Abstract

The increasing complexity of modern thermal power plants and the demands of economic as well as the operational requirements necessitate the need for continuing improvements in power plant performance monitoring and optimization techniques. The paper discusses an integrated approach for real-time plant performance monitoring. An initial investigation into performance monitoring using artificial neural networks is presented. Unit heat rate and a predictive plant model for optimal plant control are described as two case studies.