基于RBF网络自适应的Buck变换器终端滑模控制

对于Buck变换器系统,考虑到实际应用中负载变动引起系统参数的不确定性,且不确定性上界无法测量的情况,本文拟采用RBF神经网络对不确定性上界进行自适应学习。针对Buck变换器输出电压的控制问题,为了避免普通滑模控制跟踪误差渐进收敛的问题,改善其动态响应速度和稳态性能,本文拟设计一种基于RBF神经网络的上界自适应的终端滑模控制器,并通过Simulink仿真验证这种方法的可行性。

Buck Converter Terminal Sliding Mode Based on RBF Networks Adaptive Learning

In Buck converter system,considering the uncertainty of the system parameter caused by load change in prac-tical application,and the uncertain up-bound value cannot be measured properly,RBF neural network is planned to be adopt-ed to learn the uncertain up0bound value.For the control problem of the output voltage of Buck converter,in order to avoid asymptotic convergence of the tracking error in conventional sliding mode control,and improve the speed of dynamic re-sponse and steady state performance,a terminal sliding mode controller which is based on RBF neural network to learn the uncertain up-bound value will be designed.At last,simulations are used to verify the feasibility of the algorithm.