Adaptive Neural Inverse Control Applied to Power Systems

Neural networks have been extensively studied and widely used in many practical applications for identification and control of nonlinear dynamical systems in the past two decades or so. Numerous research results have been reported in the literature concerning using neural networks in the inverse control scheme or a more robust control scheme: internal model control, to control nonlinear dynamical systems to achieve desired tracking performance. A stable reference model is often times assumed to exist and is used to dictate the desired dynamic behavior of the control system. However, finding an appropriate reference model that accurately represents the desired system dynamic behavior is not a trivial matter for most cases. In addition, in many practical applications such as power systems, the admissible controls are constrained within a physically allowable range, which presents another layer of difficulties to directly apply the reference model based inverse control. Dealing with these difficulties yet achieving optimal control objectives constitutes one of the main motivations for this research effort. This paper attempts to present a design procedure of neural inverse control for a specific class of power systems to ensure the system stability in an optimal sense (for instance in minimum time), and a general adaptive optimal control framework that utilizes optimal control theory, the inverse control, and hierarchical neural networks to control uncertain power systems in an optimal manner. The simulation study is conducted on a single-machine infinite-bus (SMIB) system to illustrate the proposed design procedure and demonstrates the effectiveness of the proposed control approach.