multi-layer artificial neural network, acquiring data corresponding to the feature set and the speed loss from the preprocessed ocean vessel navigation data set and taking the data as a training set, inputting the training set into the navigational speed loss mixed prediction model for training, optimizing the super parameters of the navigational speed loss mixed prediction model based on a grid search method, acquiring an optimized navigational speed loss mixed prediction model,
and 5, predicting the navigational speed loss of the ocean vessel according to the optimized navigational speed loss mixed prediction model.
Preferably, the constructed semi-empirical model is constructed from equation (1),
wherein: the speed loss DeltaV is the actual speed V of the ship real With hydrostatic velocity V calm Difference between P e Active power for host，P b Is the main engine braking power of the ship, eta S For shaft transmission efficiency, eta D For propulsion efficiency of the main engine, the total resistance R TOTAL ＝R CALM +R AA +R AW ，R CALM 、R AA 、R AW Respectively, hydrostatic resistance, wind resistance and wave resistance.
Preferably, the feature set related to speed loss includes five features of ground speed, heading, sense wave height, average wave direction, average wave period.
Preferably, the preprocessing of the ocean vessel voyage data set based on the Z score normalization method comprises obtaining the average number and standard deviation of each parameter, preprocessing the value of each parameter according to a formula (2),
wherein X is the original data, and the data is the original data,mean, s is standard deviation.
Preferably, optimizing the super-parameters of the navigational speed loss hybrid prediction model based on the grid search method includes determining the super-parameters and the value ranges thereof, obtaining different super-parameter combinations through exhaustive search, calculating the loss function of the navigational speed loss hybrid prediction model according to the different super-parameter combinations, and selecting the optimal super-parameter combination as the super-parameters of the navigational speed loss hybrid prediction model according to the value of the loss function.
The invention provides a method for predicting the navigational speed loss of an ocean vessel based on a hybrid prediction model, which adopts the hybrid prediction model based on a semi-empirical model, builds the semi-empirical model by considering inherent characteristics of the vessel such as the braking power of a main engine, the transmission efficiency of a shaft and the propulsion efficiency of the main engine, combines the semi-empirical model with an artificial neural network model to solve the defect that a machine learning model needs a large number of data sets to train, can train to obtain a feasible prediction model by using shipborne automatic measurement data of one voyage, and improves the generalization performance of the hybrid prediction model, so that the hybrid prediction model can be applied to new vessels with less data quantity or vessels with great changes of vessel characteristics after major repair. And the grid search method is used for optimizing the super parameters of the artificial neural network model, so that the prediction accuracy of the hybrid model is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the drawings that are needed in the embodiments or the description of the prior art will be briefly described below, it will be obvious that the drawings in the following description are some embodiments of the present invention, and that other drawings can be obtained according to these drawings without inventive effort to a person skilled in the art.
FIG. 1 is a flow chart of the method of the present invention;
FIG. 2 is a schematic diagram of data fusion in accordance with the present invention;
FIG. 3 is a thermodynamic diagram of a correlation coefficient matrix for various variables in a dataset used in the present invention;
FIG. 4 is a block diagram of an artificial neural network in a hybrid predictive model of the present invention;
FIG. 5 is a scatter plot of predicted stall values versus true stall values for a hybrid predictive model of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
FIG. 1 is a flowchart of the method of the present invention, as shown in FIG. 1, the method of the present embodiment may include:
step 1, constructing a semi-empirical model for estimating the speed loss of the ocean vessel during navigation,
step 2, acquiring ocean vessel navigation data set and vessel power parameters, estimating the speed loss of the vessel according to the semi-empirical model, the vessel power parameters and the ocean vessel navigation data set, adding the estimated speed loss into the ocean vessel navigation data set, preprocessing the ocean vessel navigation data set based on a Z score standardization method,
step 3, taking parameters in ocean vessel navigation data set as characteristics, calculating correlation coefficients among the characteristics based on a pearson correlation coefficient method, constructing a correlation coefficient matrix according to the correlation coefficients among the characteristics, acquiring a characteristic set related to speed loss according to the correlation coefficient matrix,
step 4, constructing a navigational speed loss mixed prediction model based on a multi-layer artificial neural network, acquiring data corresponding to the feature set and the speed loss from the preprocessed ocean vessel navigation data set and taking the data as a training set, inputting the training set into the navigational speed loss mixed prediction model for training, optimizing the super parameters of the navigational speed loss mixed prediction model based on a grid search method, acquiring an optimized navigational speed loss mixed prediction model,
and 5, predicting the navigational speed loss of the ocean vessel according to the optimized navigational speed loss mixed prediction model.
