DC-3: Name of the DC candidate
Contact
Project
Safety-aware predictive control methods with robust contingency plans
Host organization
NTNU
Supervisors
Prof. Johannes Jäschke (Main, NTNU); Prof. Gürkan Sin (co-supervisor, DTU)
Duration
36 months
Objectives
The overall objective is to design novel algorithms for short-term optimal operation that always provide contingency plans to bring the process plant to a safe operation regime in the presence of uncertainty by combining methods from robust model predictive control, state estimation with online risk monitoring methods (DC-4) to account for fault scenarios, and to provide prioritized routes to reach a safe operation region.
The contingency plans will consider the present state of the plant, as well as possible future degradation, maintenance, and safety actions, while rigorously taking uncertainty into account. A case study for hydrogen production and liquefaction will be used to demonstrate the capability and robustness of the approach.
DC-3 will integrate the methods from DC-4 into the developed framework through close collaboration.
Expected results
-
Proposal of a robust, fault-tolerant, and safety-focused model-based control strategy that provides contingency plans for equipment and sensor failure,
-
Validation of the approach on simulations,
-
Comprehensive robustness analysis,
-
Inclusion of models from DC-4 in the framework,
-
Validation of the safety-aware predictive control methods with robust contingency plans in an industrial case study for hydrogen production,
-
Prototype software for fault-tolerant model predictive control with contingency plans.
Planned secondments
- DTU, M15, 3 months: Training on machine learning models for application in the control framework
- KAIROS, M28, 2 months: Application of predictive control to a renewable hydrogen case study