The CODATA Data Science Journal is a peer-reviewed, open access, electronic journal, publishing papers on the management, dissemination, use and reuse of research data and databases across all research domains, including science, technology, the humanities and the arts. The scope of the journal includes descriptions of data systems, their implementations and their publication, applications, infrastructures, software, legal, reproducibility and transparency issues, the availability and usability of complex datasets, and with a particular focus on the principles, policies and practices for open data. All data is in scope, whether born digital or converted from other sources.

The Collaborative Research Center 1153 is investigating an innovative process chain for the production of hybrid components. The hybrid workpieces are first joined and then formed by cross-wedge rolling. Pinion shafts were manufactured to investigate the behavior of the joining zone under increased complexity of the forming process. For this purpose, six types of workpieces produced by three types of joining processes were formed into pinion shafts. The reference process provides a shaft with a smooth bearing seat. It was found that the increased complexity did not present any challenges compared to the reference processes. A near-net shape geometry was achieved for the pinions made of steel.

Hybrid components, made of multiple materials, can meet the increasing demands for lightweight construction and functional integration in the automotive and aircraft industry. Hybrid semi-finished components are produced by applying a high-alloy cladding to a low-alloy base material before hot-forming and machining the workpiece. Throughout this process chain, workpiece deviations in the form of material distribution and material properties can occur that influence the component’s lifetime. This paper investigates whether such workpiece deviations can be detected within the process chain by analyzing process signals obtained from subsequent process steps. For this purpose, artificial workpiece deviations were introduced to hybrid semi-finished workpieces made of C22.8/X45CrSi9-3. Then, process signals during forming and machining were analyzed to determine their sensitivity to the artificial deviations. The results revealed that deviations in cladding size can be effectively monitored using signals from both forming and machining. Cladding position deviations can only be detected during machining, while forming signals are more responsive to detecting the introduced hardness deviations of approx. 100 HV0.1.

The Collaborative Research Center 1153 is investigating a novel process chain for manufacturing high-performance hybrid components. The combination of aluminum and steel can reduce the weight of components and lead to lower fuel consumption. During the welding of aluminum and steel, a brittle intermetallic phase is formed that reduces the service life of the component. After welding, the workpiece is heated inhomogeneously and hot-formed in a cross-wedge rolling process. Since the intermetallic phase grows depending on the temperature during hot forming, temperature control is of great importance. In this paper, the possibility of process-integrated contact temperature measurement with thin-film sensors is investigated. For this purpose, the initial temperature distribution after induction heating of the workpiece is determined. Subsequently, cross-wedge rolling is carried out, and the data of the thin-film sensors are compared to the temperature measurements after heating. It is shown that thin-film sensors inserted into the tool are capable of measuring surface temperatures even at a contact time of 0.041 s. The new process monitoring of the temperature makes it possible to develop a better understanding of the process as well as to further optimize the temperature distribution. In the long term, knowledge of the temperatures in the different materials also makes it possible to derive quality characteristics as well as insights into the causes of possible process errors (e.g., fracture of the joining zone).

Industrial process monitoring has a significant impact on ensuring quality and efficiency, particularly in complex manufacturing processes such as cross-wedge rolling. This study evaluates the performance of four deep learning approaches—Artificial Neural Networks (ANN), Convolutional Neural Networks (1D-CNN), Autoencoders, and Recurrent Neural Networks (RNN) using Long Short-Term Memory (LSTM)—for detecting deviations during the cross-wedge rolling of hybrid workpieces. Given the challenge of a small dataset, data augmentation techniques are applied to enhance model training. The investigation aims to determine whether deep learning can effectively classify process deviations and which architecture performs best. The results indicate that 1D-CNN models achieve the highest classification accuracy within the test group. Additionally, various hyperparameters are analyzed to identify those with the most significant impact on model performance. The findings contribute to the optimization of deep learning-based process monitoring, highlighting the potential of 1D-CNN and Autoencoder models for industrial applications.