Portfolio · Publications
Papers in continuous-time multiparametric optimization and control.
Develops a direct continuous-time multiparametric framework, gives step-by-step algorithms for switching structures and critical regions, and compares the resulting maps with discrete-time formulations.
View arXiv preprint ↗Characterizes the critical regions by the switching times, the instants at which the active constraints change. For a given sequence of active constraints and fixed switching times the optimality conditions are linear in the initial state, so every boundary, hyperplane or curved, is computed exactly from linear equations, and the complete map is assembled without solving optimal control problems.
Establishes when a continuous-time critical-region boundary is a hyperplane and provides an analytical construction for that case.
View arXiv preprint ↗Applies the continuous-time approach within a process-control workflow and shows how it can reduce the number of stored control laws while preserving switching information.
View arXiv preprint ↗Shows that the sequences of time-varying active constraints on either side of a critical-region boundary determine whether it is a hyperplane or curved, before any computation is carried out.
View arXiv preprint ↗Computes a curved critical-region boundary exactly, without solving an optimal control problem. Once the instant at which a constraint becomes active is fixed, every condition on the boundary is linear in the initial state.
Compares continuous-time and discrete-time explicit control on forced-circulation evaporators in series with one, three, and five effects. The continuous-time critical regions are few, fixed, and faster to compute, while the discrete-time ones multiply with the number of discretization intervals and with the number of states until the computation becomes intractable.