CosiMo Process Monitoring
DLR (CC BY-NC-ND 3.0).
Hier wird das Bauteil noch einmal kontrolliert
Cosimo Tooling 2
The CosiMo project aims to create a research platform for the realisation of sustainable, highly productive and economical manufacturing technologies for automotive and aero structures. This automotive research project with its focus on e-mobility expands the technology area and target group of DLR ZLP. It mainly addresses the innovative production of thermoplastic composite structures on the DLR hot press.
The aim is to manufacture automotive Automotive composite structures in a single-process step. Therefore, the advantages of an off-the-shelf resin transfer moulding (RTM) process are combined with the advantages of thermoplastic matrix systems in a thermoplastic RTM (T-RTM) process. In particular, the state-of-the-art, cost-intensive manufacturing of FRP components is optimized both economically and ecologically. Addressing these challenges, the key topics of the CosiMo project are
These key topics will be validated on a demonstrator, of which development and design is also part of the project.
Aim of the project
The following points are major goals in CosiMo for the Center of Lightweight Production Technologies (ZLP) of the German Aerospace Center (DLR):
Starting from basic scientific questions on material characteristics and process-induced material behaviour, over the development of sensor-integrated RTM tools, to an innovative process control, a wide range of physical, chemical and technological challenges is addressed by the project network. These topics are separated into the following subprojects (HAP):
HAP 1 – Tailored Nonwovens
HAP 2 – Reactive Systems
HAP 3 – Intelligent Tooling
HAP 4 – Data-driven Process Control
The research on these complex topics is conducted by companies and research institutes of distinct expertise in the named fields (see following map).
DLR ZLP work packages
The research conducted by ZLP Augsburg in the CosiMo project mainly focuses on “HAP 3 – Intelligent Tooling”. DLR ZLP is the sub-project leader of this work package. The HAP 3 work package represents the key interface between basic material characterisations (HAP 1 & 2) and data-driven process control (HAP 4). The main DLR ZLP tasks are conducted in the following work packages.
Based on a previously defined manufacturing demonstrator, sub-processes of the part manufacturing (e.g. filling simulation) and characteristics of the polymerised component (distortion simulation) are modelled with ESI PAM composites. Using these simulations, the researchers of the DLR ZLP will compare the results with real process data. In so doing, the usability of the simulation for the actual thermoplastic RTM process is investigated with the aim of identify optimisation potentials.
A sensor network is jointly developed with project partners to monitor process parameters during the polymerisation process. The network consists of pressure, temperature, ultrasound and dielectric sensors. A concept for the positioning and integration in the RTM tooling is defined. The sensor network allows tracking of the process flow and part parameters in the more complex tool geometry. It has to fulfil all requirements of the manufacturing process (e.g. reduction of defects, assurance of required process parameters such as pressure, leak tightness, homogenous temperature distribution). All these parameters are used to analyse the real polymerisation behaviour and to optimise simulation models. The main advantage of these investigations is to potentially eliminate a separate quality assurance step with the finished part.
Optimisation of process and machines
The validation of the T-RTM process is realised by the integration of an injection machine by KraussMaffei Technologies GmbH into the infrastructure of DLR ZLP. To do this, the injection machine will be combined with the DLR hot press (Wickert). In this way, safety and control concepts are consolidated and optimized for smooth machine operation.
First, the T-RTM process is demonstrated on a simple plate tooling. Machine, process and material parameters are investigated intensively and optimised for best part qualities. A further goal of the project is to enhance the machine with the aim of achieving at an automated process and simulation data-driven process control based on machine-learning methods. Finally, the manufacturing process of a complex demonstrator component is supported by machine-learning training data generated by the Institute for Software & Systems Engineering (ISSE) of Augsburg University.
The final demonstrator component is tested with a non-destructive testing method to define the gathered part qualities. To do this, several testing methods are available at DLR ZLP, such as microscopy, thermography and air-coupled ultrasound. On the one hand, gathered values for the tensile strength of FRP parts reinforced with nonwoven fabric can be compared to values of off-the-shelf FRP parts. The correlation of process-induced sensor data and resulting part qualities are also investigated. On the other hand, the part quality achieved for the demonstrator component is essential for evaluating of the industrialisation of the manufacturing process.