Intelligent control of cupola furnaces

The objective of this research is to improve the efficiency of the cupola operation in the presence of desired output changes. This is achieved by developing an automatic controller for the cupola furnace. A controller was previously developed that provided tight control over the output parameters of the cupola under little or no desired change. The controller considers physical and practical limitations on the possible inputs. Previous research described optimal pairing of cupola outputs and inputs. This is exploited in a fuzzy controller that pairs oxygen enrichment and blast rate to the outputs of metal temperature and melt rate, respectively. The controller is modified to determine action when problems in production require a slow down or shut down of the melting process. A model is developed using a combination of experimental data and data obtained from the AFS cupola model. The model is a first order multivariable dynamical system. Due to the nonlinearities of the cupola, the model is only accurate for a narrow operating band. By varying the model parameters, simulations show the control to be robust. Simulations in the presence of input disturbances and output sensor noise show the controller efficiently rejects the common noise of a cupola furnace environment.