Based on the scheme, the ocean vessel navigational speed loss prediction method based on the hybrid prediction model is provided, the hybrid prediction model based on the semi-empirical model is adopted, the semi-empirical model is built by considering inherent characteristics of vessels such as main engine braking power, shaft transmission efficiency and main engine propulsion efficiency, meanwhile, the defect that a machine learning model needs a large number of data sets to train is overcome by combining the semi-empirical model with an artificial neural network model, the feasible prediction model can be obtained by training on-board automatic measurement data of one voyage, and the generalization performance of the hybrid prediction model is improved, so that the model can be applied to new vessels with small data quantity or vessels with great changes in characteristics after overhaul. And the grid search method is used for optimizing the hyper-parameters of the artificial neural network model, so that the prediction accuracy of the hybrid model is improved.
In particular, the present embodiment gives details of a speed loss prediction method, including,
step 1, constructing a semi-empirical model for estimating the speed loss of the ocean vessel while underway, the semi-empirical model being constructed by the formula (1),
wherein: the speed loss DeltaV is the actual speed V of the ship real With hydrostatic velocity V calm Difference between P e For the effective power of the host, P b Is the main engine braking power of the ship, eta S For shaft transmission efficiency, eta D For propulsion efficiency of the main engine, the total resistance R TOTAL ＝R CALM +R AA +R AW ，R CALM 、R AA 、R AW Respectively, hydrostatic resistance, wind resistance and wave resistance.
And 2, acquiring a ocean vessel navigation data set, wherein variables (or characteristics and fields) contained in the data set include weather information such as vessel navigation speed, pitching, rudder angle, wind, waves and currents, and the abbreviations and units of the variables are shown in a table 1. Wherein the relevant variables for vessel voyage can be obtained from deck logs and turbine logs, and the meteorological data is obtained by accessing public data sets of European weather forecast centers (European center for medium-range weather forecasts, ECMWF) and real-time ocean surface flow analysis (Ocean Surface Current Analysis Real-time, OSCAR) published by the national environmental forecast center (National centers for environmental prediction, NCEP).
Table 1 abbreviations and units for the various variables
Obtaining ship power parameters, wherein the ship power parameters comprise navigational speed, draft, host power and the like, estimating the speed loss of the ship in the ocean ship navigation data set according to a semi-empirical model, and adding the estimated speed loss to the ocean ship navigation data set, wherein stall values calculated by the semi-empirical model are taken as a series of data to be put into the data set according to time and ship position in the data set, an example of data fusion is shown in fig. 2, and the main flow is as follows:
(1) Calculating weather data under the current ship position according to the weather data in the navigation data set by using a proportional interpolation method;
(2) And aligning the calculated meteorological data and stall values into a navigation data set according to the time, longitude and latitude information to form a new data set for training a machine learning model.
Furthermore, since each feature value in the dataset has a different range of values, this will lead to a decrease in the accuracy and convergence of the model. In order to eliminate such adverse effects, the ocean vessel voyage data set is preprocessed based on a Z-score normalization method, that is, each data in the data set is normalized according to a Z-score calculation formula, the preprocessing of the ocean vessel voyage data set based on the Z-score normalization method includes obtaining an average number and a standard deviation of each parameter, preprocessing the value of each parameter according to a formula (2),
wherein X is the original data, and the data is the original data,mean, s is standard deviation.
And 3, taking parameters in ocean vessel navigation data sets as features, calculating correlation coefficients among the features based on a Pearson correlation coefficient method, constructing a correlation coefficient matrix according to the correlation coefficients among the features, and acquiring a feature set related to speed loss according to the correlation coefficient matrix, wherein the feature set related to the speed loss comprises five features including a speed to earth, a heading, a sense wave height, an average wave direction and an average wave period.
