An adaptive optimum SMES controller for a PMSG wind generation system

An artificial neural network based online adaptive control of superconducting magnetic energy storage system (SMES) controller has been proposed to improve the dynamic performance of a permanent magnet synchronous generator (PMSG) wind system. The training data for the neural network has been generated through an improved particle swarm optimization (IPSO) algorithm. The weighting matrix for the radial basis function is obtained from a large input-output data set representing various operating conditions. The control parameters were updated for transient variations in the system through an adaptation procedure of the weighting functions. The proposed adaptive algorithm was tested on the PMSG system for different disturbances such as wind gust as well as low voltage condition on the grid. The adaptive radial basis function neural network (RBFNN) based SMES control exhibited excellent transient behavior following large disturbances on the wind system.