The fractional- order PI
α
D
β controller is more flexible and gives an opportunity to better adjust dynamical properties of a fractional-order control
system than the traditional PID controller. However, the parameter selection is more difficult for such a controller, because it introduces two additional
parameters α and β. For this problem, this paper proposes a fractional-order PI
α controller
... [Show full abstract] with self-tuning parameters based on neural network, and discusses the discretization method and the design method
of the PI
α controller. The architecture of back-propagation neural networks and parameters self-tuning algorithm of the controller are
described. Experiment results show that the controller presented not only maintains the performance of the normal fractional-order
PI
α controller, but also has better flexibility and parameters self-tuning ability