Design And Implementation Of Generalized Predictive Controller For Reheating Furnace Temperature

Steel is widely used as raw material in many industries. Meanwhile it is a very important kind of material for national defense. Steel industry is one of the pillar industries for domestic economy, so that steel industry reflects the overall national power of a country. China has a very large scale production of steel, however the technologies of producing steel are still evolving. In recent years, the steel production of China has been excessive, which leads to a persistent declining steel market. Under this kind of industry background, enterprises are forced to upgrade their production specifications to enhance their competitive. Applying advanced control technology would notably reduce the energy consumption and yield appreciable economic benefit.The control problem of temperature of the furnace used for reheating steel to be rolled is studied. A specific furnace is chosen as study object. A generalized predictive controller is designed for the furnace temperature. The original control facilities and control loops are considered to be effectively utilized at the designing stage. Then a temperature-flow cascade control loop is implemented, generalized predictive controller is used in the outer loop to control furnace temperature, PID controller is used in the inner loop to control gas flow.While implementing the controller, saturation problem is encountered and then solved by proposing a dynamic output constraint method.At the same time, a multi-model predictive control strategy is proposed to solve the model mismatch problem when the production line periodically gets to maintaining status.