论文部分内容阅读
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.
Based on the generalized ellipsoidal basis function fuzzy neural network (GEBF-FNN) algorithm, a novel dynamic steering response model of the tanker is proposed. The training data are obtained by manipulating a set of oil tanker nonlinear differential equations in advance. The GEBF-FNN algorithm On the line, the parameters K and T of the Nomoto steering response model are identified. Specifically, the GEBF-FNN model begins with no fuzzy rules and generates fuzzy rules online based on rule growth rules and parameter estimation methods to learn a set of fuzzy rules Aiming at verifying the effectiveness of this dynamic response model, a simulation study of a typical Z-shaped manipulator was carried out and a wide range of comparative studies were carried out. The simulation results show that the dynamic response model based on the GEBF-FNN The algorithm of tanker dynamic response model has ideal approximation and prediction performance.