Abstract
In this paper, we present a multiple neural control and stabilization strategy for nonlinear and unstable systems. This control strategy method is efficient especially when the system presents different behaviors or different equilibrium points and when we hope to drive the whole process to a desired state ensuring stabilization. The considered control strategy has been applied on a nonlinear unstable system possessing two equilibrium points. It has been shown that the use of the multiple neural control and stabilization strategy increases further the stability domain of the system further than when we use a single neural control strategy.
Highlights
Multiple models characterizing different plant operation modes are used to predict the system behaviors
The application of multiple neural control strategy is based on neural models, which incorporates a set of pair model/controller
We have presented a multiple neuronal control and stabilization strategy which can be applied on systems characterized by different behaviors in different regions of the functional domain
Summary
Multiple models characterizing different plant operation modes are used to predict the system behaviors They are the one that best describes the plant and used to initialize new adaptation and/or generate new control input. The second is known as indirect switching, where local models are used at each moment in which controller will be used [3]. Narendra et al [4] presented a general methodology to design a multiple model adaptive control of uncertain systems. This methodology makes systems to operate effectively in an environment with a high degree of uncertainty. The banks of neural models and neural controllers are made after learning steps from sub-databases representing different behaviors of the controlled system
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have