基于IGA-BP算法的船舶航向智能自适应控制系统设计

在深入研究基于BP学习算法的前向神经网络以及模糊神经网络控制器的基础上,针对模糊神经网络控制器难以设计以及传统BP学习算法易于陷入局部收敛的不足,结合免疫遗传算法的全局收敛特性以及BP学习算法局部收敛的快速性,提出了一种基于混合计算智能方法的IGA-BP算法的神经网络参数的优化设计方法.将设计的控制器用于建立船舶航向控制系统模型,仿真结果表明,在船舶无干扰和存在随机干扰的情况下,基于IGA-BP算法设计的船舶航向控制系统均能使船舶转向控制无超调,跟踪快,比BP学习算法的控制效果更理想.

Design of an intelligent self-adaptive control system for ship course based on IGA-BP algorithms

Aiming at the design difficulty of fuzzy neural network controller and the shortage of classical BP algorithm which leads to easily local convergence, after deep research of feed-forward neural network and fuzzy neural network controller based on BP algorithm, an optimization design method is proposed for the neural network parameters of IGA-BP algorithms based on blend compute intelligent method, that combines global convergence characteristics of the immune genetic algorithm with quick local convergence of BP algorithm. The designed controller is used to build up the model of ship course control system. The simulation results show that with or without disturb or random disturb, the ship course control system based on IGA-BP algorithms can lead to no overshoot of ship turning and quick following, which is better than the controlling result of BP algorithm.