CN113868765B - Ship main scale parameter optimization method based on approximate model - Google Patents
Ship main scale parameter optimization method based on approximate model Download PDFInfo
- Publication number
- CN113868765B CN113868765B CN202111142269.XA CN202111142269A CN113868765B CN 113868765 B CN113868765 B CN 113868765B CN 202111142269 A CN202111142269 A CN 202111142269A CN 113868765 B CN113868765 B CN 113868765B
- Authority
- CN
- China
- Prior art keywords
- model
- ship
- main scale
- resistance
- parameters
- 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.)
- Active
Links
Classifications
-
- 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]
- G06F30/10—Geometric CAD
- G06F30/17—Mechanical parametric or variational 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Geometry (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Computational Mathematics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Aviation & Aerospace Engineering (AREA)
- Automation & Control Theory (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Aerodynamic Tests, Hydrodynamic Tests, Wind Tunnels, And Water Tanks (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a ship main scale parameter optimization method based on an approximate model, which comprises the following steps of S1, constructing a ship rapidness model test database; s2, selecting an approximate model type, establishing a corresponding approximate model function related to the mapping relation of the main scale, the ship type parameter and the residual resistance coefficient, and carrying out error analysis and validity check on the approximate model function; s3, selecting a group of ship main scale and ship model parameter data except a database as input, forecasting resistance coefficients based on each approximate model, comparing the resistance coefficients with corresponding test values, and verifying the accuracy of forecasting resistance of the built approximate model; s4, optimizing the main scale parameters by using a single-target optimization algorithm of the self-adaptive simulated annealing algorithm and taking the minimum resistance of the full-speed section as a target based on the approximate model. The invention is based on utilizing ASA single-target optimization algorithm, and can quickly realize the optimization selection of the main scale parameters of the ship by taking the minimum resistance of the full-speed section as the target.
Description
Technical Field
The invention belongs to the technical field of optimization of main scale parameters of ships, and particularly relates to a method for supporting demonstration optimization of main scale and ship type parameters of ships by using an optimization algorithm based on an approximation model and taking the minimum resistance of a full-speed section as a target.
Background
In the demonstration design process, the modern ship overall design provides higher requirements for model line design optimization, and according to the overall scheme requirements, a plurality of iterative designs can be developed on the basis of a mother ship with excellent performance, a large number of different main scale and ship type parameter schemes are generated, the rapidity of the ship type schemes is required to be evaluated through a large number of simulation calculations, the analysis workload is large, the timeliness is poor, the requirement of rapid design is difficult to adapt, and great inconvenience is brought to finding the influence rule and the model line optimization direction of model line parameter variation.
Disclosure of Invention
Aiming at the defect of low optimization efficiency of the ship main scale parameters in the prior art, the invention provides a ship main scale parameter optimization method based on an approximate model, which establishes an approximate model of main scale, ship type parameters and residual resistance coefficients based on sample data with excellent performance, realizes quick and accurate forecasting of the resistance of a full-speed section, provides an efficient response capability for forecasting ship resistance with a ship type characteristic relatively close to the ship type characteristic, establishes a high-precision analysis model for corresponding main scale optimization, and further realizes quick optimization selection of the ship main scale parameters by using a single-target optimization algorithm with minimum resistance of the full-speed section as a target.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a ship main scale parameter optimization method based on an approximate model comprises the following steps:
s1, constructing a ship rapidness model test database: screening main hull resistance test data of different ship types under various drainage capacities, selecting main scale, ship type parameters and rapidness parameters which mainly influence resistance, and constructing a main scale, ship type parameters and rapidness parameter database;
s2, constructing an approximate model: selecting an approximate model type based on the database sample points in the S1, establishing a corresponding approximate model function related to the mapping relation of the main scale, the ship type parameters and the residual resistance coefficients, and carrying out error analysis and validity check on the approximate model function;
s3, forecasting resistance: selecting a group of ship main scale and ship type parameter data outside the database as input, calculating friction resistance by adopting a Prandtl-Schlichting formula, forecasting resistance coefficients based on each approximate model established in S2, comparing with corresponding test values, and verifying the accuracy of forecasting resistance of the established approximate model; finally, selecting a model with a more stable error range and a smaller error mean value as an approximate model for rapidly forecasting the ship resistance;
s4, automatically optimizing the main scale parameters of the ship by using an optimization algorithm: based on the finally selected approximate model in the step S3, optimizing the main scale parameter by using a single-target optimization algorithm of a self-adaptive simulated annealing algorithm (ASA) and taking the minimum resistance of the full-speed section as a target.
