Michael Baldea

Integrated production scheduling and process control: A systematic review

Abstract Production scheduling and process control have the common aim of identifying (economically) optimal operational decisions, and it is reasonable to expect that significant economic benefits can be derived from their integration. Yet, the scheduling and control fields have evolved quite independently of each other, and efforts aimed at integrating these two decision-making activities are quite recent. In this paper, we review progress made thus far in this direction. We identify key elements of control and scheduling, and carry out a systematic investigation of their use as building blocks for the formulation and solution of the integrated scheduling/control problem. On the basis of our review, we define several necessary directions for future development as well as a complement of promising applications.

Control of integrated process networks-A multi-time scale perspective

Abstract In this work, we consider a generic class of process networks that feature large material recycle streams and in which impurities are present, and are removed by a small purge stream. We show that such networks exhibit a dynamic behavior featuring three time scales. Employing a nested application of singular perturbation arguments, we propose a method for deriving non-linear, non-stiff, reduced order models for the dynamics in each time scale. Subsequently, we propose a hierarchical controller design framework that takes advantage of the identified time scale multiplicity. The developed concepts are exemplified through a case study on the dynamics and control of a reactor–single-stage separator network core, and illustrative numerical simulation results are provided.

A time scale-bridging approach for integrating production scheduling and process control ☆

Abstract In this paper, we propose a novel framework for integrating scheduling and nonlinear control of continuous processes. We introduce the time scale-bridging model (SBM) as an explicit, low-order representation of the closed-loop input–output dynamics of the process. The SBM then represents the process dynamics in a scheduling framework geared towards calculating the optimal time-varying setpoint vector for the process control system. The proposed framework accounts for process dynamics at the scheduling stage, while maintaining closed-loop stability and disturbance rejection properties via feedback control during the production cycle. Using two case studies, a CSTR and a polymerization reactor, we show that SBM-based scheduling has significant computational advantages compared to existing integrated scheduling and control formulations. Moreover, we show that the economic performance of our framework is comparable to that of existing approaches when a perfect process model is available, with the added benefit of superior robustness to plant-model mismatch.

Dynamics and control of chemical process networks: Integrating physics, communication and computation

Abstract This paper provides the theoretical foundation for the modeling, analysis and control of integrated chemical process networks, or, in short, “process networks.” The dynamics of process networks is represented using state-space descriptions derived from classical irreversible thermodynamics and constrained by the second law so that dissipation is always non-negative. The state descriptions (models) derived from this point of view provide exact process representations. A unique, quadratic Lyapunov function for stability analysis and control design is derived directly from the entropy. The resulting process models are complex and simplifications may be needed in practical applications. Time-scale decomposition and singular perturbation theory provide the basis for exploring the network-level dynamic behavior that emerges as a result of tight inventory integration, and developing appropriate reduced-order models and a hierarchy of control systems for managing inventories and inventory flows. Model-based networked control and Lyapunov theory are leveraged to develop an integrated control and communication strategy that manages the information flows between the network components and explicitly accounts for communication constraints.

From process integration to process intensification

Abstract In this paper, we establish a connection between process integration and process intensification. Focusing on processes with material recycle, we use an asymptotic analysis to demonstrate that intensification represents a limit case of tight integration through significant material recycling. Based on this result, we propose a novel avenue for discovering intensification opportunities at the process design stage. Subsequently, we investigate the dynamics and control implications of the transition from process integration to process intensification. We demonstrate that, for the same steady-state performance, the dynamic response of an integrated process is slower than that of its intensified equivalent. Also, we provide a theoretical justification for existing empirical arguments concerning the loss of control degrees of freedom caused by process intensification. The theoretical developments are applied on a reaction–separation–recycle process example.

