论文部分内容阅读
通过应用具有二阶线性收敛速度的拟牛顿法于多层前向网络 ,以作为非线性预测控制中的预测模型 ,结合非线性优化方法 ,实现对于一般意义非线性系统的预测控制。仿真表明文中算法大大提高了网络学习收敛速度 ,使非线性预测控制算法的实时性能有很大改观。
The quasi-Newton method with second-order linear convergence rate is applied to the multi-layer forward network as the predictive model in the nonlinear predictive control, combined with the nonlinear optimization method to achieve the predictive control of the general nonlinear system. Simulation results show that the proposed algorithm greatly improves the speed of network learning convergence and greatly improves the real-time performance of nonlinear predictive control algorithms.