In the above scheme, in step S1, the main dimensions mainly affecting the resistance include the waterline Length, the water line width Beam, the draft T, the drainage volume V, the main hull wet surface area S, the ship-type parameters include the square coefficient Cb, the mid-cross section coefficient Cm, and the rapidity parameters include the navigational speed Vs and the residual resistance coefficient 1000Cr.
In the above scheme, in step S1, when the database is constructed, each parameter is distributed uniformly as much as possible in a reasonable range, and the upper and lower boundaries are as large as possible.
In the above scheme, in step S1, when the database is constructed, a ship type scheme with similar ship type characteristics is selected.
In the above scheme, in step S2, the approximation model includes a response surface model, where the response surface model uses Length, beam, T, V, S, cb, cm, vs as an input variable and 1000Cr as an output variable; corresponding approximation model functionThe method comprises the following steps:
wherein x is i 、x j Designing sample points for the ith and jth, wherein k is the number of the sample points and alpha is 0 、α i 、α ij 、α ii 、α iii 、α iiii Is a fourth order polynomial coefficient to be determined.
In the above scheme, in step S2, the approximation model includes a radial basis model, where the radial basis model includes an input layer, an hidden layer, and an output layer, the input layer and the hidden layer exhibit nonlinear transformation, the hidden layer and the output layer exhibit linear transformation relationships, and a Radial Basis Function (RBF) is a learning function from the neural network input layer to the hidden layer, and is a typical nonlinear function; the corresponding approximation model function is:
r k ＝R(||x-T k ||)
wherein x= [ x ] 1 ,x 2 ,…,x n ]Represents an n-dimensional input vector, i is an euclidean norm, T k Is the center of the kth hidden node (k=1, 2, N T )，N T To train the number of sample points, R () is an RBF function, R k Is an output distance function.
In the above scheme, in step S2, when performing the approximate model function error analysis and the validity check, the root mean square R and the determination coefficient error R are adopted 2 Checking the precision of the model, wherein R-0 represents small error of the model, high approximation precision and R 2 And 1, the similarity degree of the model and the original model is high.
In the above scheme, in step S3, the Prandtl-Schlichting formula is:
wherein Re is the Reynolds number,l is the characteristic length, V is the characteristic speed, and V is the dynamic viscosity coefficient.
In the above scheme, in step S4, the objective function of the optimization algorithm is selected as the sum of the remaining resistance coefficients of the full-speed segment, and the optimized mathematical model is as follows:
Min[SUM(Cr1,Cr2,…,Crn)]
in the optimization process, under the condition of meeting the main scale constraint, the minimum objective function is the optimal solution, and the corresponding main scale parameters are the optimal main scale parameter combination.
In the above scheme, in step S4, the optimization procedure of the main scale parameter specifically includes the following steps:
s4.1, providing a primary scale of the ship and an initial value of a ship type design parameter by adopting an ASA optimization algorithm;
s4.2, calculating the residual resistance coefficient of each navigational speed based on the design parameters in the step S4.1 by using the approximate model between the main scale, the ship type parameters and the residual resistance coefficient constructed by the method to obtain a target function value;
s4.3, evaluating target fitness according to the objective function value, updating design parameters according to a fitness evaluation result by an optimization algorithm, and iterating S4.3-S4.1-S4.2-S4.3 …; the optimization flow termination criteria is typically determined by the maximum number of iteration steps or by reaching an optimal solution (minimum objective function).
The invention has the beneficial effects that:
according to the invention, an approximate model is established based on the mapping relation between the abstract main scale, the ship type parameter and the residual resistance coefficient based on the ship resistance test database, and the model is used for resistance prediction, so that compared with numerical calculation, the workload and time of resistance prediction are greatly reduced; based on the approximate model, the ASA single-target optimization algorithm is utilized, and the ship main scale parameter optimization selection can be rapidly realized by taking the minimum resistance of the full-speed section as the target.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a graph comparing RSM model fitting values with an original model in an embodiment of the present invention;
FIG. 2 is a graph showing the RBF model fitting values versus the original model in an embodiment of the present invention;
FIG. 3 is a graph showing comparison of resistance errors of a real ship predicted by an RSM model and an RBF model in an embodiment of the invention;
FIG. 4 is a schematic view of an optimization flow of the main scale parameters of the ship according to the present invention;
FIG. 5 is a schematic diagram of an objective function optimization process for drag in a full range in an embodiment of the invention.