In particular, we use as input the features (parameters or variables) in the dataset. In order to reveal the relationship between stall and each sailing parameter, we calculate the correlation coefficient matrix of each feature in the dataset by adopting a correlation analysis method. The correlation analysis is to study whether the two groups of variables have linear correlation, so the precondition of the correlation analysis is to assume that the variables have linear correlation, and the obtained result is also to describe the linear correlation degree between the variables. According to the invention, the pearson correlation coefficient among the variables is calculated to form a correlation coefficient matrix among the variables, so that a plurality of variables with larger correlation with ship stall are found. Meanwhile, the multiple collinearity of the variables can be caused by the large correlation among the dependent variables, and the invention combines the navigation practice to make a choice on the variables which can cause the multiple collinearity, and only one representative variable is reserved. The specific analysis is as follows:
we found that the characteristics significantly correlated with the speed loss are (the values in brackets are correlation coefficients, near 1 is positive correlation, near-1 is negative correlation), ground speed (-0.97), heading (-0.65), sense wave height (0.72), surge height (0.76), average wave period (0.59), average wave direction (0.62), average surge period (0.54), average wave direction (0.62) and average surge direction (0.56), where ground speed and heading exhibit a high negative correlation.
In addition, we note that the average swell direction shows a high positive correlation with average swell period (0.94), average wave direction (0.97), average wave period (0.9), swell height (0.77) and sense swell height (0.77) and a negative correlation with ground speed (-0.57) and heading (-0.71). Other characteristics also present the problem of multiple collinearity, so, in order to avoid multiple collinearity, we select the speed to earth, heading, sense wave height, average wave direction, average wave period as the input variables after characteristic selection in combination with navigation practice, and the correlation coefficient matrix of each variable is shown in fig. 3.
Step 4, constructing a navigational speed loss mixed prediction model based on a multi-layer artificial neural network, acquiring data corresponding to the feature set and the speed loss from the preprocessed ocean vessel navigation data set and taking the data as a training set, inputting the training set into the navigational speed loss mixed prediction model for training, optimizing the super parameters of the navigational speed loss mixed prediction model based on a grid search method,
the grid search method-based super-parameter optimization method for the navigational speed loss hybrid prediction model comprises determining super-parameters and the value range thereof, wherein the super-parameters are shown in a table 2, different super-parameter combinations are obtained through exhaustive search, the loss function of the navigational speed loss hybrid prediction model is calculated according to the different super-parameter combinations, the optimal super-parameter combination is selected as the super-parameter of the navigational speed loss hybrid prediction model according to the value of the loss function, the super-parameter of the model of ANN is optimized by the grid search method-based super-parameter optimization method, so that navigational speed loss prediction with higher precision and higher applicability is realized,
table 2 super parameters and ranges thereof
Obtaining an optimized navigational speed loss mixed prediction model,
specifically, in this embodiment, the basic process of a neural network can be represented by the following formula, and the neural network model structure adopted in this embodiment is shown in fig. 4,
wherein w is ij Weights, x, for layer i and neuron j i B is the input value of the ith layer i For the bias weight of the ith layer, n is the total number of connected neurons, u i Is the output result of a unit or neuron in the neural network model.
In general, the layers of the neural network are generally fully connected, i.e., any neuron of the nth layer must be connected to any neuron of the n+1th layer. While ANNs appear to be large and complex as a whole, they are, in terms of microscopic local models, in fact a linear relationship plus an activation function. Mathematically, the effect of the activation function is to map the input data onto 0 to 1 (where tanh is mapped onto-1 to +1). As for the reasons of mapping, besides regularization of data, it is probably control data so that it is only within a certain range. Of course, there are also additional details, such as Sigmoid functions, that focus more on small changes in data around the center point, while ignoring changes in data at extremes. In general, sigmoid is used mostly for fully connected layers, while ReLU is used mostly for convolutional layers. The invention adopts a 3-layer hidden layer with a full-connection layer, so that a Sigmoid activation function is selected, and the calculation formula is as follows:
where x represents the input variable. As can be seen from analysis of the functional expressions and Sigmoid curves, sigmoid activation functions have the advantage that curve smoothing is everywhere conductive and therefore are often used for regression and fitting problems.
