An improved neuro-adaptive-optimal control for oscillation damping in wind-integrated power systems

This article proposes an improved neuro-adaptive-optimal control scheme, based on online system identification and simultaneous control, to replace power system stabilizer in the renewable-energy-penetrated power systems. A simple, linear neural identifier, with a few adjustable connection weights is used, which ensures minimal computational burden, reduced development time, and makes the controller practically realizable. An adaptive learning rate, derived using Lyapunov stability theorems, guarantees stability of convergence of the learning algorithm as well as an optimal speed of convergence. It is demonstrated that a simple linear neural identifier, which approximates a local linear model of a system, by adjustment of its parameters online, is faithfully able to track the varying dynamics of the system. Improved oscillation-damping performance over a wide range of operating conditions and disturbances, in comparison with a well-established IEEE-PSS1A and fuzzy-logic-control-based PSS, was validated through simulation studies on a single-machine infinite-bus power system and a wind-integrated two-area power system. The computational superiority of the proposed scheme in comparison to complex and non-linear neural networks and fuzzy-logic-based control was also established. The novelty of the controller lies in its structure which, in-spite of being purely linear, performs robustly for highly complex and non-linear power system models.