N.A. Lehtomaki, D. Castanon, B. Levy, G. Stein, N.R. Sandell, Jr., M. Athans. / "June 1983" / Bibliography: p. [23-24] / "NGL-22-009-124"

Gunter Stein. / "August 30, 1985." "October 1985" / Bibliography: leaf 15. / "This work was supported in part by Honeywell Internal IR&D, and by the NASA Ames and Langley Research Centers under Grant NAG-2-297."

The entire thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file; a non-technical public abstract appears in the public.pdf file. Title from PDF of title page (University of Missouri--Columbia, viewed on November 18, 2009). Thesis advisor: Dr. Roger Fales. Includes bibliographical references.

M.Ing. / Fuzzy controllers are increasingly being accepted by engineers and scientists alike as a viable alternative for classical controllers. The processes involved in fuzzy controllers closely imitate human control processes. Human responses to stimuli are not governed by transfer functions and neither are those from fuzzy controllers. The fuzzy approach is of course not the answer to all problems, but it can clearly be very successful, and can also be helpful to anyone involved in developing control systems. This study however is devoted to the environmental control task within greenhouses and the fuzzy approach is proposed in order to fulfil this task. To create near optimal conditions within a greenhouse for plant growth two environmental factors are proposed to be controlled namely the temperature and relative humidity. These factors are interdependent and they make the environmental control within a greenhouse a multi-variable control problem. Furthermore, the non-linear physical phenomena governing the dynamics of temperature and relative humidity in such a process makes it very difficult to model and to control using traditional techniques. Thus, it can be said that the environmental control in greenhouses is an art, that only expert growers bring to near perfection. The central theme of this study is the development of a multi-input-single-output heuristic rule-based fuzzy logic control algorithm, for environmental control within a greenhouse. This study is intended to improve existing environmental control systems by implementing this control technique. The control algorithm is tested in an experimental greenhouse and the results obtained indicate that fuzzy logic control is viable for environmental control within greenhouses.

The first experimental demonstration of feedback suppression of rotating external kink modes near the ideal wall limit in a tokamak using Kalman filtering to discriminate the n = 1 kink mode from background noise is reported. In order to achieve the highest plasma pressure limits in tokamak fusion experiments, feedback stabilization of long-wavelength, external instabilities will be required, and feedback algorithms will need to distinguish the unstable mode from noise due to other magnetohydrodynamic activity. When noise is present in measurements of a system, a Kalman filter can be used to compare the measurements with an internal model, producing a realtime, optimal estimate for the system's state. For the work described here, the Kalman filter contains an internal model that captures the dynamics of a rotating, growing instability and produces an estimate for the instability's amplitude and spatial phase. On the High Beta Tokamak-Extended Pulse (HBT-EP) experiment, the Kalman filter algorithm is implemented using a set of digital, field-programmable gate array controllers with 10 microsecond latencies. The feedback system with the Kalman filter is able to suppress the external kink mode over a broad range of spatial phase angles between the sensed mode and applied control field, and performance is robust at noise levels that render feedback with a classical, proportional gain algorithm ineffective. Scans of filter parameters show good agreement between simulation and experiment, and feedback suppression and excitation of the kink mode are enhanced in experiments when a filter made using optimal parameters from the experimental scans is used.

Thesis (Nucl.E)--Massachusetts Institute of Technology, Dept. of Nuclear Engineering, 1980. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND SCIENCE. / Includes bibliographical references. / by Abdelhamid Chenini. / Nucl.E

No description available.

A crucial component of instructional design for simulation-based training systems involves optimizing the presentation of complex material in order to maximize knowledge acquisition and application. One approach toward facilitating the learning of this complex information is to instantiate instructional strategies within the training systems themselves. However, there are few established guidelines in place which are meant specifically for real-time guidance strategies within simulation-based environments. Consequently, this study aims to apply findings from the literature on instructional information presentation to drive decisions for how to most effectively provide real-time feedback during training of simulated decision-making tasks. Research has shown that presenting text information in an auditory mode during direct instruction of operational tasks enhances learning and reduces the probability of learners experiencing cognitive overload. Similar effects have been found regarding the presentation modality of feedback during operational tasks. In the current study, this principle was extended by comparing text versus verbal real-time feedback presentation during learning of higher-level cognitive skills in a virtual environment. Participants were instructed on how to perform a simulated decisionmaking task, while receiving text, verbal or no instructional feedback in real-time, based on their performance. Participants then completed an assessment scenario in which no feedback was provided to any group. It was hypothesized that a linear relationship would exist across each of the three conditions, with the verbal group making the best decisions, followed by the text group, and then by the control group. Additionally, reduced cognitive load was expected throughout the instructional process for those receiving verbal feedback prompts compared to those receiving text prompts and the control. Analyses revealed several significant linear trends across iv conditions regarding measures of knowledge acquisition and application. The results provide support for the hypothesis that verbal real-time feedback is more effective than text during training of primarily visual tasks for the acquisition of higher-order cognitive skills such as decision making. There were no significant linear trends regarding the amount of cognitive load subjectively reported during training and assessment. The results of this study indicate that instructional systems intended to train primarily visual tasks should present real-time feedback in verbal rather than text form.

A root locus graphics routine was written in Turbo Pascal for the analysis and design of a linearized dual tank control system. The routine is a subprogram to be incorporated with an editor written by L. Fadden. This editor allows for the saving and changing of parameters to the system. The dual tank system is a good example for classical feedback control analysis. A brief description of the process and system is presented. The system may be described by linearized differential and algebraic equations. From these, a characteristic equation is derived, which gives rise to the root locus. The root locus is a plot of the poles of the closed loop system. Poles or roots of the characteristic equation are found using the Lin-Bairstow algorithm. This method may be used to solve for the zeroes of an nth degree polynomial. The root locus plotter was exercised by attempting to optimally tune the system’s controller. Corroboration of the results was provided by step response plots from the TUTSIM simulation program. Minor modifications allow the root locus plotter to run without the editor. Graphics subroutines are provided by the Turbo Graphix Toolbox. When run under the editor, the plotter is one interactive design module of the dual tank system analysis and design program. The subprogram was designed principally for user ease, error checking, and effective graphics.

A model (NONLINRK) was developed for a closed tank system under feedback control by an ideal proportional-integral-derivative controller. Under servo action the fluid level in the tank is altered from its equilibrium set point. Under regulatory action the feed pressure to the inlet valve and/or the outlet valve percentage opening are varied from equilibrium settings. The numerical model uses Gill’s fourth-order Runge-Kutta algorithm to solve the system equation. The equation was made separable by approximating an exponential factor by the tangent at the beginning of each time step in the numerical solution. NONLINRK simulation trials exhibited many characteristics of linear system including unequal offset under proportional control for the setpoint changes equal in magnitude but opposite in sign, harmonics in the response to a sine wave input on fluid level setpoint and bounded response in spite of increased gain settings. In addition, further simulation trials showed the system response converges to that of a linear system for sufficiently small setpoint of load variations. A second model using the modeling language TUTSIM provided corroboration of the results produced by NONLINRK. Proportional and proportional-integral control simulations differed by less than .1% and the models showed the same rates of convergence as the time step was decreased. Under PID control TUTSIM simulations developed severe instabilities, but NONLINRK exhibited the expected trends in the increased ability to react to a ramp function disturbance and the decrease in phase lag in response to a sinusoidal setpoint function.