Implementation of a Multi-Agent System in Java Agent Development Framework for a large-scale fossil-fuel electrical power unit control.

Abstract

With technology and a larger-than-ever population, the society demand for energy has achieved records in the last few years. Currently, large-scale fossil fuel power plants, ranging in the hundreds of MW’s, produce most of the power in the United States and must do efficiently with more added regulation and more in the short term future. The control for these large-scale power plants can be complex but must still address several different operating objectives. Implementing advanced control for a large-scale power plant targeting multiple objectives can be computationally intensive as well as difficult when attempted in a centralized control configuration. In order for this control to be more flexible, adaptive and robust, it needs to be coordinated in distributed environments. The method presented in this thesis involves Multi-Agent System (MAS), which addresses the challenges of computing optimal multi-objective control using Java Agent Development (JADE) Framework, a much more communication-efficient, network-oriented environment.