Simulation-based Design System for
Flow Control in Liquid Composite
Department of Mechanical Engineering
University of South Alabama
NSF/DOE/APC Workshop: Future of Modeling in Composites Molding Processes
June 9-10, 2004, Arlington, VA
Role of Flow Simulation in LCM Optimization
Final intuitive design GA/simulation-based design
 K.T. Hsiao, M. Devillard, and S. G. Advani, “Simulation Based Flow Distribution Network Optimization for Vacuum Assisted
Resin Transfer Molding Process,” Modeling and Simulation in Materials Science and Engineering, 12(3), pp. S175-S190, 2004.
Flow Disturbance in LCM
Small variations on the local permeability and fiber volume fraction sometimes
make the filling pattern very different and cause unexpected dry spot!
Need reliable flow control to counteract the disturbance.
1 V f P
Design LCM Flow Control with
Simulation-based Liquid Injection Control
Mesh Resin Viscosity
Fiber Volume Fraction 3. Optimally Place Sensors and Create
1. Gates/Vents Design. Database for Mold Filling Monitoring and
Permeability Characterization. [2,3]
2. Layout of Flow Runners and
4. Optimally Place Auxiliary Gates and
Flow Distribution Media.  Create Mold Filling Control Strategies .
 K.T. Hsiao and S. G. Advani, “Flow sensing and control strategies to address race-tracking disturbances in resin transfer molding---
Part I: design and algorithm development,” Composites Part A: Applied Science and Manufacturing, (in press).
 M. Devillard, K.T. Hsiao, A. Gokce, and S. G. Advani, “On-line characterization of bulk permeability and race-tracking during the
filling stage in resin transfer molding process,” Journal of Composite Materials, 37(17), pp. 1525-1541, 2003.
Case Study: Online Flow Monitoring &
Strategic (On/Off) Injection Control
TekscanTM Sensor Area
(Pressure Grid Film)
Experimental resin Disturbance Mode Implement the
arrival times 29 is selected customized
t0, t1, t2, t3, t4 are all from the Database control action for
collected Mode 29
AG1 Control action Mode 29 is taking place.
IG1 CS2 IG2 • CS1 >>> Close IG2
Initial injection gate (IG) CS3
with flow runner • CS2 >>> Open AG1
CS1 • CS3 >>> Close IG1
Fixed vent AG2 • Vent Sensor >>> Close All Gates.
Auxiliary gate (AG)
Disturbance detection sensor (DS)
Control action trigger sensor (CS)
 M. Devillard, K.T. Hsiao and S. G. Advani, “Flow sensing and control strategies to address race-tracking disturbances in resin
transfer molding---Part II: automation and validation,” Composites Part A: Applied Science and Manufacturing (submitted).
Other Types of LCM Flow Control
Simulation-based Artificial Neural Adaptive Control (Numerical
Network and Simulation-Annealing Simulations may NOT be Necessary)
Control . .
Predicted flow front
Actual flow front
Q1 Q2 Q3
 D. Nielsen, R. Pitchumani “Intelligent model-based control of preform permeation in liquid composite molding processes, with online
optimization”, Composites: Part A 32 (2001) 1789-1803.
 B. Minaie, W. Li, S. Jiang, K. Hsiao, R. Little “Adaptive Control of Non-Isothermal Filling in Resin Transfer Molding”, Proceedings of
49th International SAMPE Symposium and Exhibition, Long Beach, CA, May 16-20, 2004.
Sensors Available for LCM Flow Monitoring
•DC point sensor
•DC linear sensor
•Dielectric linear sensor Electrical Admittance?
•Optic fiber sensor
•Electric time-domain reflectometry sensor
Time of Flight?
•Tekscan sensor (pressure grid film)
Interpretation algorithms to figure out the details of LCM flow from the
limited (point, linear, 2-D) sensor feedback.
1. Reduce mold tooling/equipment cost using modular approach.
2. Reduce the process development time and cost by minimizing the use of
3. Enhance the capability of manufacturing large, complex, and net-shaped
4. Reduce the cycle time by optimally merging the mold filling stage and cure
5. Need to gain better process controllability against disturbance during
6. Need complete and rigorous heat transfer models for non-isothermal LCM
7. Include dimension tolerance modeling into LCM design.
8. Need a systematic approach to tie the final part quality with processing
9. Need reliable sensors and interpretation algorithms.
10. Reduce the portion of human factor in LCM operation.
Vision: Computer Controlled LCM System - Integration
of Process Design, Automation, and Quality Control
Fiber Preform Resin
How do we formulate the building blocks and connect them by exploiting the
knowledge of composites manufacturing, information technology and robotics?
Database for Past Processes LCM Process
Raw Material Database Server
Implementation of Process
Monitoring and Control
Challenges of the Future Integrated LCM System
System Reliability Process Simulation
•Sensor and Sensing Algorithm •Non-isothermal Molding
•Control Algorithm •3-D Simulation
•Controllability •Preform Deformation in LCM
•Algorithm/Methodology to Integrate •Micro-Voids Formation/Migration
the Design, Automation, and Quality •Residual Stress/Strain
•Influence of Defects
Process Physics •Influence of Residual Stress/Strain
•New Resins •Influence of Other Processing Parameters
•New Fillers such as Pressure, Cure Cycle, Moisture
•New Fiber/Fabric Systems Content, Mold Tools, etc.