Research on Resistance Prediction Method of Conventional Planning Hull Based on BP Network

Abstract

Resistance performance is an important factor affecting the usability and economy of planning hulls. This paper proposes to use the neural network to build the relationship model between the characteristic parameters of the planning hull and its hydro-resistance performance. Six form parameters (such as overall length, breadth of chine line, et al ) and six state parameters (such as displacement, velocity, et al.) are selected as characteristic parameters. The resistance coefficient is selected as the research indicator. The neural network response model is constructed by taking characteristic parameters as input variables and the research indicator as the output of the network. The mean squared error and determination coefficient are treated as performance function and assessment indicators respectively. After training, we compare the output of the network with the experimental results. It shows that the predicted value of the network agrees well with model test results with the determination coefficient reaching 0.99. This illustrates that the trained network has high precision in capturing the relationship between characteristic parameters and resistance coefficient, which provides a new way of resistance prediction for the conventional planning