Browsing by Subject "Predictive control"
Now showing 1 - 5 of 5
Results Per Page
Sort Options
Item Experimental comparison of advanced control strategies(Texas Tech University, 1995-08) Joshi, Ninad V.The objective of this research endeavor is to compare experimentally several advanced control strategies on a heat exchanger and fluid flow system. The experimental set-up was established a few years back and consists of a shell-and-tube heat exchanger with several control valves. This heat exchanger uses steam or hot water on its shell side to heat either cold or hot or a mixture of hot and cold water passing through its tube side to a desired temperature. The apparatus also contains many pneumatic control valves for controlling the flow rates of hot or cold water or steam. An experimental comparison of three control strategies (classical PID, internal model control [IMC], and process modelbased control [PMBC]) was done a couple of years earlier. The objective of this study, along with the previous one, was to implement some advanced control strategies, and present a broad-based overall perspective on the advantages and disadvantages of different control strategies. This study picks up where the last study left off, and implemented some more control strategies under similar experimental conditions. The different advanced control strategies ultimately to be implemented were neural network-based control (both inverse and normal model), model predictive control, a combination of model predictive control and neural network-based control, and heuristic-based fuzzy logic control. Thus, as a part of this study, eight different strategies were implemented. Studies on the fuzzy logic strategy were carried out separately by another graduate student.Item Multiple-model predictive control framework for multi-input multi-output continuous processes(Texas Tech University, 2003-12) Tian, Zhenhua; Hoo, Karlene A.; Mann, Uzi; Dayawansa, Wijesuriya P.; Vaughn, Mark W.This work proposes and develops an approach to transition control of chemical plants based on the development of a state-shared model in a model-predictive control (MPC) framework. Transition control over a large operating space presents a challenging problem, especially for nonlinear multiple-input multiple-output (MIMO) constrained systems. Attractive features of a transition control structure should include rapid and stable closed-loop response. A large number of recent approaches address this issue using multiple fixed and adaptive models, and single or multiple controllers. Regardless, the controller must not only successfully regulate the plant at the initial and final operating points, but also track the reference set point during the transition. In fact, the problem can be considered in two parts - the identification of a model that estimates the plant outputs during the transition and the synthesis of a suitable controller that produces smooth and realistic control action. Satisfactory closed-loop and stable performance of the controller and nonlinear plant is inferred, if the performance of the model in closed-loop with the controller can be guaranteed. By a state-shared model, we mean a linear time-invariant model structure that is a realization of the system and driven by the measured signals - the plant outputs and the manipulated variables. The coefficient matrices in the state-shared model are selected to be a controllable pair by the designer; however, the equation that represents the measured outputs of each model is unique. The description of the measurements is embedded in the coefficient matrix. Any such model necessarily fulfills the requirement that the output of the state-shared model must be an asymptotically-correct estimate of the plant's output, if the process models were selected appropriately. The parameters of the adaptive state-shared models are modified using a stable and convergent adaptive law. By means of adapting the parameters of the equation that represents the measured outputs and switching among fixed and adaptive models, accurate estimates of the plant can be obtained. The use of the state-shared model necessarily relaxes the assumption that the fixed models cover the large operating space. The theoretical underpinnings that permit the development of the state-shared model are stated and proven. Using the state-shared model, the MPC optimal controller or an Hoo robust controller can be designed. Conditions for both controllers to produce stable closed-loop responses for certain classes of systems can be used to establish closed-loop stability in the case of transition control. Critical to the tracking problem is the identification of the transition reference trajectories. Generally, the reference trajectory is determined by experience. Transition control using a state-shared model in either an HQO or MPC framework is demonstrated on several single-input single-output and multiple-input multiple-output continuous, nonlinear processes. Finally, the state-shared model based controller design approach is demonstrated on a plant-wide scale using the widely known Tennessee Eastman (TE) plant.Item