Design and research of digital supervisory control systems of a crude oil distillation column

The new control systems for crude oil distillation column are developed and investigated on the base of the new usage of supervisory approach. It is shown that such approach again is spreading in world practice of technological processes automation, especially in the petroleum industry. The advantages of linking the control system on the PID-family controller base and the control system on the linear-quadratic and model predictive controller base are researched. Four types of supervisory systems were researched. Analysis of transitional processes in closed control systems in case of load disturbances shows that all researched supervisory control systems increasе control quality in comparison with decentralized control systems on PID-family controller base and provide equal quality in comparison with much more complex multivariable model predictive controllers.