Browsing by Author "McAvoy, Thomas J."
Now showing 1 - 14 of 14
Results Per Page
Sort Options
Item An Analytical Approach to Approximate Dynamic Modeling of Distillation Towers.(1987) Kapoor, Naveen; McAvoy, Thomas J.; ISRThis paper presents analytical expressions for predicting distillate and bottoms composition transient reponses due to changes in manipulative and disturbance variables of a tower. The authors in an earlier paper showed that towers are inherently recycle structures and linearization techniques should be applied at perturbed conditions of tower models to evaluate realistic tower time constants. In this earlier paper, a numerical approach to estimating time constants was presented. The numerical approach is extended in this paper to an analytical approach that requires only steady state and design information of a tower. The analytical approach is relatively simple to use and much less time consuming compared to dynamic simulation. Further, the analytical technique gives insight into why towers respond nonlinearly.Item Analyzing Composition Control Schemes for Heat Integrated Distillation Towers Using the Relative Disturbance Gain.(1987) McAvoy, Thomas J.; Wang, Y.; ISRThe relative disturbance gain (RDG) is a recently introduced index that can be used to assess control loop operability. It is applied to the problem of controlling composition at both ends of heat integrated towers. Using the RDG the ability of the towers to respond to set point changes, feed flow, feed composition and heat input upsets is assessed. A general approach to calculating the RDG is given.Item Application of Dynamic Matrix Control to Moderate and High Purity Distillation Towers.(1987) McDonald, Karen A.; McAvoy, Thomas J.; ISRDynamic Matrix Control, a predictive computer control algorithm developed at Shell Oil Co. (Cutler, 1983) is applied to dual composition control of moderate and high purity, binary distillation towers. The composition control loops in these towers are highly interacting and, in addition, very nonlinear. The difficulty in obtaining a representative process model for nonlinear systems is demonstrated and methods to overcome this problem are suggested. Multivariable gain and time constant scheduling are evaluated using simple analytical models to update key process model parameters (process gains and time constants) on-line. This approach results in a significant improvement in regulatory control performance over the standard application of DMC.Item Application of Neural Networks on the Detection of Sensor Failure During the Operation of a Control System.(1989) Naidu, S.R.; Zafiriou, E.; McAvoy, Thomas J.; ISRNeural computing is one of the fastest growing branches of artificial intelligence. Neural Nets, endowed with inherent parallelism hold great promise owing to their ability to capture highly nonlinear relationships. This paper discusses the use of the back propagation neural net for failure cognition in chemical process systems. The backpropagation paradigm along with traditional fault detection algorithms such as the finite integral square error method and the nearest neighbor method are discussed. The algorithm is applied to an IMC controlled first order linear time invariant plant subject to high model uncertainty. Compared to traditional methods, the backpropagation technique is shown to be able to accurately discern the supercritical failures from their subcritical counterparts. The use of backpropagation fault detection systems in on-line adaptation of nonlinear plants has been investigated.Item Computer-Enhanced Analysis of Multicomponent Fluorescence Spectra.(1988) Wang, N.S.; Rinaudo, P.J.B.; McAvoy, Thomas J.; ISRLaser-induced culture fluorescence is one of the few nondestructive measurements that are currently available for the on-line monitoring of the internal state of a microorganism. Many biomolecules, including NADH, exhibit characteristic fluorescence spectra that are dependent on the excitation wavelength. Many fluorophores are simultaneously present in a typical fermentor, and the fluorescence spectra for a mixture of fluorophores must be deconvoluted to yield individual concentrations, after the noisy spectral data are treated in the Fourier domain. Because of the secondary optical effects such as the repeated re-absorption and re-emission of light in a mixture, the principle of superposition does not hold true. Furthermore, the overall emitted signals are shifted toward the higher wavelengths. A rigorous nonlinear model is developed for the most commonly used optical arrangement to relate the detected signal level to the fluorophore concentration. As a model system, the emission spectra of binary and ternary mixtures of simple aromatic amino acids (tyrosine, tryptophan, and phenylalanine) are analyzed with a range of computer algorithms. The predictions resulting from different analytical approaches are compared to the true component concentrations, and the effectiveness of each method is evaluated. In general, it is most difficult to detect a small amount of one component either in the presence of larger concentrations of other fluorophores or in the presence of an unknown fluorophore. The presence of an unknown fluorophore is determined by spectral matching and hypothesis testing. The application of culture fluorescence measurements to fermentation control will be discussed.Item An Expert Approach Towards State Estimation of Bioreactors.