Controllability analysis and optimal control of biomass drying with reduced order models

Abstract

Complex industrial processes such as the drying of combustible biomass can be described with partial differential equations and finite volume methods. It is not straightforward to use these models to monitor or analyze process parameters due to the complexity of these methods. We show that reduced order models are capable to overcome this drawback and can be used in model based observers to determine quantities that cannot be measured directly.

Epidemic model dynamics and fuzzy neural-network optimal control with impulsive traveling and migrating: Case study of COVID-19 vaccination

“…For the optimal-control problems with ICS, they have a few schemes such as a linear-quadratic (LQ) controller [35], adaptive dynamic programming (ADP) [36] and Pontryagin's maximum principle [26]. For the dynamics with partial knowledge of models, the neural optimal controller [37], the piecewise-constant optimal controller [38], models reducing order [39], neuro-fuzzy inference system [40] and the optimal control based on passivity [41] have been developed with the full state observer and measurement. Moreover, the model dynamics must be well-defined with the appropriate accuracy because of the impulsive data acquisition, [34].…”