Title: MPC course materials
1Model Predictive Control Dr.Ir. Ton van den
Boom Prof.Dr.Ir. Ton Backx
October 27, 2003
2Model Predictive Control
- Fourth lecture
- Robustness
- Distillation Case
December 20, 1999
3Model Predictive Control
- Distillation is one of the most applied
separation processes in the processing
industries. It is one of the processes where MPC
can bring significant benefits - Tight control of quality of products
- Energy saving
- Minimization of give away
- Robustness
- Distillation Case
4Model Predictive Control
FF inlet feed flow rate XF inlet feed
composition Fst steam flow rate Fref reflux
flow rate FD distillate (top) product flow
rate XD distillate product composition FB
bottom product flow rate XB bottom product
composition LC condenser holdup LB reboiler
holdup
- Robustness
- Distillation Case
5Model Predictive Control
- Robustness
- Distillation Case
Increase of the reflux flow
6Model Predictive Control
- Robustness
- Distillation Case
Responses of distillate purity and bottom product
impurity to a step in the reflux flow
7Model Predictive Control
- Robustness
- Distillation Case
Increase of the steam flow
8Model Predictive Control
- Robustness
- Distillation Case
Responses of distillate purity and bottom product
impurity to a step in the steam flow
9Model Predictive Control
- Robustness
- Distillation Case
Staircase signal applied to the reflux flow
10Model Predictive Control
- Robustness
- Distillation Case
Responses of top product purity and bottom
product impurity to a staircase change of the
reflux flow
11Model Predictive Control
- Robustness
- Distillation Case
Responses of linearized top and bottom product
impurity to a staircase change of the reflux flow
12Model Predictive Control
- Robustness
- Distillation Case
Step responses of the reflux flow, reboiler steam
flow and feedflow (Disturbance variable) to
log(top impurity) and log (bottom impurity)
13Model Predictive Control
Prioritised Optimisation
Fulfill operational requirements
Satisfy MV constraints Satisfy Priority 1 zones
and targets, if not possible balance between all
priority 1 requirements Satisfy priority i zones
and targets, if not possible balance between all
priority i requirements Bring CVs as close as
still possible to optimal values Bring MVs as
close as still possible to optimal values
Fulfill quality requirements
Fulfill economic performance conditions
- Robustness
- Distillation Case
14Model Predictive Control
- Robustness
- Distillation Case
Responses of the top product purity and bottom
product impurity to the setpoint changes
15Model Predictive Control
- Robustness
- Distillation Case
Reflux flow and reboiler steam flow manipulations
applied by the control system
16Model Predictive Control
Directionality of process transfer
- Robustness
- Distillation Case
17Model Predictive Control
- Robustness
- Distillation Case
Top product purity and bottom product impurity
responses to further setpoint changes
18Model Predictive Control
- Robustness
- Distillation Case
Reflux flow and reboiler steam flow manipulations
applied by the control system
19Model Predictive Control
Top product purity and bottom product impurity
responses related to release of the constraints
- Robustness
- Distillation Case
20Model Predictive Control
- Robustness
- Distillation Case
Reflux flow and reboiler steam flow manipulations
applied by the control system