Structured Representation of Simulation and Annotation Data for Machine Learning in Forming Technologies

Schumann, Markus; Moske, Jonas; Wüst, Antonia; Divo, Felix; Gelbich, Daria; Niemietz, Philipp

Open Access

Structured Representation of Simulation and Annotation Data for Machine Learning in Forming Technologies

Description

The use of machine learning (ML) in manufacturing requires structured, especially standardized, access to both simulation data and domain knowledge. This paper introduces a JSON-based data format for representing synthetic force-time series alongside expert annotations. The schema captures simulation metadata, tool and material parameters, and allows explicit expert knowledge, such as failure indicators, to be linked to signal segments. The proposed structure enables process-aware ML methods that leverage both domain knowledge and raw data for improved learning and generalization. A deep drawing use case illustrates how the format facilitates knowledge-guided learning. The approach aims to bridge the gap between real and simulated production data, supporting scalable integration in modern manufacturing systems. The dataset filename encodes the preprocessing configuration. <N>pts denotes the number of uniformly resampled time points per stroke. aug-gaussian-force-Fx-Ty-<k>x specifies the augmentation strategy, where Fx is the standard deviation of additive Gaussian noise applied to the force signal, Ty the standard deviation of optional time jitter, and <k>x the number of augmented copies per original stroke. tFULL indicates that the full time series is used, while tA-Bs denotes a cropped time window from A to B seconds.