Analysis and optimization of the structural mass of a ship designed by classification rules

• Grant number: 12/03644-3
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: July 01, 2012
• End date: June 30, 2013
• Field of knowledge: Engineering - Naval and Oceanic Engineering - Marine and Ocean Structures
• Principal Investigator: Thiago Pontin Tancredi
• Grantee: Vitor Tonzar Chaves
• Host Institution: Escola Politécnica (EP). Universidade de São Paulo (USP). São Paulo , SP, Brazil

Abstract

The structural design is, since the nineteenth century, one of the areas where optimization techniques have been applied more intensively to reduce the weight of the structure without compromising its integrity. This is the case of structural design in the marine industry. The reduction of the light weight of the vessel and possibly the cost of manufacturing are directly associated with the minimization of structural weight. This research plan proposes to develop a model of synthesis and optimization of the structural configuration of vessels to minimize its structural weight since the early stages of the project. This model will be based on techniques of neural networks and allow to estimate the structural weight, the position of center of gravity and the main features of the structural design of ships with displacement less than 100 meters long made of steel. The function of this model will give the designer a tool capable of, even in the early stages of the project; provide rapid estimates of the optimized structural arrangement of a vessel that meets the criteria for the classification standards of vessels. To construct the model will consider a number of vessels initially described only by its main geometrical characteristics (length, width, draft, depth and block coefficient). Each of these vessels will have its structural arrangement optimized according to the criteria of the standard. The results of the optimized structural design, as well as their respective weights and structural positions of center of gravity will then be used to train a neural network. Finally a new set of vessels will have the structural arrangements designed based on neural network model built and the results are compared with those obtained by traditional techniques, including classical empirical models presented in the literature to estimate the position of center of mass and the structural weight. (AU)