Multivariable self-tuning temperature control of a thermo-analyzer

The paper deals with the multivariable self-tuning temperature control of a thermo-analyzer. The controller parameters are derived on the basis of the deadbeat approach. For parameter estimation of the process model, the recursive least squares method with LD decomposition and directional forgetting is applied. This multivariable controller has been designed especially for technological processes, where the process output variables without overshoot or with prescribed overshoot are desired and reference signals are changed with steps and linear increasing.