Detailed Description
For a clearer understanding of technical features, objects and effects of the present invention, a detailed description of embodiments of the present invention will be made with reference to the accompanying drawings.
The invention provides a ship main scale parameter optimization method based on an approximate model, which takes a single square stern ship (the displacement is 3000-10000 t) as a research object.
S1, constructing a ship rapidness model test database: and screening main hull resistance test data of different ship types under various drainage capacities, and selecting main scale, ship type parameters and rapidness parameters which mainly influence resistance, wherein the main scale, the ship type parameters and the rapidness parameters are specifically a waterline Length, a water line width Beam, a draft T, a drainage volume V, a main hull wet surface area S, a square coefficient Cb, a middle cross section coefficient Cm, a navigational speed Vs and a residual resistance coefficient 1000Cr, and a main scale, ship type parameters and rapidness parameter database is constructed.
When the database is built, the parameters are uniformly distributed as much as possible in a reasonable range, and the upper and lower boundaries are as large as possible, because after an approximate model is built based on the database, the parameters cannot exceed the parameter range of the database when the corresponding resistance of a certain main scale and ship type parameters is predicted, otherwise, extrapolation is caused, and calculation is not converged. In addition, in order to improve the prediction precision, a ship type scheme with similar ship type characteristics is selected as much as possible when a database is constructed, so that the abrupt change of the ship type characteristics and the low fitting precision of an approximate model are avoided.
The database construction format is as follows:
| length | Beam | T | V | S | Cb | Cm | Vs | 1000Cr |
| L1 | B1 | T1 | V1 | S1 | Cb1 | Cm1 | Vs1 | 1000Cr1 |
| L2 | B2 | T2 | V2 | S2 | Cb2 | Cm2 | Vs2 | 1000Cr2 |
| … | … | … | … | … | … | … | … | … |
s2, constructing an approximate model: and (3) selecting an approximate model type based on the database sample points in the step (S1), establishing a corresponding approximate model function related to the mapping relation of the main scale, the ship type parameters and the residual resistance coefficients, and carrying out error analysis and validity test on the approximate model function.
The process of establishing the approximate model mainly comprises the following steps:
1) Selecting a sample point input;
2) Selecting a model function type to express and simulate the sample data;
3) And (5) checking the validity.
In this embodiment, two approximation models are created for comparison, namely a fourth order Response Surface Model (RSM) and a radial basis model (RBF).
(a) RSM model
The response surface model adopts a nonlinear polynomial with higher fourth-order accuracy as a response surface approximation function, and a plurality of groups of sample points and output values are used for creating the response surface model by fitting, regression and other methods. For the embodiment, a constructed sample point database is selected, length, beam, T, V, S, cb, cm, vs is an input variable, 1000Cr is an output variable, and for the 8 variables selected, at least 61 sample points are needed for constructing a fourth-order response surface model, and 162 sample points in the constructed database meet the requirements.
The RSM approximation function is as follows:
in the method, in the process of the invention,as a response function x i 、x j Designing sample points for the ith and jth, wherein k is the number of the sample points and alpha is 0 、α i 、α ij 、α ii 、α iii 、α iiii Is a fourth order polynomial coefficient to be determined.
Judging the effectiveness of the fitted response surface model, wherein the common error index has the maximumAbsolute value error, sum of squares error, root mean square error, and determination coefficient error R 2 Etc., the present embodiment uses the root mean square R and the determination coefficient error R 2 Checking the precision of the model, wherein R-0 represents that the RSM model has small error, R 2 And 1, the RSM model and the original model are high in similarity.