And 5, predicting the navigational speed loss of the ocean vessel according to the optimized navigational speed loss mixed prediction model.
In this embodiment, a Semi-empirical model (SEM) is used to calculate a preliminary stall prediction value, and a new data set is constructed as a new feature. Then, the inputs to the hybrid predictive model (Hybrid prediction model, HPM) consist of the ground speed, heading, sense wave height, average wave direction, average wave period 5 variables or features plus SEM calculated stall prediction value resulting from the feature selection process.
The model parameters are reselected for the new dataset. In the parameter selection stage, the network searching method is used for optimizing the super parameters of the model in the super parameter value range. The grid search method is exhaustive search, and each possibility is tried through circulation traversal in the range of the values of the super parameters, and the parameter combination with the best performance is the final super parameter combination. After the super-parameter adjustment, the prediction precision of the model is improved to a certain extent, and the accuracy of the hybrid prediction model is shown in fig. 5.
The whole beneficial effects are that:
the invention provides a method for predicting the navigational speed loss of an ocean vessel based on a hybrid prediction model, which adopts the hybrid prediction model based on a semi-empirical model, builds the semi-empirical model by considering inherent characteristics of the vessel such as the braking power of a main engine, the transmission efficiency of a shaft and the propulsion efficiency of the main engine, combines the semi-empirical model with an artificial neural network model to solve the defect that a machine learning model needs a large number of data sets to train, can train to obtain a feasible prediction model by using shipborne automatic measurement data of one voyage, and improves the generalization performance of the hybrid prediction model, so that the hybrid prediction model can be applied to new vessels with less data quantity or vessels with great changes of vessel characteristics after major repair. And the grid search method is used for optimizing the hyper-parameters of the artificial neural network model, so that the prediction accuracy of the hybrid model is improved.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention.
Claims (5)
1. A marine vessel navigational speed loss prediction method based on a mixed prediction model is characterized by comprising the following steps of,
step 1, constructing a semi-empirical model for estimating the speed loss of the ocean vessel during navigation,
step 2, acquiring ocean vessel navigation data set and vessel power parameters, estimating the speed loss of the vessel according to the semi-empirical model, the vessel power parameters and the ocean vessel navigation data set, adding the estimated speed loss into the ocean vessel navigation data set, preprocessing the ocean vessel navigation data set based on a Z score standardization method,
step 3, taking parameters in ocean vessel navigation data set as characteristics, calculating correlation coefficients among the characteristics based on a pearson correlation coefficient method, constructing a correlation coefficient matrix according to the correlation coefficients among the characteristics, acquiring a characteristic set related to speed loss according to the correlation coefficient matrix,
step 4, constructing a navigational speed loss mixed prediction model based on a multi-layer artificial neural network, acquiring data corresponding to the feature set and the speed loss from the preprocessed ocean vessel navigation data set and taking the data as a training set, inputting the training set into the navigational speed loss mixed prediction model for training, optimizing the super parameters of the navigational speed loss mixed prediction model based on a grid search method, acquiring an optimized navigational speed loss mixed prediction model,
and 5, predicting the navigational speed loss of the ocean vessel according to the optimized navigational speed loss mixed prediction model.
2. The method for predicting the navigational speed loss of the ocean vessel based on the mixed prediction model according to claim 1, wherein the constructed semi-empirical model is constructed by a formula (1),
wherein: the speed loss DeltaV is the actual speed V of the ship real With hydrostatic velocity V calm Difference between P e For the effective power of the host, P b Is the main engine braking power of the ship, eta S For shaft transmission efficiency, eta D For propulsion efficiency of the main engine, the total resistance R TOTAL ＝R CALM +R AA +R AW ，R CALM 、R AA 、R AW Respectively, hydrostatic resistance, wind resistance and wave resistance.
3. The ocean vessel navigational speed loss prediction method based on the hybrid predictive model according to claim 1, wherein the set of characteristics related to speed loss includes five characteristics of ground speed, heading, sense wave height, average wave direction, average wave period.
4. The ocean vessel voyage loss prediction method based on the mixed prediction model according to claim 1, wherein the preprocessing of the ocean vessel voyage data set based on the Z-score normalization method comprises the steps of obtaining the average number and the standard deviation of each parameter, preprocessing the value of each parameter according to a formula (2),
wherein X is the original data, and the data is the original data,mean, s is standard deviation.
5. The ocean vessel navigational speed loss prediction method based on the mixed prediction model according to claim 1, wherein the optimization of the hyper-parameters of the navigational speed loss mixed prediction model based on the grid search method comprises determining the hyper-parameters and the value ranges thereof, obtaining different hyper-parameter combinations through exhaustive search, calculating the loss function of the navigational speed loss mixed prediction model according to the different hyper-parameter combinations, and selecting the optimal hyper-parameter combination as the hyper-parameters of the navigational speed loss mixed prediction model according to the value of the loss function.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311452226.0A CN117332510A (en) | 2023-11-01 | 2023-11-01 | Ocean vessel navigational speed loss prediction method based on hybrid prediction model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311452226.0A CN117332510A (en) | 2023-11-01 | 2023-11-01 | Ocean vessel navigational speed loss prediction method based on hybrid prediction model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117332510A true CN117332510A (en) | 2024-01-02 |
Family
ID=89293254
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311452226.0A Pending CN117332510A (en) | 2023-11-01 | 2023-11-01 | Ocean vessel navigational speed loss prediction method based on hybrid prediction model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117332510A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118133697A (en) * | 2024-05-10 | 2024-06-04 | 无锡九方科技有限公司 | Application method and system of ship stall model based on ensemble learning |
-
2023
- 2023-11-01 CN CN202311452226.0A patent/CN117332510A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118133697A (en) * | 2024-05-10 | 2024-06-04 | 无锡九方科技有限公司 | Application method and system of ship stall model based on ensemble learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106599995B (en) | A method of according to maritime meteorology information prediction ship's navigation meteorological factor | |
| CN107292451B (en) | Ship speed optimization method and equipment | |
| Dalgic et al. | Investigation of optimum crew transfer vessel fleet for offshore wind farm maintenance operations | |
| Tsou et al. | An Ant Colony Algorithm for efficient ship routing | |
| CN117332510A (en) | Ocean vessel navigational speed loss prediction method based on hybrid prediction model | |
| CN107944648A (en) | A kind of accurate Forecasting Methodology of large ship speed of a ship or plane rate of fuel consumption | |
| CN112446096A (en) | Ship host engine oil consumption dynamic estimation method based on big navigation data | |
| CN112836893B (en) | Method for predicting ship oil consumption under severe sea conditions based on sea condition and ship navigation condition | |
| CN114912077B (en) | Sea wave forecasting method integrating random search and mixed decomposition error correction | |
| Wang et al. | Sea-water-level prediction via combined wavelet decomposition, neuro-fuzzy and neural networks using SLA and wind information | |
| CN112182747A (en) | Simulation test system and method for ship wave drag performance analysis | |
| CN115204058B (en) | Ground wave radar flow field calculation method and device based on BP neural network | |
| CN113507118B (en) | Wind power prediction method and system | |
| CN110778398B (en) | Marine diesel engine fuel management control system | |
| CN112488208A (en) | Method for acquiring remaining life of island pillar insulator | |
| CN108832623B (en) | Physical-statistical hybrid two-stage wind power prediction method | |
| CN111488974B (en) | Ocean wind energy downscaling method based on deep learning neural network | |
| Lin et al. | Research on the Speed Optimization Model Based on BP Neural Network and Genetic Algorithm (GA) | |
| Lei et al. | Prediction of main engine speed and fuel consumption of inland ships based on deep learning | |
| Huu et al. | A real-time model using interpolation method and semi-empirical formulas to estimate fuel consumption for cargo ships | |
| Li et al. | Prediction of ship fuel consumption based on Elastic network regression model | |
| CN105550495A (en) | Method for evaluating ship navigation increasing resistance level | |
| Krata et al. | The Development of a Combined Method to Quickly Assess Ship Speed and Fuel Consumption at Different Powertrain Load and Sea Conditions | |
| CN117421601B (en) | Sea surface evaporation waveguide near-future rapid forecasting method | |
| Zhang et al. | A Combination Forecasting Model Based on AdaBoost_GRNN in Depth-Averaged Currents Using Underwater Gliders |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |