Using Artificial Neural Networks to Implement Real-Time Optimized Multi-objective Power Plant Control in a Multi-Agent System

There are many benefits to using a multi-agent system for power plant control, such as the ability to perform multiple computationally intensive tasks in parallel such that effective optimized real-time control can be achieved. These parallel tasks include neural network training, parameter optimization, and system monitoring. The multi-agent system approach also allows for intelligent control that is robust and flexible in that it can autonomously make decisions and adjust to partial control system failure to maintain control with minimal performance degradation, to name a few of the potential benefits. It is the goal of this paper to detail the implementation of neural networks as models of the power plant system to allow effective online performance evaluation of the power plant in the optimization processes, where the optimization processes provides optimized control setpoints and feedback gain values. Though equations are available for the power plant model used here, the larger, more complex power plant systems for which this MAS control system is ultimately being designed will either not have equations available or the model is too computationally complex for use in real-time operation, making it necessary to have a simpler model such as an artificial neural network.