In the RSM model constructed in this example, r=0.0224, R 2 ＝0.9976。
Randomly selecting 4 sample points in a database for verification, wherein the sample point data comprises main scale and ship type parameter input values and 1000Cr test values, fitting the sample points based on a constructed response surface model, and the result shows that the fitting value of the RSM model and the 1000Cr test value corresponding to the sample points are shown in figure 1, and R, R is combined 2 The RSM model constructed in the method has higher precision.
(b) RBF model
The Radial Basis Function (RBF) is a learning function from the input layer to the hidden layer of the neural network, and is a typical nonlinear function. The corresponding approximation model function is:
r k ＝R(||x-T k ||)
wherein x= [ x ] 1 ,x 2 ,…,x n ]Representing an n-dimensional input vector, |, the term "T" is an European norm k Is the center of the kth hidden node (k=1, 2, N T )，N T To train the number of sample points, R () is an RBF function, R k Is an output distance function.
The validity of the fitted radial basis model is determined, again using the root mean square R and the decision coefficient error (R 2 ) Checking the precision of the model, wherein R-0 represents that the RSM model has small error, R 2 And 1, the RSM model and the original model are high in similarity.
In the RBF model constructed in this embodiment, r=0.00211, R 2 ＝0.99996。
Randomly selecting 5 sample points for verification, wherein the sample points comprise main scale and ship type parameter input values andand (2) fitting the sample points based on the constructed radial basis model to obtain 1000Cr test values, wherein the result shows that the RBF model fitting values correspond to the test values 1000Cr pairs of the sample points, such as shown in FIG. 2. Bond R, R 2 The RBF model constructed in this embodiment is higher than the RSM model.
S3, forecasting resistance: selecting a group of ship main scale and ship type parameter data outside the database as input, calculating friction resistance by adopting a Prandtl-Schlichting formula, forecasting resistance coefficients based on each approximate model established in S2, comparing with corresponding test values, and verifying the accuracy of forecasting resistance of the established approximate model; and finally, selecting a model with a more stable error range and a smaller error mean value as an approximate model for rapidly forecasting the ship resistance.
The Prandtl-Schlichting formula is:
wherein Re is the Reynolds number,l is the characteristic length, V is the characteristic speed, and V is the dynamic viscosity coefficient.
The following table shows the predicted drag coefficient and the actual ship drag error value.
Because the absolute value of the residual resistance coefficient is small, the maximum Cr error of the RSM model is 12 percent, and the maximum Cr error of the RBF model is 7 percent; the prediction error of the RSM model is about 5% and the prediction error of the RBF model is about 3%.
As shown in FIG. 3, in the full-speed section, the error range of the RBF model is more stable, the mean value is smaller than that of the RSM model, and compared with the real ship resistance test value, the error is controlled within 3%. Therefore, the RBF model is selected as an approximate model for rapidly forecasting the ship resistance.
S4, automatically optimizing the main scale parameters of the ship by using an optimization algorithm: based on the finally selected approximate model in the step S3, a single-target optimization algorithm of a self-adaptive simulated annealing algorithm is utilized, and the main scale parameter is optimized by taking the minimum full navigational resistance as a target.
In this embodiment, based on the established RBF approximation model, an adaptive simulated annealing algorithm (ASA) single-target optimization algorithm is utilized, and the objective function is selected as the sum of the residual resistance coefficients of the full-speed section (10-32 kn). In the optimization process, under the condition of meeting the main scale constraint, the minimum objective function is the optimal solution, the corresponding main scale parameters are the optimal main scale parameter combination, and the rapid and beneficial support can be provided for the main scale demonstration work of the ship overall design through the optimization process. The optimized mathematical model is as follows:
Min[SUM(Cr1,Cr2,…,Crn)]
as shown in fig. 4, the optimization flow specifically includes the following steps:
s4.1, providing a primary scale of the ship and an initial value of a ship type design parameter by adopting an ASA optimization algorithm;
s4.2, calculating the residual resistance coefficient of each navigational speed based on the parameters in the step S4.1 by using the approximate model between the main scale, the ship type parameters and the residual resistance coefficient constructed by the method to obtain a target function value;
and S4.3, evaluating target fitness according to the objective function value, updating design parameters according to the evaluation result by an optimization algorithm, and iterating S4.3-S4.1-S4.2-S4.3 …. The optimization flow termination criteria is typically determined by the maximum number of iteration steps or by reaching an optimal solution (minimum objective function).
