A Technique for Developing Low Order Models of Power Plants

The development of a reduced order nonlinear mathematical model of a power plant is described. An integrated model development approach was used. It is shown that the integrated approach can efficiently assess a model´s performance capabilities in a quantified manner. By utilizing stochastic modeling, Maximun Likelihood Estimation, and Kalman Filtering, the approach provides many enhanced features, such as statistically based validity testing and accuracy evaluation. Discussions describe the motivations for each step of the approach. The approach is general and can be used for the development of other types of nonlinear system models, as well