Vent Location Optimization Using Map-Based Exhaustive Search in Liquid Composite Molding Processes

In liquid composite molding (LCM) processes, the resin is injected into the mold cavity, which contains preplaced reinforcement fabrics, through openings known as gates, while the displaced air leaves the mold through openings called vents. Under nominal conditions, the last points to fill are chosen as vent locations. However, due to imperfect preform cutting and placement, gaps and channels may form along the edges and curvatures in a mold, offering a path with less resistance for resin flow. The faster advance of resin through these gaps and channels, a common disturbance known as racetracking, will cause the last filled regions to vary, which complicates the vent selection process. In this study, probabilistic racetracking modeling is used to capture last-filled region distribution over the mold geometry. Success criteria for mold filling are defined in terms of dry spot tolerances, and vent fitness maps, which display potential vent locations, are created. Next, exhaustive search algorithm is coupled with vent fitness maps to determine optimal vent configurations. The map-based exhaustive search is demonstrated on three geometries and results are compared with existing combinatorial search results. The performance of the optimal vent configurations is evaluated in a virtual manufacturing environment. Sensitivity analyses are conducted to determine the influence of optimization parameters on the results.     

Combinatorial Search to Optimize Vent Locations in the Presence of Disturbances in Liquid Composite Molding Processes

In liquid composite molding processes, the resin is injected into a mold cavity containing preplaced reinforcement fabrics through openings known as gates, and the air leaves the mold through openings called as vents. Under nominal conditions, the last points to fill are chosen as vent locations. However, due to imperfect preform placement, gaps and channels may form along the edges and at curvatures in a mold, offering a path with less resistance for resin flow. The faster advance of resin through these gaps and channels, a common disturbance known as racetracking, will cause the last filled regions to vary, which complicates the vent selection process. We introduce a methodology that uses set and probability theories to forecast racetracking conditions in a mold and have developed a combinatorial search algorithm to locate corresponding optimal vent locations. The accuracy, efficiency and usefulness of the approach is illustrated with three case studies.