Institut de Robòtica
i Informàtica Industrial

Description

In this proposal, a research on the design of fault-tolerant predictive controllers for large-scale systems is proposed. Fault detection and isolation are based on set invariance theory and interval observers. The research objective for fault detection is to combine both invariant set-based methods and interval observer-based methods and develop new ideas to detect faults not only in transient state but also in steady state. This point will be the core of this proposed research. Regarding fault tolerance of large-scale systems, only faults occurring in sensors and actuators are considered. Both fault detection and system reconguration are based on multi-sensor and multi-actuator schemes, which can keep stability of systems when several conditions are satised even under faults. Other works on multi-sensor and multi-actuator schemes are to improve them with considering model uncertainties and broaden them with interval observers.

In this work, model predictive control is used to implement controllers of multi-sensor and multi-actuator schemes. Besides, model predictive control strategy is also used to design the reference governor which can ensure to meet separation conditions of invariant sets. The nal achievements of this work will are attempted to solve problems related to faults of actuators and sensors in the Barcelona Drinking Water Network. However, preliminary results of this proposed methodology takes a probable fault in the pitch system of wind turbine as the case study.

The work is under the scope of the following projects:

• WATMAN: Analysis and Design of distributed optimal control strategies applied to large-scale WATer systems MANagement (web)