FORinFPRO-HIMD: Multimodal Sensor Dataset for Hybrid Injection Molding of Continuous Fiber-Reinforced Polypropylene Composites

This dataset contains data from 50 hybrid injection molding experiments conducted at the University of Augsburg within the framework of the research project FORinFPRO (Project No. AZ-1596-23), funded by the Bayerische Transformations- und Forschungsstiftung. The experiments were performed on a hybrid injection molding demonstrator based on the Erlanger Träger, which serves exclusively as a research platform for investigating process monitoring and data-driven manufacturing approaches. The manufacturing process involves the in-situ forming of continuous glass-fiber reinforced polypropylene (PP) organosheets followed by overmolding with polypropylene in a single hybrid injection molding cycle. The experiments were carried out using an ENGEL V-Duo injection molding machine, which is part of the DFG-funded large-scale research facility "Hybrid Injection Molding Cell" (INST 94/124-1, Project No. 466727809), specifically designed for the production and investigation of hybrid composite components. The dataset comprises both machine signals and synchronized in-mold sensor data. Machine data were acquired directly from the injection molding machine, while additional sensing systems were integrated into the mold at the University of Augsburg. The included sensing modalities are: Machine process data (e.g. machine signals and process parameters), Active pulse-echo ultrasonic measurements, Cavity pressure and temperature measurements, Dielectric analysis (DEA) measurements, and Associated process metadata. The ultrasonic measurement system consists of custom high-temperature pulse-echo sensors integrated into the mold, enabling the observation of material and process-state evolution during injection, holding, and cooling phases. The multimodal dataset is intended to support research in: Process monitoring and state estimation, Sensor fusion, Hybrid injection molding of fiber-reinforced thermoplastics, Signal processing, Machine learning and representation learning, Digital twins and intelligent manufacturing systems. The dataset was generated as part of ongoing research on resilient and adaptive composite manufacturing processes and is made publicly available to facilitate reproducible research and the development of advanced monitoring and data-driven methods.