Session: 08-06-01 non-presentations

Paper Number: 126635

126635 - Research on Quantitative Model of Ship Performance Evaluation Index Based on Neural Network 

Research on Quantitative Model of Ship Performance Evaluation Index Based on Neural Network

Abstract

         The quantification of evaluation indicators is an important foundation to ensure the objectivity of evaluation. In order to reasonably quantify the performance evaluation indicators of ships, a quantitative model for ship performance evaluation indicators based on neural network technology is investigated. Research and analyze the correlation between ship performance experimental data and various professional attributes, establish a quantitative calculation model for ship performance indicators based on expert experience and fuzzy set theory. And use neural network technology to verify and correct the consistency of expert knowledge. On the basis of reasonable classification of expert knowledge, complete the optimization of quantitative models based on expert knowledge, and analyze and predicte the convergence of the model. Additionally, a quantitative model for assessing ship performance was verified using data that was made accessible to the public from both local and foreign sources, including the "Mozi" database in order to verify the validity of the model. The test results show that the proposed model has good convergence and can provide a basis for the evaluation and optimization of conceptual schemes. 

Key words: Neural network; Quantitative model; Ship performance; Evaluation index

Presenting Author: Shengchen Ji Wuhan University of Technology

Presenting Author Biography: Master student in Wuhan University of Technology

Authors:

Hao Wang Wuhan University of Technology
Shengchen Ji Wuhan University of Technology