Observer‐based fuzzy adaptive optimal control for nonlinear continuous‐time saturation interconnected systems

Abstract

Summary A fuzzy adaptive optimal output‐feedback control problem is investigated for nonlinear continuous‐time interconnected systems with saturation constraints. The system dynamics and the mismatched interconnections of each subsystem are unknown, and each subsystem contains unavailable states. Based on fuzzy logic systems, a fuzzy state observer is designed to approximate the unknown system states and the interconnections of the nonlinear interconnected system. By using adaptive dynamic programming technology, the decentralized optimal controller of each subsystem is given and the saturation control input is updated via the critic fuzzy logic systems. The proposed fuzzy adaptive observer‐based optimal control scheme can guarantee that the closed‐loop interconnected system is ultimately uniformly bounded stable. A simulation example validates the effectiveness of the presented scheme.

Data-Driven Adaptive Optimal Control Under Model Uncertainty: An Application to Cold Atmospheric Plasmas

“…Several approaches for adaptive optimal control of nonlinear systems with model uncertainty have been developed recently, mostly using a data-based paradigm. For example, methods based on fuzzy control and combination of feedforward and feedback control [31], [32] or adaptive state observers [33], [34] were developed to implement adaptive, decentralized, and fault-tolerant optimal control for nonlinear systems, including large-scale systems, and a similar method based on neural networks was proposed by [35]. However, these methods typically require a specific structure for either the model or mismatch, which may not encompass the true system.…”

Robust decentralized optimal stabilizability of continuous time uncertain interconnected systems: Application to large scale power system

“…These limitations motivate the design of decentralized control schemes, which leads to a wide variety of new concepts and results. [5][6][7] The advantage of such methods is that they reduce the complexity and therefore make the implementation of the control law more feasible. Decentralized control is a control design where local decisions are based only on local information of the subsystems.…”

“…Designing a centralized control for these systems may not be efficient due to the natural systems modularity, which may prevent a viable way on sharing information across the subsystems, and often it may be too costly when it is implemented. These limitations motivate the design of decentralized control schemes, which leads to a wide variety of new concepts and results 5‐7 . The advantage of such methods is that they reduce the complexity and therefore make the implementation of the control law more feasible.…”

Enhanced fuzzy state estimation of discrete‐time nonlinear systems via a multi‐instant united switch‐type observer

“…State estimation belongs to a kind of methodology that the internal state of a dynamic system is estimated by using measurable system measurement output 1 . Due to the well‐known effectiveness of state estimation, it has been playing an important role in a large number of control processes, for example, References 2,3. In the meantime, it also attracts a quantity of researchers to investigate the problem of state estimation of nonlinear systems, especially for the reason that most of industrial plants can be provided with complex nonlinear dynamics 4 .…”