Bow lines optimization and resistance prediction are important contents in ship design. In this paper, the bow lines optimization of the 47500-DWT bulk carrier is carried out, the resistance of the optimized ship is calculated by the CFD method, and the ship resistance prediction model is established based on the BP neural network. This study verifies the feasibility of the optimization method and the accuracy of using BP neural network to predict ship resistance.
Ship resistance prediction is one of the important contents of ship hydrodynamic performance prediction, for which many scholars have done a lot of work. Due to the complexity of ship resistance calculation, there is no mature pure theoretical formula. At present, the commonly used resistance prediction methods include model test, CFD numerical simulation, and parameter estimation method.
Generally, after the hull lines are determined, the resistance of the ship model can be obtained by the method of model test. The ship model test method mainly includes the process of model building, pool test, and test result conversion (Tu et all., 2018). However, in the early stage of ship design, the ship shape line has not yet been determined, and other methods must be adopted if the model test is not ready. The parameter estimation method is an empirical method developed on the basis of model test data, among which the representative method is the standard series atlas method. Standard series atlas methods include Taylor series, SSPA series at all (Williams, 1969). It draws a large number of standard series of ship resistance data into graphs or tables, so that the manual calculation of the ship's resistance follows a relatively simple and direct procedure, so as to study the relationship between the hull parameters and the hull's resistance, and guide the ship design (Shaher, 1971). With the development of computer technology and computational fluid dynamics theory, numerical simulation technology has been widely used in the prediction of ship hydrodynamic performance (Xu at all., 2019). In these papers, the ship's trim was optimized, and the CFD numerical simulation method was used to calculate the optimized ship's resistance (Lv at all., 2013; Sun at all., 2016; Lyu at all., 2018). And the data required for the study was obtained by this method. In this paper, CFD simulation and ship model test method are used to study the resistance and flow field of the super cruise ship, and analyze the influence of the turbulence model on the CFD calculation results (Guo at all., 2013). This paper uses the CFD method to demonstrate the effect of scale on ship resistance and investigate the uncertainty of full-scale resistance simulations. (Terziev at all., 2019).