(1987) Minderman, Peter A.; McAvoy, Thomas J.; Gasner, Larry L.; ISRPrevious investigators have demonstrated the theory and applications of macroscopic elemental and material balances to the estimation of bioreactor states. However most previous applications have been limited to processes which have only one growth phase. An expert systems knowledge base is presented which is used to extend this empirical estimation technique to procesess which have multiple phases (e.g. a lag phase, a growth phase, and a production phase). This approach is illustrated by a simple example from the literature.Item Fluorescence Monitoring of Aromatic Compounds in a Multicomponent System.(1988) Wang, N.S.; Simmons, M.B.; Rinaudo, P.J.B.; McAvoy, Thomas J.; ISRThere have been many studies on the use of fluorescence measurements to monitor the concentrations of aromatic compounds. Recently, there are a few commercial fluorescence probes on the market that are tuned to a fixed emission fluorescence frequency. Because of the non-invasive nature, fluorescence and optical techniques are ideally suited for the on-line monitoring of aromatic compounds. A rigorous sensor model was developed for a commercial fluorescence probe to describe the single frequency excitation and emission fluorescence behavior of a mixture of fluorophores. This model is needed to correlate the measured signals to the concentration of fluorescent compounds. The relevant parameters of the model are the absorbance of the medium at both the excitation and the emission frequencies by the solvent and other absorbing species, the background signals, the light path length of the fermentor vessel, the fluorescence yield, the lamp-detector configuration, and a light-scattering coefficient. The effects of temperature, pH, and spatial inhomogeneity (bubbles and insoluble solids) are implicitly included in the model. The model shows that the relationship between the level of fluorescence signal and the concentration of the fluorophore of interest in the presence of other interferences is intrinsically nonlinear. The signal level is independent of the fluorophore concentration for the ideal case of a single fluorophore and an infinitely long light path. The validity of the model was verified experimentally with a simple system of known fluorophores and extended to a complex system containing microbial cells in an aerated fermentor.Item Laser-Induced Fluorescence Monitoring of Multicomponent Aromatic Amino Acids.(1988) Wang, N.S.; Rinaudo, P.J.B.; McAvoy, Thomas J.; ISRThere have been many studies on the use of laser induced fluorescence measurements to monitor the concentrations of aromatic compounds. The emission spectra of binary and ternary mixtures of simple aromatic amino acids (tyrosine, tryptophan, and phenylalanine) are analyzed based on a rigorous model that has been previously developed by us to relate the detected signal level to the fluorophore concentration. The fluorescence signals are generally noisy, and the contributions from each fluorophore in a mixture are not at all additive due to the secondary optical effects such as the reabsorption of the fluorescence signal and the reemission of photons at lower frequencies. Fourier transform methods are employed for noise reduction, and a range of algorithms are proposed to extract information in an optimal manner from the strongly nonlinear system. Much improved concentration estimates are obtainable over those calculated directly from the linear additivity assumption when confounding effects are considered explicitly. Our results to date show that it is most difficult to detect a small amount of one component either in the presence of larger concentrations of other fluorophores or in the presence of an unknown fluorophore. The application of on-line fluorescence measurements in bioprocessing will be discussed.Item Novel Control Strategies for Catabolite Repressible Fermentations.(1988) Minderman, Peter A.; Gasner, Larry L.; McAvoy, Thomas J.; ISRBecause of the problem of catabolite repression, a sign)ficant class of industrially important fermentations, e.g. enzyme and antibiotic production, are controlled by the fed-batch policy in which the catabolite concentration is maintained at a low, constant, and acceptable level. An improved policy, the pulsed batch policy, is proposed which takes advantage of the sign)ficantly accelerated production rates observed when a microorganism adapts from one environment to another. Experimental and modeling results are presented for the ,BETA galactosidase from Escherichia coli case.Item On-Line Optimization Using Steady State Models.