论文部分内容阅读
针对航空发动机是一个不确定性的强非线性系统,借鉴预测控制的思想,提出了基于径向基函数RBF(Radical Basis Function)网络的航空发动机预测滑模控制.首先利用RBF网络建立航空发动机预测模型,进而得到滑模预测模型;其次在线修正网络参数实时反馈校正滑模预测模型,滚动优化求取控制量;然后采用另外一个RBF神经网络实现了全包线建模和控制;最后分析了控制系统的收敛性.仿真结果表明,所设计的控制器性能良好,能有效地抑制参数摄动和干扰的影响.
Aero-engine predictive sliding mode control based on radial basis function RBF (Radical Basis Function) network is proposed by reference to the idea of predictive control, aero-engine prediction is a strong nonlinear system.An aircraft engine prediction Model, and then get the sliding mode prediction model; Secondly, online modify the network parameters real-time feedback to correct the sliding mode prediction model, scroll optimization to obtain the control volume; then use another RBF neural network to realize the all- envelope modeling and control; finally analyze the control The convergence of the system.The simulation results show that the designed controller has good performance and can effectively restrain the influence of parameter perturbation and disturbance.