In the optimization process of the embodiment, as shown in fig. 5, as the iteration steps increase, the objective function tends to converge, approaches to the minimum value, reaches the optimization target, and finds the main scale and ship type parameter combination with the minimum resistance of the full-speed section by optimizing, thereby providing rapid and effective support for the main scale demonstration of the ship.
The method of the invention has the following advantages:
1. a ship rapidness model test database is established, sample data with excellent performance is selected, two approximate models of mapping relation between main scale, ship type parameters and residual resistance coefficients are established, and the accuracy of model fitting errors is over 99 percent; and a better approximation model is selected through verification.
2. The method takes the main scale and the ship type parameters of the ship as input, realizes the quick prediction of the main hull resistance of the full-speed section based on the approximate model, has the error of about 3 percent compared with the actual ship resistance test value, greatly saves solving time compared with CFD numerical calculation, and provides high-efficiency response capability for the ship resistance prediction with a similar ship type characteristic.
3. Based on the established approximate model, the self-adaptive simulated annealing optimization algorithm is utilized to optimize the main scale parameters by taking the minimum resistance of the full navigational speed section as a target, and reasonable main scale parameters are effectively and objectively selected.
The embodiments of the present invention have been described above with reference to the accompanying drawings, but the present invention is not limited to the above-described embodiments, which are merely illustrative and not restrictive, and many forms may be made by those having ordinary skill in the art without departing from the spirit of the present invention and the scope of the claims, which are to be protected by the present invention.
Claims (7)
1. The ship main scale parameter optimization method based on the approximate model is characterized by comprising the following steps of:
s1, constructing a ship rapidness model test database: screening main hull resistance test data of different ship types under various drainage capacities, selecting main scale, ship type parameters and rapidness parameters which mainly influence resistance, and constructing a main scale, ship type parameters and rapidness parameter database; the main scale mainly influencing the resistance comprises a waterline Length, a water line width Beam, a draft T, a drainage volume V and a main hull wet surface area S, the ship type parameters comprise a square coefficient Cb, a middle cross section coefficient Cm and a rapidness parameter comprise a navigational speed Vs and a residual resistance coefficient 1000Cr;
s2, constructing an approximate model: selecting an approximate model type based on the database sample points in the S1, establishing a corresponding approximate model function related to the mapping relation of the main scale, the ship type parameters and the residual resistance coefficients, and carrying out error analysis and validity check on the approximate model function;
the approximation model includes a response surface model and a radial basis model:
the response surface model takes Length, beam, T, V, S, cb, cm, vs as an input variable and 1000Cr as an output variable; corresponding approximation model functionThe method comprises the following steps:
wherein x is i 、x j Designing sample points for the ith and jth, wherein k is the number of the sample points and alpha is 0 、α i 、α ij 、α ii 、α iii 、α iiii Is a fourth order polynomial coefficient to be determined,
the radial basis model comprises an input layer, an hidden layer and an output layer, wherein the input layer and the hidden layer show nonlinear transformation, the hidden layer and the output layer show linear transformation relation, and the radial basis function is a learning function from the input layer to the hidden layer of the neural network and is a typical nonlinear function; the corresponding approximation model function is:
r k ＝R(||x-T k ||)
wherein x= [ x ] 1 ,x 2 ,…,x n ]Represents an n-dimensional input vector, i is an euclidean norm, T k Is the center of the kth hidden node, k=1, 2, N T ，N T To train the number of sample points, R () is an RBF function, R k Is an output distance function;
s3, forecasting resistance: selecting a group of ship main scale and ship type parameter data outside the database as input, calculating friction resistance by adopting a Prandtl-Schlichting formula, forecasting resistance coefficients based on each approximate model established in S2, comparing with corresponding test values, and verifying the accuracy of forecasting resistance of the established approximate model; finally, selecting a model with a more stable error range and a smaller error mean value as an approximate model for rapidly forecasting the ship resistance;
s4, automatically optimizing the main scale parameters of the ship by using an optimization algorithm: based on the finally selected approximate model in the step S3, a single-target optimization algorithm of a self-adaptive simulated annealing algorithm is utilized, and main scale parameters are optimized by taking the minimum resistance of the full-navigational speed section as a target.