(1988) Koninckx, Jan; McAvoy, Thomas J.; Marlin, T.E.; ISRMany sectors of the chemical industry suffer from a production overcapacity. Efficient production strategies are important. This situation combines with dropping computer costs to form an excellent environment for on-line optimization. If the transient periods of an operation are short relative to the steady intervals, a major part of the operation economics is determined by the steady state. Therefore, an optimization limited to the steady state will cover the main part of the attainable profit of the plant. A structure for an on-line optimizer is proposed. The optimization is conceived as a calculation of a set of optimal setpoints for the plant. The on-line optimizer is composed of a number of modules. The most important modules perform the optimization and identify the model. Although steady state models are much morre readily available and accurate than dynamic models, they still are approximate and contain parameters that have to be updated regularly to correct for the plant model mismatch and for slow changes in the plant. Sensitivity analysis of the optimization results and statistical analysis of the model identification results are combined in short-cut feasibility studies and on-line accuracy estimation. Data reconciliation improves the robustness of the application. Two examples serve as illustrations. The first example concerns a propane-propylene splitter. This example shows many of the interesting issues on a system of reduced size. The results are therefore easier to interpret. The second example is a boiler load allocation problem. This example is more involved and shows a realistic application.Item Parameter Identification and Simulation of a Thin Film Conducting Polymer Gas Sensor(2001) Li, Yong; Adomaitis, Raymond A.; McAvoy, Thomas J.; ISRThin film conducting polymer gas composition sensors can be used forindustrial safety monitoring, environmental pollution control, and numerousother applications. In this paper, a sensor system model is developed andthis model is used to compare a lumped circuit to a transmission lineformulation of the equivalent circuits describing the polymer film andelectrodes; the sensor model includes the adsorption/desorption processand the electric properties of the sensor. The simulation approachcombines physically based modeling and parameter estimation, in which asubset of model parameters are estimated using published experimentaldata. The simulation results are found to be useful for interpretingconduction polymer sensor responses to different gas species.Item Short-Cut Analysis of Pressure Control in Steam Headers.(1987) Bertrand, Christophe R.; McAvoy, Thomas J.; ISRAlmost all chemical/petrochemical plants have steam networks to supply power and heating requirements. Good control of steam systems can have important economic implications. In this paper control of pressure in the high pressure header is analyzed. A simple short-cut model is used. The steam system is represented by a third order transfer function with widely separated time constants. This model compares very well with a detailed non- linear simulation and requires only readily available process information. The question of tuning PI controllers for the integrating pressure loop is treated in detail. A root locus analysis is used to develop tuning rules. Typical upsets such as boiler and turbine trips are evaluated for various steam systems. In each case the tuning rules provide very satisfactory control performance.Item Short-Cut Operability Analysis. Part II - Estimation of f_1 Detuning Parameter for Classical Control Systems.(1987) Marino-Galarraga, Maria; McAvoy, Thomas J.; Marlin, T.E.; ISRIn this paper a short-cut tuning method for multiloop PID controllers is presented. The method is an integral part of a short-cut operability analysis that provides a quick estimate of closed loop integral error. The tuning procedure provides a simple method of calculating how much multiloop controllers have to be detuned because of interaction. Tuning each loop is based on the nature and relative speeds of its interacting partners. A steady state version of the method is also presented with recommendations on its region of applicability.Item Short-Cut Operability Analysis. Part III - Short-Cut Methodology for the Assessment of Process Control Designs.(1987) Marino-Galarraga, Maria; McAvoy, Thomas J.; Marlin, T.E.; ISRIn this paper a methodology for the evaluation and comparison of several control strategies at different stages of the design is developed. The two parts of this methodology extend the previously developed short-cut operability analysis. The first part of the methodology is based on steady state information. It uses the Relative Disturbance Gain (RDG) in conjunction with the Relative Gain Array (RGA) to distinguish between clearly unfavorable control schemes and those which might result in acceptable transient performance. Also, it uses a steady state approach to decide whether a decoupling controller is beneficial and what its structure should be. The second part of the methodology makes use of approximate dynamic information to evaluate which control structures meet the process specifications. Internal Model Control (IMC) with the assumption of a perfect process model is used as an upper bound for the performance obtainable with advanced control.