【摘要】 针对未知非线性不确定系统 ,提出了一种新的基于神经网络的自适应控制策略 ,该方法只需辨识对象的正向模型 ,将神经网络与优化方法相结合 ,对控制量进行优化迭代求解 ,使被控对象的输出能较好地跟踪期望输出 ,并且分别针对单输入 -单输出系统和多输入 -多输出系统进行了控制算法的推导 ,其中的优化方法分别采用梯度法和高斯 -牛顿法·仿真结果表明 ,该算法能精确跟踪设定输出 ,超调量小 ,响应速度快 ,无稳态误差 ,控制效果是非常令人满意的更多还原

【Abstract】 For unknown nonlinear uncertain systems,a new adaptive control scheme based on neural network was proposed. It only needs the forward model of the plant. In this method,the neural network was combined with optimum method to solve the control law. It makes the output of plants follow the desired output very well. Control algorithm was deduced for SISO systems and MIMO systems respectively. The optimum method is gradient and Gauss Newton method respectively. It can follow desired output presicely and has small overshoot. Its response is faster and has no steady error. The effectiveness of these algorithms was verified by simulation.更多还原