Application of Swarm Intelligence to a digital excitation control system

This paper describes an application of swarm intelligence (SI) technique to a digital excitation control system. Some of the modern voltage regulator systems are utilizing the proportional, integral, and derivative (PID) control for stabilization. Based on given excitation system parameters, several PID tuning approaches are reported. Since in general, these parameters are not available during commissioning, specifically the machine time constants, this lack of information causes a considerable time delay and cost of fuel usage for commissioning the automatic voltage regulator (AVR). To reduce the commissioning time and cost, the excitation system parameters are automatically identified and the PID gains are calculated using well-developed algorithms. A swarm intelligence (SI) technique is proposed to identify the system parameters and compared with recursive least square (RLS) with linearization via feedback. The performance of the proposed method is evaluated with several generator sets. With self-tuned PID gains, commissioning is accomplished very quickly with excellent performance results.