Ship hydrodynamic performance evaluation method

Abstract

The invention discloses a ship hydrodynamic performance evaluation method. The method comprises the following steps: selecting a parent ship; based on the ship type, constructing a series of sample points by adopting a Latin hypercube algorithm, constructing a series of sample ship types similar to the parent ship by adopting an arbitrary curved surface deformation algorithm, calculating six target values of all the sample ship types in waves by adopting a Reynolds average-fluid volume function method, and forming a sample data set; and then training a deep belief network model by adopting thesample data set, determining the accuracy of the constructed deep belief network model through accuracy verification, and finally evaluating a new ship type similar to the parent ship by the trainedmodel. The characteristics of ship sailing on the sea are comprehensively considered, the ship type with better performance in waves can be obtained, the applicability of ship type optimization is improved, the six hydrodynamic performance of the ship under the wave working condition can be effectively and rapidly calculated, and meanwhile the accuracy of data calculation can be guaranteed.

Description

technical field [0001] The invention relates to a ship evaluation method, in particular to a ship hydrodynamic performance evaluation method based on wave conditions. Background technique [0002] At present, in the field of ship design, the ship performance evaluation method based on RANS-VoF technology has become the current mainstream method. However, the time-consuming problem based on RANS-VoF technology has brought great difficulties to the fast and efficient production of the shipbuilding industry. How to accurately and efficiently evaluate the hydrodynamic performance of ships has become one of the key issues in ship design. [0003] In response to the above problems, an approximate method for quickly predicting ship performance has appeared in the prior art. For example: use BP neural network to predict the total resistance of the ship in still water; use artificial neural network to predict the remaining resistance of the ship; use the Kriging approximate model a...

AI Technical Summary

Problems solved by technology

However, the existing rapid and accurate assessment of the hydrodynamic performance of ships generally only considers the resistance performance of the ship in still water, but does not predict and evaluate the overall performance of the ship in waves

Images