一种改进的神经网络非线性预测控制

从建立神经网络非线性预测模型出发，针对BP网络存在收敛速度慢，容易陷入局部最小的缺点，该文在BFGS拟牛顿法的基础上，提出了一种基于并行拟牛顿优化算法的并行拟牛顿神经网络。该并行拟牛顿优化算法采用两个含有不同参数的拟牛顿校正公式，在每次迭代过程中，利用这两个不同的校正公式得到相应的搜索方向，并通过不精确搜索法求取最优步长，最后根据一性能指标取最优的一个搜索方向和相应的步长对网络各层之间的权值进行修正。Matlab仿真结果表明，同BP神经网络和BFGS拟牛顿神经网络相比，该神经网络具有收敛速度快、模型精度高的特点，更适合于实时非线性控制。

A Novel Predictive Control Method Based on Neural Networks

To improve the rate of convergence of the BP neural networks for nonlinear system identification in nonlinear predictive control, based on the BFGS quasi - Newton algorithm, a novel parallel quasi - Newton optimization technique is proposed and as a multi - step predictive model of nonlinear industrialized process. simulation experiments have shown that the algorithm is able to increase the precision of nonlinear predictive model greatly and improve both the static and dynamic performances of the nonlinear predictive control system.