Catálogo de publicaciones - libros

This chapter considers the identification and control of free-radical polymerization reactors. A discussion of the modeling and simulation of such reactors is followed by an optimal control study that demonstrates the potential of optimal control of the molecular-weight distribution based on mechanistic models. Achieving this potential in a batch reactor requires an accurate estimation of the free-radical polymerization kinetic parameters. The remainder of the chapter describes an experimental investigation of the free-radical polymerization of methyl methacrylate, in which modern sensing techniques are used to estimate kinetic parameters. The monomer conversion is measured using inline ATR-FTIR spectroscopy and robust chemometrics, and the molecular-weight distribution is measured by gas-permeation chromatography. The resulting parameter estimates and confidence intervals are used to discuss the importance of various reactions in the free-radical polymerization reaction mechanism. Discrepancies between theory and experiments are discussed.

To obtain maximal return from a mining operation it is important that the sequence of mining steps be carefully planned. In this chapter we show how this problem can be converted into a closed-loop receding-horizon optimal control problem. Of particular interest is the formulation of the associated optimization problem in the face of uncertainty, , future ore prices. We show how one can formulate “open-loop”, “reactive” and “closed-loop” policies to deal with price uncertainty. A “toy” example is presented to give insight into the problem. Also, a realistic mine-planning exercise is briefly described to highlight discrepancies between theory and practice.

This chapter presents an application of advanced control techniques on a copper smelter. The main objective of the control strategy is to keep the gas-circuit pressure at its desired value while achieving energy saving. Another objective of the control strategy is to reduce the risk of emissions. This chapter describes the design and implementation of the gas-circuit control.

The design phase includes an identification procedure. This is a multivariable process where a thorough analysis is needed for input-output matching. The identification phase included the determination of the best input-output pairing.

The control strategy has been devised taking into account not only system performance but also implementation issues. The designed controller runs on a distributed control system (DCS) using the available single-loop blocks and is able to perform a predictive control strategy with feedforward action using existing PID and lead-lag blocks.

(RTC) of sewer-network systems plays an important role in meeting increasingly restrictive environmental regulations to reduce release of untreated wastewater to the environment. This chapter presents the application of hybrid model predictive control (HMPC) on sewer systems. It is known from the literature that HMPC has a computational complexity growing exponentially with the size of the system to be controlled. However, the average solution time of modern (MIP) solvers is often much better than the predicted worst-case-solution time. The problem is to know when the worst-case computational complexity appears. In addition to presenting the application, a secondary aim of the chapter is to discuss the limits of applicability due to real-time constraints on computation time when HMPC is applied on large-scale systems such as sewer networks. By using a case study of a portion of the Barcelona sewer system, it is demonstrated how the computational complexity of HMPC appears for certain state and disturbance combinations.

Identification methods for NARMAX models are applied to aeroelastic dynamics and its properties demonstrated via continuous-time simulations and experimental conditions. Identification of parametric nonlinear models involves estimating unknown parameters and detecting its underlying structure. Structure computation is concerned with selecting a subset of parameters to give a parsimonious description of the system that may afford greater insight into the functionality of the system or a simpler controller design. Structure-detection methods applicable to NARMAX modeling are applied to aeroelastic dynamics. Among other methods, a least absolute shrinkage and selection operator (LASSO) technique is investigated for computing efficient model descriptions of nonlinear systems. Simulation results from a nonlinear dynamic aircraft model demonstrate that methods developed for NARMAX structure computation provide good accuracy for selection of the exact model structure from an over-parameterized model description. Applicability of the method to more complex systems such as those encountered in aerospace applications is shown by identifying parsimonious system descriptions of the F/A-18 active aeroelastic wing (AAW) using flight-test data.

To obtain maximal return from a mining operation it is important that the sequence of mining steps be carefully planned. In this chapter we show how this problem can be converted into a closed-loop receding-horizon optimal control problem. Of particular interest is the formulation of the associated optimization problem in the face of uncertainty, , future ore prices. We show how one can formulate “open-loop”, “reactive” and “closed-loop” policies to deal with price uncertainty. A “toy” example is presented to give insight into the problem. Also, a realistic mine-planning exercise is briefly described to highlight discrepancies between theory and practice.

