Genetic algorithm-aided design of a fuzzy logic stabilizer for a superconducting generator

An important aspect of the design of superconducting generators concerns stability following a major system disturbance. Because the superconducting field winding has a very long time constant, turbine governor control is crucial for improving transient and dynamic stability. This paper describes the design of a fuzzy logic stabilizer using a genetic algorithm to enhance the stability of a superconducting generator whose turbine is equipped with fast acting electro-hydraulic governors. The stabilizing signal is based on the instantaneous speed deviation and acceleration of the superconducting generator and on a set of simple control rules. A new approach is proposed to generate the control rules, and thus increase the effectiveness of the fuzzy logic stabilizer. A genetic algorithm is used to search for optimal settings of the fuzzy stabilizer parameters. Simulation results, compared with those using a conventional stabilizer, show a significant improvement in the system performance over a range of operating conditions.