Nonlinear model predictive control of a reactive distillation column(Texas Tech University, 2004-05) Kawathekar, RohitModel Predictive Control (MPC) is an optimal-control based method to select control inputs by minimizing the predicted error from setpoint for the future. Industrially popular MPC algorithms use linear convolution models for predicting controlled variable response in future. For highly nonlinear processes, the linear MPC might not provide satisfactory performance. Nonlinear Model Predictive Control (NLMPC) employs nonlinear models of the process in the control algorithm for controlled variable response in future. Reactive distillation modeling and control poses a challenging problem because the simultaneous separation and reaction leads to complex interactions between vapor-liquid equilibrium, vapor-liquid mass transfer and chemical kinetics. Hence, reactive distillation processes are highly nonlinear in nature. Application of reactive distillation for the production of ethyl acetate is considered for this dissertation. A detailed steady-state and dynamic mathematical model of reactive distillation is developed. This model is used for control studies of the reactive distillation column. Nonlinear Model Predictive Control algorithm is developed for centralized multivariable control of reactive distillation column. The performance of NLMPC is compared with decentralized PI control structure.Item Statistical process control performance characterization under field conditions(Texas Tech University, 1997-05) Karim, Mehmud ZaglulPerformance characterization of SPC technology is necessary in order to assess the impact of potential SPC strategies and actions. In this research a simulation-based tool (SPClab) is defined, designed and developed that can be used to study SPC options and its performance characteristics, considering both iid and non-iid data streams with and without step shifts in the data stream. The step shifts can be in location and/or in dispersion. The tool has three modules (1) simulation module, (2) performance module and (3) Output/report module and is developed using Borland C++ version 3.1. The number of programming lines needed to complete the tool is 20,019. The tool was tested and demonstrated in the cases of both iid and cyclical response data streams. The results from the simulation agree closely with those of theoretical values and other investigations found in the literature for the normally distributed iid case, indicating the tools validity. The tool, SPClab, was also used to demonstrate how to assess the impact of different SPC strategies and actions by considering different sampling scheme. The adverse consequences of applying SPC models inappropriately to non-iid data streams were illustrated. An appropriate physical-covariate based modeling approach is also introducedItem Superfractionator process control(Texas Tech University, 1998-08) Hurowitz, Scott EdwardAn in-depth study is conducted regarding product composition control of superfractionators with an emphasis on control configuration selection. A propylenepropane (C3) splitter is chosen as a representative column by which to investigate superfractionator process control issues. An ethylene-ethane (C2) splitter is also investigated for comparative purposes. Detailed steady state and dynamic simulations of a C3 and C2 splitter are developed and benchmarked against industrial C3 and C2 splitter process data. These simulations are used to investigate single-ended and dual Proportional-Integral (PI) composition control. For C3 splitter single-ended PI composition control, the (L, V) configuration provides the best control performance. For C3 splitter dual PI composition control, the (L, B) and (L, V/B)configurations provide the best control performance. The (L, V) and (L, V/B) configurations are determined as optimal for dual PI composition control of the C2 splitter. The control benefits provided by the use of decoupling techniques and feedforward compensation for dual PI composition control are also investigated. An evaluation of the control benefits realized by feedforward compensation indicate that, when a material balance (product) stream is used to control composition, feedforward compensation will provide a significant improvement in composition control performance. Dynamic Matrix Control (DMC), a model-based control algorithm, is applied to the C3 and C2 splitters, and its performance is compared to that obtained by PI control. Dynamic Matrix Control generally provides control performance that is equal to or better than that obtained by PI control for unconstrained, 2x2 distillation composition control, provided that the process is adequately modeled by the DMC controller. A technique is developed for predicting closed-loop product variabilities based on a signal processing analysis of feed composition data, from which usefiil information can be extracted and used to predict closed-loop product variabilities. This technique is applied to a C3 splitter for demonstrative purposes and is shown to accurately predict the product variabilities that result from feed composition disturbances.