Adaptive RBF neural network based on sliding mode controller for active power filter Online publication date: Thu, 04-Jun-2020
by Huiyue Zhang; Yunbo Liu
International Journal of Power Electronics (IJPELEC), Vol. 11, No. 4, 2020
Abstract: In this paper, an adaptive radical basis function neural network (RBFNN) is proposed to deal with dynamic tracking error problems which are the mathematic model uncertain or complex for the three-phase active power filter (APF). The adaptive RBFNN systems are employed to approximate the system function term f^(x) in the sliding mode controller. Different from back propagation neural network (BPNN), RBFNN is local uniformly convergence so that it enhances the convergence speed of weights. According to Lyapunov stability analysis guarantee the control algorithm stable implementing that is working out the adaptive law and dragging the states onto the sliding surface and sliding along it. This approach is almost model-free requiring a minimal amount of a priori knowledge and is robust in the face of parameter changes. The simulation results of APF demonstrate the outstanding compensation performance and strong robustness.
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Power Electronics (IJPELEC):
Login with your Inderscience username and password:
Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.
If you still need assistance, please email subs@inderscience.com