OF THE DISSERTATION Local Planning for Continuous Markov Decision Processes

Determinantal point processes (DPPs) are random point processes well-suited for modeling repulsion. In machine learning, the focus of DPP-based models has been on diverse subset selection from a discrete and finite base set. This discrete setting admits an efficient sampling algorithm based on the eigendecomposition of the defining kernel matrix. Recently, there has been growing interest in usi...

This work investigates the effects of different seeding strategies and residence time characteristics on the dynamics of a Mixed Suspension Mixed Product Removal (MSMPR) crystallizer equipped with a fines dissolution unit and applied to continuous preferential enantioselective crystallization. The fines dissolution is included as recycle streams around the MSMPR crystallizer. Moreover, primary ...

The Process Analytical Technology (PAT) initiative has encouraged the development of new technology to improve upon the current manufacturing paradigm. As a result substantial attention has recently focused on continuous processing due to the ability to control disturbances online, avoiding the loss of processing materials and enabling effective process scale-up. In this paper, a pharmaceutical...

Among the known predicate calculus formalisms for axiomatizing commonsense reasoning about actions, the Fluent Calculus stands out in offering a solution not only to the representational but also the inferential aspect of the fundamental Frame Problem. In this paper we extend this formalism to modeling hybrid systems, which involve both discrete and continuous change. We borrow basic notions fr...

BACKGROUND Probabilistic reaction norms (PRNs) are an extension of the concept of reaction norms, developed to account for stochasticity in ontogenetic transitions. However, logistic regression based PRNs are restricted to discrete time intervals, whereas previously proposed models for continuous transitions are demanding in terms of modelling effort and data needed. METHODOLOGY/PRINCIPAL FIN...

Models of industrial processes often contain discrete phenomena superimposed on the continuous system behavior. Simulation of batch processes, start-up and shutdown procedures, fault diagnosis and alarms fall under this category. Models for such processes require a mathematical theory for both its continuous and discrete state transitions. A key problem in hybrid simulation lies in the detectio...

In this paper, a multivariate fault prognosis approach for continuous processes with hidden faults is proposed based on statistical process monitoring methods and multivariate time series prediction. It is assumed that the fault is a slowly time-varying autocorrelated process and can be completely reconstructed. Fault magnitude is estimated first via reconstruction, then predicted by a vector A...