DAVID R. NIELSEN
M.S. in Materials Science, June, 2001
M.S. in Mechanical Engineering, 1998
Advanced Materials and Technologies Laboratory
University of Connecticut
Recipient of the OUTSTANDING MASTERS STUDENT AWARD IN THE SCHOOL OF ENGINEERING, 2001
Presently with Pratt and Whitney
MODEL-BASED CONTROL STRATEGIES FOR FLOW IN RESIN TRANSFER MOLDING OF COMPOSITE MATERIALS
Manufacturing of quality products via liquid composite molding calls for a precise control of resin permeation during the process. Resin permeation during mold filling can be controlled by steering the flowfront using multiple injection ports. The implementation of process simulations for on-line simulation-based control requires that the simulation time scales be less than the time scales of the process. In addition, the on-line monitoring and control schemes needs to encompass uncertainties in uncontrollable process variables, such as the preform permeability which adversely affect the process model's accuracy.
Toward addressing this challenge, three control architectures are explored: One controller employs process trained neural networks to model flowfront permeation, and a simulated annealing-based optimizer that optimizes both flowrate injection decisions or pressure injection decisions. This duo provides for a quick and accurate control that works in real-time. Another control scheme uses the foregoing architecture but provides preform permeability values, from a fuzzy-based model, as an additional input to the neural network-based permeation simulator. A third architecture uses direct finite difference based flow simulations in real-time control. All the controllers are implemented in a LabVIEW environment, and were systematically assessed for a wide range of processing scenarios. Present work includes using a full range of actual numerical simulations to predict flowfront permeation in real-time to provide control.
Movies of controller experiments: movie1 movie2 movie3 movie4 movie5 movie6 movie7
Publications
C-98-03: D. Nielsen, R. Pitchumani, and P. Lafferty, "An Intelligent Model-Predictive Process Control Framework for RTM," in: Proceedings of the 13th Technical Conference of the American Society for Composites, A. J. Vizzini, ed., pp. 213-225, 1998.
C-99-02: D. Nielsen and R. Pitchumani, "Intelligent Simulation-based Optimal Control of Liquid Composite Molding Processes," Conference on On-Line Sensing & Control for Liquid Molding of Composite Structures, 1999. Invited Presentation.
C-00-03: D. Nielsen and R. Pitchumani, "Real-Time Model-Predictive Control of Preform Permeation in Liquid Composite Molding Processes," in Advances for Sensing and Control of Thermal Processing in Manufacturing, ASME Edited CD Volume, National Heat Transfer Conference, Topic Area: T9-35, Paper No. NHTC2000-12158, 10 pp., 2000.
C-00-06: D. Nielsen and R. Pitchumani, "Neural Network-based Control of Preform Permeation in Resin Transfer Molding Processes with Real-time Permeability Estimation," in Proceedings of ASME Heat Transfer Division, ASME-HTD-366-3, 159-170, 2000.
J-01-02: [Abstract | Request Reprint] D. Nielsen and R. Pitchumani, "Intelligent Model-based Control of Preform Permeation in Liquid Composite Molding Processes, with Online Optimization," Composites A: Applied Science and Manufacturing, 32(12), pp. 1789-1803, 2001.
J-02-01: [Abstract | Request Reprint] D. Nielsen and R. Pitchumani, "Closed-loop Flow Control in Resin Transfer Molding Using Real-Time Numerical Process Simulations," Composites Science and Technology, 62(2), pp. 283-298, 2002.
J-02-04: [Abstract | Request Reprint] D. Nielsen and R. Pitchumani, "Control of Flow in Resin Transfer Molding with Real-time Preform Permeability Estimation," Polymer Composites, 23(2), pp. 1087-1110, 2002.
Presentations
Sponsors: