Publication
Health-aware LPV-MPC based on a reliability-based remaining useful life assessment
Conference Article
Conference
IFAC Symposium on Fault Detection Supervision and Safety for Technical Processes (SAFEPROCESS)
Edition
10th
Pages
1285-1291
Doc link
https://doi.org/10.1016/j.ifacol.2018.09.569
File
Abstract
One of the relevant information provided by the prognostics and health management algorithms is the estimation of the Remaining Useful Life (RUL). The prediction of the expected RUL is very useful to decrease maintenance cost, operational downtime and safety hazards. This paper proposes a new strategy of health-aware Model Predictive Control (MPC) for a Linear Parameter Varying (LPV) system that includes as an additional goal extending the system RUL via their estimation using reliability tools. In this approach, the RUL maximization is included in the objective function of the LPV-MPC controller. The RUL is included in the MPC model as an extra parameter varying equation that considers the control action as scheduling variable. The proposed control approach allows the controller to accommodate to the parameter changes. Through computing an estimation of the state variables during prediction, the MPC model can be modified to the estimated state evolution at each time instant. Moreover, for solving the optimization problem by using a series of Quadratic Programs (QP) in each time instant, a new iterative approach is exhibited, which improves the computational efficiency. A pasteurization plant control system is used as a case study to illustrate the performance of the proposed approach.
Categories
control theory, optimisation.
Author keywords
Remaining Useful Life(RUL),Model Predictive Control,Linear Parameter Varying, Relability
Scientific reference
F. Karimi, V. Puig and G. Cembrano. Health-aware LPV-MPC based on a reliability-based remaining useful life assessment, 10th IFAC Symposium on Fault Detection Supervision and Safety for Technical Processes, 2018, Warsow, Poland, Vol 51-24 of IFAC Papers Online, pp. 1285-1291, IFAC-Elsevier.
Follow us!