2. The method for optimizing the main scale parameters of the ship based on the approximate model according to claim 1, wherein in the step S1, when the database is constructed, each parameter is uniformly distributed as much as possible in a reasonable range, and the upper and lower boundaries are as large as possible.
3. The method for optimizing the main scale parameters of the ship based on the approximate model according to claim 1, wherein in the step S1, a ship type scheme with more similar ship type characteristics is selected when the database is constructed.
4. The method for optimizing principal scale parameters of a ship based on an approximation model according to claim 1, wherein in step S2, the error analysis and the validity check of the approximation model function are performed by using a root mean square R and a determination coefficient error R 2 Checking the precision of the model, wherein R-0 represents small error of the model, high approximation precision and R 2 And 1, the similarity degree of the model and the original model is high.
5. The method for optimizing the main scale parameters of a ship based on an approximation model according to claim 1, wherein in step S3, the Prandtl-Schlichting formula is:
wherein Re is the Reynolds number,l is the characteristic length, V is the characteristic speed, and V is the dynamic viscosity coefficient.
6. The method for optimizing the main scale parameters of a ship based on an approximation model according to claim 1, wherein in step S4, the objective function of the optimization algorithm is selected as the sum of the remaining drag coefficients of the full-speed segment, and the optimized mathematical model is as follows:
Min[SUM(Cr1,Cr2,…,Crn)]
in the optimization process, under the condition of meeting the main scale constraint, the minimum objective function is the optimal solution, and the corresponding main scale parameters are the optimal main scale parameter combination.
7. The method for optimizing the main scale parameters of the ship based on the approximate model according to claim 6, wherein in the step S4, the optimization flow of the main scale parameters specifically comprises the following steps:
s4.1, providing a primary scale of the ship and an initial value of a ship design parameter by adopting a self-adaptive simulated annealing algorithm;
s4.2, calculating the residual resistance coefficient of each navigational speed based on the design parameters in the step S4.1 by using the approximate model between the main scale, the ship type parameters and the residual resistance coefficient constructed by the method to obtain a target function value;
s4.3, evaluating target fitness according to the objective function value, updating design parameters according to a fitness evaluation result by an optimization algorithm, and iterating S4.3-S4.1-S4.2-S4.3 …; the optimization flow termination criteria is typically determined by the maximum number of iteration steps or the optimal solution reached.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111142269.XA CN113868765B (en) | 2021-09-28 | 2021-09-28 | Ship main scale parameter optimization method based on approximate model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111142269.XA CN113868765B (en) | 2021-09-28 | 2021-09-28 | Ship main scale parameter optimization method based on approximate model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113868765A CN113868765A (en) | 2021-12-31 |
| CN113868765B true CN113868765B (en) | 2023-09-05 |
Family
ID=78991868
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111142269.XA Active CN113868765B (en) | 2021-09-28 | 2021-09-28 | Ship main scale parameter optimization method based on approximate model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113868765B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115481494B (en) * | 2022-10-17 | 2023-08-15 | 武汉理工大学 | Method for generating model line pedigree of Yangtze river all-line passenger ship |
| CN117150652B (en) * | 2023-08-30 | 2024-04-30 | 武汉理工大学 | Sample point selection method, system and terminal based on ship-shaped constraint space |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109681299A (en) * | 2019-01-23 | 2019-04-26 | 黄冈师范学院 | Diesel engine Urea-SCR system ammonia the amount of injection and ammonia coverage rate Synchronous fluorimetry method |
| KR20200046994A (en) * | 2018-10-26 | 2020-05-07 | 삼성중공업 주식회사 | Apparatus and method for optimizing PID parameters for ship |
| CN111506970A (en) * | 2020-04-21 | 2020-08-07 | 常熟理工学院 | Ship hydrodynamic performance evaluation method |
| CN111797464A (en) * | 2020-06-05 | 2020-10-20 | 长江三峡通航管理局 | Ship-mechanical traction system design integrated system of ship lift |
