Model predictive control of the superheater temperature based on a piecewise affine model

In the conditions of today´s deregulated electricity market, load following capability of power plants is of cardinal importance. However, it is also a big challenge for the steam power plant control system because in the wide range load following operation the inherent nonlinearity of power plant units is much more apparent than in the base load operation. This paper is focused on one important aspect of this complex problem: control of the main steam temperature at the superheater output. The superheater is a highly nonlinear system and there is little chance that advanced control based on single linearized model may achieve good control performance. The paper shows that nonlinear superheater model can well be approximated with a piecewise affine (PWA) system and controlled using model predictive control (MPC) based on PWA model. The responses to fast wide range load changes (in the range from 50 to 100% of the nominal load) and periodic input disturbances due to irregularities in coal mills are evaluated. It is shown that in both cases MPC control based on PWA model can achieve considerable performance improvement over classical control system based on cascades of gain scheduled PI controllers.