期刊名称：International Journal of Applied Mathematics and Computer Science
电子版ISSN：2083-8492
出版年度：2018
卷号：28
期号：1
页码：1-14
DOI：10.2478/amcs-2018-0007
出版社：De Gruyter Open
摘要：This paper proposes two different health-aware economic predictive control strategies that aim at minimizing the damage
of components in a pasteurization plant. The damage is assessed with a rainflow-counting algorithm that allows estimating
the components’ fatigue. By using the results obtained from this algorithm, a simplified model that characterizes the health
of the system is developed and integrated into the predictive controller. The overall control objective is modified by adding
an extra criterion that takes into account the accumulated damage. The first strategy is a single-layer predictive controller
with an integral action to eliminate the steady-state error that appears when adding the extra criterion. In order to achieve
the best minimal accumulated damage and operational costs, the single-layer approach is improved with a multi-layer
control scheme, where the solution of the dynamic optimization problem is obtained from the model in two different time
scales. Finally, to achieve the advisable trade-off between minimal accumulated damage and operational costs, both control
strategies are compared in simulation over a utility-scale pasteurization plant.