| CN111797471A (en) * | 2020-06-24 | 2020-10-20 | 中国第一汽车股份有限公司 | Engine hood lightweight design method based on radial basis function neural network approximate model |
| CN112487715A (en) * | 2020-11-27 | 2021-03-12 | 江苏科技大学 | Method for optimizing reliability of process parameters of key hole system of marine diesel engine body |
| CN112699497A (en) * | 2021-01-22 | 2021-04-23 | 上海船舶研究设计院（中国船舶工业集团公司第六0四研究院） | Method and system for establishing multi-target joint optimization model of air route speed |
-
2021
- 2021-09-28 CN CN202111142269.XA patent/CN113868765B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20200046994A (en) * | 2018-10-26 | 2020-05-07 | 삼성중공업 주식회사 | Apparatus and method for optimizing PID parameters for ship |
| CN109681299A (en) * | 2019-01-23 | 2019-04-26 | 黄冈师范学院 | Diesel engine Urea-SCR system ammonia the amount of injection and ammonia coverage rate Synchronous fluorimetry method |
| CN111506970A (en) * | 2020-04-21 | 2020-08-07 | 常熟理工学院 | Ship hydrodynamic performance evaluation method |
| CN111797464A (en) * | 2020-06-05 | 2020-10-20 | 长江三峡通航管理局 | Ship-mechanical traction system design integrated system of ship lift |
| CN111797471A (en) * | 2020-06-24 | 2020-10-20 | 中国第一汽车股份有限公司 | Engine hood lightweight design method based on radial basis function neural network approximate model |
| CN112487715A (en) * | 2020-11-27 | 2021-03-12 | 江苏科技大学 | Method for optimizing reliability of process parameters of key hole system of marine diesel engine body |
| CN112699497A (en) * | 2021-01-22 | 2021-04-23 | 上海船舶研究设计院（中国船舶工业集团公司第六0四研究院） | Method and system for establishing multi-target joint optimization model of air route speed |
Non-Patent Citations (1)
| Title |
|---|
| Dynamic sampling method for ship resistance performance optimisation based on approximated model;Haichao Chang等;Ships and Offshore Structures;386-396 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113868765A (en) | 2021-12-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109858647B (en) | Regional flood disaster risk evaluation and estimation method coupled with GIS and GBDT algorithm | |
| CN113868765B (en) | Ship main scale parameter optimization method based on approximate model | |
| Fan | A modification to particle swarm optimization algorithm | |
| CN112288164B (en) | Wind power combined prediction method considering spatial correlation and correcting numerical weather forecast | |
| CN112557034B (en) | Bearing fault diagnosis method based on PCA _ CNNS | |
| CN113190688B (en) | Complex network link prediction method and system based on logical reasoning and graph convolution | |
| CN109165421B (en) | Ship shafting bearing load value prediction method based on genetic algorithm optimization BP neural network | |
| CN112836377B (en) | Ship simulation test system based on digital simulation | |
| CN105844334B (en) | A kind of temperature interpolation method based on radial base neural net | |
| CN113919221A (en) | Fan load prediction and analysis method and device based on BP neural network and storage medium | |
| CN111200141A (en) | Proton exchange membrane fuel cell performance prediction and optimization method based on deep belief network | |
| CN112036651A (en) | Electricity price prediction method based on quantum immune optimization BP neural network algorithm | |
| CN115860170A (en) | Power quality optimization method of power distribution system considering power electronic load | |
| Zeng et al. | Short-term load forecasting of smart grid systems by combination of general regression neural network and least squares-support vector machine algorithm optimized by harmony search algorithm method | |
| CN110880044B (en) | Markov chain-based load prediction method | |
| CN104156878B (en) | Determination method for upgrading of rural power grids upgrading engineering evaluation index weights | |
| CN116050241A (en) | Submarine pipeline corrosion rate prediction method based on PCA-TSO-BPNN model | |
| CN103887823B (en) | Micro-grid connection position based on Fuzzy Level Analytic Approach choosing method | |
| Ferreira et al. | Genetic assisted selection of RBF model structures for greenhouse inside air temperature prediction | |
| CN112861418A (en) | Short-term icing thickness prediction method for stay cable based on GA-WOA-GRNN network | |
| CN114818128B (en) | Modeling method and optimizing method for ship body local curved surface optimizing neural network | |
| CN116415177A (en) | Classifier parameter identification method based on extreme learning machine | |
| CN114819107A (en) | Mixed data assimilation method based on deep learning | |
| CN114372618A (en) | Student score prediction method and system, computer equipment and storage medium | |
| CN110543724B (en) | Satellite structure performance prediction method for overall design |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |