一种简化的减摇鳍自适应神经网络控制

A Novel Adaptive Neural Network Control for Fin Stabilizer

基于Lyapunov稳定性定理和动态面控制技术,针对船舶减摇鳍系统提出了一种自适应神经网络控制设计方法。文章采用动态面控制技术消除传统Backstepping方法中存在的"计算爆炸"问题,同时从实际应用的角度出发,充分考虑系统中存在的显著不确定性,并运用径向基神经网络对不确定性进行估计及补偿。此外,采用最少学习参数技术减少控制器的计算负担,使所设计的控制器更贴近工程应用。文中所设计的控制器保证了船舶减摇鳍系统信号一致最终有界,使系统输出收敛到一个较小区间,最后通过数值仿真验证了所提算法的有效性。

Based on the Lyapunov stability theory and the technology of dynamic surface control, a direct adaptive neural network controller is proposed for the ship fin stabilizer in the presence of system uncertainties and nonlinearities. The algorithm incorporates the technology of dynamic surface control to avoid the problem of ＂explosion of complexity＂ which is in the traditional Backstepping design procedure. In addition, the radial basic function neural network is used to estimate and compensate the unknown system uncertainties. The technology of minimum learning parameter is used to reduce the computational burden of the algorithm, so the controller is convenient to implement in applications. The proposed NN based controller guarantees that all the close-loop signals of the ship fin stabilizer are uniform ultimate bounded and that the tracking errors converge to a small neighborhood of the desired trajectory. Finally, the MATLAB simulation results are used to demonstrate the effectiveness of the proposed scheme.