Patricio Luppi

Decentralized plantwide control strategy for large-scale processes. Case study: Pulp mill benchmark problem

Abstract The plantwide control (PWC) complexity increases for highly-integrated and large-scale chemical processes. This work presents a novel framework for decentralized PWC which includes: (i) the selection of the controlled variables (CVs), (ii) the pairing between the manipulated variables (MVs) and the CVs, and (iii) the determination of the controller algorithms as well as their tuning parameters for closed-loop operation. The proposal is to solve the steps (i) and (ii) simultaneously, driving the selection of the most effective PWC structure from a Pareto optimal set. Here, algorithms based only on steady-state information are considered to give a systematic procedure which tries to minimize the use of heuristic considerations. Genetic algorithms (GA) and the Hungarian algorithm (HA) are used here because they provide a good trade-off between computational effort and acceptable results. The proposed methodology is completely tested in a pulp mill benchmark and compared with a previous one.

Real-Time Optimization via Modifier Adaptation Integrated with Model Predictive Control

Abstract In order to deal with plant-model mismatch, real-time optimization schemes use some adaptation strategy based on measurements. The modifier-adaptation approach is to correct the constraints and gradients in the optimization problem by adapting the values of bias modifiers expressing the difference between the constraints and gradients of the plant and the model. The approach has the ability to converge to the plant optimum but does not guarantee feasibility prior to convergence if the evaluated optimal inputs are applied directly to the plant. In this paper, an approach for integrating modifier adaptation with model predictive control is presented where both automation layers use the same decision variables. The control targets are included as equality constraints in the real-time optimization problem, and are continuously enforced by the model predictive controller. In order to apply modifier adaptation, new gradient modifiers are defined that correct the projected gradients in the tangent space of the equality constraints, rather than the full gradients. An illustrative case study shows the applicability of the approach.

Energy Management of a Hybrid Controlled Process with Renewable Sources

Abstract This work focuses on the multivariable control of a hydrogen production plant from bio- ethanol to feed a fuel cell and solar energy power generation system. The concept of combining bio-ethanol with solar energy to vehicular applications is new. A central power bus structure to receive and give the power generation is proposed. So the designed Energy Management Strategies (EMS) must act at every instant to achieve the most efficient way to handle the available energy sources. Some assumptions are adopted for defining the conceptual engineering of an integrated dynamic model of this system. It includes solar subsystem, the bio ethanol processor system (BPS), as well as a battery bank and supercapacitors to best satisfy the electric motor power demand. Several simulations are carried out to illustrate the applicability and effectiveness of the proposed design and strategies. The solar profile and standard driving cycles for Rosario city in Argentine are considered for the tests. In addition, the hydrogen quality for the fuel cell is guaranteed by improving the base conventional control strategy with Model Predictive Control.

Global Optimization for Flexible Heat Exchanger Network Synthesis of Chemical Plant

Abstract In this work a new systematic methodology of global optimization to perform synthesis of a flexible heat exchanger network (HEN) is presented. It is based on a previous model, named SynFlex, which uses a promising superstructure to obtain flexible HEN. The synthesis is projected to operate over a specified range of expected variations in the inlet temperatures and flow rates of the process streams. The main objective of SynFlex is that the total annual cost (TAC) involving the utility consumption and the investment be optimized simultaneously. The framework is based on the operability analysis where the design variables are chosen and favors to follow with the control structure design as a next step. Hence, SynFlex is enhanced by introducing a new special-purpose global optimization procedure, which relies on turning as convex specific non convex terms. Two examples are given in this work to illustrate the achievements since the SynFlex model can exclude optimal configurations due to the isothermal mixing assumption. Specially, the obtained application results from a hydrogen production plant from bioethanol with fuel cell, illustrates about the global optimization and computational efficiency for this complex case study.

Improvements on hydrogen production efficiency based on switching multiple renewable power sources

Abstract This work focuses on the Energy Management Strategies (EMS) of a stand-alone hybrid wind-solar energy generation system which provides power to produce hydrogen from bio-ethanol. This concept is new, so some assumptions are adopted for the construction of an integrated dynamic model of this system. It allows to develop an EMS and test if it is able to properly coordinate the wind and solar subsystems as well as a battery bank to best satisfy the power demand of the hydrogen production and later storage. Simulations are carried out to illustrate the applicability and effectiveness of the proposed EMS design.

Synthesis of Flexible Heat Exchanger Networks Integrated with Reconfigurable Control Design

Abstract In this work, a new systematic methodology, which extends the synthesis of a flexible heat exchanger network (HEN),proposed by Braccia et al. (2013), with a designing technique of a fully-decentralized reconfigurable control structure(CS) given by Luppi et al. (2014), is presented. The strategy to design the HEN allows the optimal flexible operation inside an expected range of the plant conditions. The CS is done in the context of reconfigurable control structure which are characterized by: (i) the integration of the variables selection task (CVs and MVs) with the corresponding pairings definition (CVs-MVs), (iii) the consideration of scalar indexes to set a rank of the potential solutions. The advantage of the method is that provides alternative control structures able to act at different operating points. In addition, it only utilizes steady-state process information. One example is given in this work to illustrate the potentiality of the proposed methodology and to support the final conclusions.

Control Oriented Dynamic Model of the Bio-Ethanol Processor System

In this chapter the pseudo-dynamic rigorous model of the bio-ethanol processor system (BPS) and proton exchange membrane fuel cell (PEMFC), working at the optimal efficiency point, is presented. It is based on the steady state HYSYS model given in the previous chapter and is able to deal with this kind of integrated plant for proposing and testing a suitable control structure. The model can be implemented thanks to the use of a specific communication protocol between the programs HYSYS and MATLAB to coordinate the calculations. The heat integration configuration is the same given in Chap. 9. The optimal heat exchangers network is initially considered with a faster dynamic than the rest of units of the plant so as to simplify the solution of the problem. It allows assuming that the stream temperatures are instantaneously at their optimum values. Moreover, the dynamic behavior of the reactors and the PEMFC are modeled in MATLAB. The principal objectives given previously for the BPS and PEMFC can be tested here through the dynamic simulations. They are performed using the critical control loops defined through the sensitivity analysis presented in the previous chapter. The controllers tuning for doing the preliminary tests are done thanks to the obtention of black box models, using the simulated plant excited with step changes and recording the reaction curves data.

Improvements on Model Predictive Control for a Pulp Mill Process

Abstract This work constitutes a contribution to the previous one presented by Castro and Doyle (2004a). They decided the incorporation of four Model Predictive Control (MPC) for specific parts of the complex chemical Pulp and Paper plant to improve its global dynamic and economic performance. Meanly the authors supported the decision of including MPC based on the RGA information. In this paper, a deep analysis about each MPC implementation is performed so as to test if the used methodology could guide efficiently for adopting this kind of decisions. Initially, the study begins with a systematic procedure for adjusting the key MPC tuning parameters. The economic and dynamic performance indexes are evaluated to demonstrate for which specific cases a real benefit can be achieved. The results presented here were obtained through dynamic simulations using the computational benchmark model of 8200 states for the same scenarios evaluated by Castro and Doyle (2004b).