Image based control system for improving fiber injection molding process

Fiber injection molding is an innovative process for the resource-efficient production of near net-shape long fiber preforms. The filling of the mold is crucial for the repeatability and uniformity of the produced preforms. For improving the fiber injection process a control system based on image processing has been developed. With a camera the current mold filling is recorded and processed by artificial neural networks. This information on the filling state is used for an adaptive control of the injection nozzle. The control system is validated experimentally with results showing improved reproducibility of the fiber injection molding process.