Recent hardware advances have rendered active-vision a viable option for a diverse spectrum of applications ranging from MEMS manufacture to assisting individuals with disabilities. However, there are relatively few instances where these techniques have been successfully applied in uncontrolled environments. This can be traced to the of activevision systems designed using classical methods. In this chapter we show how a combination of robust identification and model (in)validation tools can be used to obtain models of these systems that combine the dynamics of the physical hardware and the image-processing algorithms, along with an associated worst-case uncertainty description, suitable to be used in the context of modern robust control tools. As shown here, this combination leads to systems that achieve good tracking performance under a wide range of conditions. These results are experimentally validated using a BiSight/UniSight robotic head-eye platform.

The purpose of this chapter is to apply robust identification and control techniques to perform active noise control (ANC). The identification phase is based on a control-oriented robust identification approach that considers both parametric and nonparametric descriptions of the system, and quantifies the uncertainty. The controller design compares the feedback (FB), feedforward (FF) and hybrid (FB/FF) control structures. The feedback control is synthesized and evaluated in the robust control framework. It is designed using optimal control as a mixed-sensitivity problem, and the robust performance of the loop is analyzed the structured singular value (). The FF controller is an adaptive identifier, based on the robustly normalized -algorithm. Two approaches are developed to decide what control structure is more efficient on a 4-m duct example with broadband noise. In addition, the compromises between identification and control, the inherent limitations of feedback and implementation issues in ANC are explicitly pointed out. Relations between performance, controller order, parametric/nonparametric models and digital signal processor (DSP) implementation are discussed. Theoretical and experimental results on the duct are compared. Finally the gaps that still remain between theory and practice in this type of applications, are outlined.

When using model-based controller design methodologies in order to control a plant whose model is not known, there are several practical issues to be taken into account. Theoretically, the procedure is to identify an accurate enough model of the plant and, then, apply any model-based controller design technique. But the is much more complex and issues as the presence of nonlinearities in the plant or/and in the actuators, the noise that always corrupts the data, the existing actuator constraints or computational limitations, . arise, imposing natural limitations over the achievable performance of the plant to be controlled as well as in the possible model to be experimentally obtained. In this contribution, an iterative framework has been used to overcome some of these problems. The proposed algorithm starts by using a very rough plant model and low-demanding specifications. These and the plant model are progressively improved only as needed, . until the final desired performance is achieved or no improvements are attained in consecutive iterations. This procedure somehow avoids useless effort: on the one hand, the identification stage is only a tool for controller design, not aiming at achieving a very accurate model at any working condition but just at those frequencies that are interesting for control purposes; on the other hand, the bandwidth is increased as long as the plant allows it, avoiding undesirable experimental results trying to achieve what is not possible at all.

This chapter reports on the modeling, identification and control system design of a magnetic-bearing system. The model used in order to carry out the design of the control system is first obtained in a process with two steps. In the first one, a nonlinear model based upon physical laws is derived. After a linearization step, and with a set of frequency-response samples in hand, an identification process follows that renders a linear model that adequately fits the experimental frequency-response samples.

The control-system design problem is addressed through three different techniques. The first one resorts to simplifying the multiple input multiple output (MIMO) 4 degrees of freedom (DOF) dynamics of the magnetic-bearing system to a set of single input single output (SISO) transfer functions, one per DOF. A classical phase-lead controller is designed, its performance being evaluated in terms of the phase margin. The second technique is robust control. Through considerations about the high-frequency behavior of the system, in particular the rotating beam’s flexible modes, a family of transfer functions is derived that represents the uncertain systems’ dynamics. The design deals with the problem of robustly rejecting sinusoidal disturbance that actually represent the beam’s imbalance. Finally, evolving from the design, the parameter variation nature of the system (as the rotation rate of the beam changes) is introduced in order to derive a linear parameter varying (LPV) model. This approach seeks rejection of the sinusoidal imbalance force carrying out an LPV adaptation to the online measured rotation rate. The design process is thoroughly covered with remarks concerning the discrepancies between theory and practice that appear when applying the different methodologies.