基于广义椭球基函数模糊神经网络的油轮转向动态响应模型(英文)

基于广义椭球基函数模糊神经网络(GEBF-FNN)算法,提出一种新颖的油轮转向动态响应模型.通过事先建立好的一组油轮操纵非线性微分方程获得训练数据,GEBF-FNN算法用于在线辨识Nomoto型油轮转向响应模型的参数K和T.具体地,GEBF-FNN模型从没有任何模糊规则开始,基于规则生长准则和参数估计方法,在线生成模糊规则,从而学习出由一组模糊规则构成的具有高精度和精简系统结构的油轮转向动态响应模型.为验证该动态响应模型的有效性,针对典型的Z形操纵进行仿真研究,并进行广泛的比较研究,仿真结果显示基于GEBF-FNN算法的油轮动态响应模型具有理想的逼近和预测性能.

Dynamical response model for tanker steering using GEBF based fuzzy neural networks

A novel dynamical response model is proposed for tanker steering by employing the promising generalized ellipsoidal basis function based fuzzy neural network(GEBF-FNN) algorithm.Based on a group of well established nonlinear differential equations for tanker maneuvering dynamics,training data samples are generated for the GEBF-FNN method to online identify the Kand Tparameters of the tanker response model in the form of Nomoto steeringmodel.The GEBF-FNN model starts with zero fuzzy rules and online recruits efficient fuzzy rules via rule node generation criteria and parameter estimation.As a consequence,it results in adynamical response model for tanker steering with high accuracy and transparent structures consisting of a group of fuzzy rules.In order to demonstrate that the proposed response model is effective,simulation studies are conducted on typical zig-zag maneuvers.Moreover,comprehensivecomparisons are carefully presented.Simulation results indicate that the GEBF-FNN based response model can achieve promising performance in terms of approximation and prediction.