Power balance control using evolutionary algorithm

This paper described a nonlinear optimization problem to enable independent system operators (ISOs) to make least-cost decisions for the activation of ancillary services. A self-organizing migration algorithm (SOMA) based on the competitive-cooperative behavior of intelligent creatures solving a common problem was tested on 2 scenarios based on real values obtained from a transmission system operator (TSO) in the Czech Republic. The 6 hour datasets were used to demonstrate the prediction of error between production and consumption. The characteristics and capacity of the ancillary services were used as system and operating constraints for the optimization. History values from the datasets were then simulated. The SOMA algorithm then searched for solutions to minimize the cost function. Scenarios for a short and long outage of a large generating unit were used to demonstrate the algorithm's ability to balance power in the system. 4 refs., 1 tab., 3 figs.