Tight energy integration: Dynamic impact and control advantages

Process integration is a key enabler to increasing efficiency in the process and energy generation industries. Efficiency improvements are obtained, however, at the cost of an increasingly complex dynamic behavior. As a result, tightly integrated designs continue to be regarded with caution owing to the dynamics and control difficulties that they pose. The present work introduces a generic class of integrated networks where significant energy flows (either arising from energy recycling or of external origin) result in dynamic models with a multi-time-scale structure. Such networks feature a clear distinction between the fast dynamics of individual units and the slow dynamics of the entire network. We draw a connection between specific (steady-state) design features and structural properties that afford the development of a framework for the derivation of low-order, non-stiff, nonlinear models of the core network dynamics. Furthermore, we demonstrate that tight energy integration and the presence of significant energy flows can facilitate, rather than hinder, control structure design and performance, and propose a cadre for hierarchical control predicated on the use of fast, distributed control for the individual units and nonlinear supervisory control for the entire network. The developed concepts are illustrated with examples.

Nonlinear model predictive control of energy-integrated process systems

Abstract Improving energy efficiency has become particularly important in chemical processes in view of recent increases in energy prices, growing environmental concerns, and regulatory pressure. In this paper, we consider a class of process systems with significant energy recovery. Extending our previous results concerning the two time scale dynamics of such systems, we demonstrate that the fast component of the dynamics is asymptotically stable in practical cases. Using this result, we develop a hierarchical control framework, consisting of a linear control system for the fast dynamics and a MISO nonlinear model predictive controller for the slow dynamics, and prove that it guarantees exponential stability for the overall system. Subsequently, we explore the implications of this approach in economic model predictive control and optimal energy management. We illustrate our theoretical developments with a benchmark chemical process application.

Modeling, dynamics and control of process networks with high energy throughput

This paper studies the energy dynamics of process networks and/or staged processes in the presence of large energy input and output flows. A generic model structure is developed for such networks and singular perturbation arguments are used to document that the variables in the energy balance evolve over a short time horizon while the variables in the material balance may exhibit both fast and slow transients. Nonlinear reduced order models for the fast and slow dynamics that can be used for model based controller design are also derived. A high-purity distillation column and a reactor with an external heat exchanger are used as simulation case studies to illustrate the theoretical results.

Integrative FourD omics approach profiles the target network of the carbon storage regulatory system.

Abstract Multi-target regulators represent a largely untapped area for metabolic engineering and anti-bacterial development. These regulators are complex to characterize because they often act at multiple levels, affecting proteins, transcripts and metabolites. Therefore, single omics experiments cannot profile their underlying targets and mechanisms. In this work, we used an Integrative FourD omics approach (INFO) that consists of collecting and analyzing systems data throughout multiple time points, using multiple genetic backgrounds, and multiple omics approaches (transcriptomics, proteomics and high throughput sequencing crosslinking immunoprecipitation) to evaluate simultaneous changes in gene expression after imposing an environmental stress that accentuates the regulatory features of a network. Using this approach, we profiled the targets and potential regulatory mechanisms of a global regulatory system, the well-studied carbon storage regulatory (Csr) system of Escherichia coli, which is widespread among bacteria. Using 126 sets of proteomics and transcriptomics data, we identified 136 potential direct CsrA targets, including 50 novel ones, categorized their behaviors into distinct regulatory patterns, and performed in vivo fluorescence-based follow up experiments. The results of this work validate 17 novel mRNAs as authentic direct CsrA targets and demonstrate a generalizable strategy to integrate multiple lines of omics data to identify a core pool of regulator targets.

Data cleaning in the process industries

Abstract In the past decades, process engineers are facing increasingly more data analytics challenges and having difficulties obtaining valuable information from a wealth of process variable data trends. The raw data of different formats stored in databases are not useful until they are cleaned and transformed. Generally, data cleaning consists of four steps: missing data imputation, outlier detection, noise removal, and time alignment and delay estimation. This paper discusses available data cleaning methods that can be used in data pre-processing and help overcome challenges of “Big Data”.