During turning of quenched and tempered AISI4140 surface layer states can be generated, which degrade the lifetime of manufactured parts. Such states may be brittle rehardened layers or tensile residual stresses. A soft sensor concept is presented in this work, in order to identify relevant surface modifications during machining. A crucial part of this concept is the measurement of magnetic characteristics by means of the 3MA-testing (Micromagnetic Multiparameter Microstructure and Stress Analysis). Those measurements correlate with the microstructure of the material, only take a few seconds and can be processed on the machine. This enables a continuous workpiece quality control during machining. However specific problems come with the distant measurement of thin surface layers, which are analyzed here. Furthermore the scope of this work is the in-process-measurement of the tool wear, which is an important input parameter of the thermomechanical surface load. The availability of the current tool wear is to be used for the adaption of the process parameters in order to avoid detrimental surface states. This enables new approaches for a workpiece focused process control, which is of high importance considering the goals of Industry 4.0.

The paper addresses the investigation of microstructures from AISI 52100 and AISI 4140 in hardened as well as in quenched and tempered conditions. The specimens are compared in terms of their magnetic hysteresis and their microstructural and mechanical properties. Material properties were determined by hardness, microhardness, and X-ray diffraction measurements. Two different approaches were used to characterize magnetic properties via a hysteresis frame device, aiming, on the one hand, to record the magnetic hysteresis with established proceedings by setting a constant magnetic flux and, on the other hand, by offsetting a constant field strength to facilitate reproducibility of the results with other micromagnetic measurement systems. Comparable differences in both the micromagnetic and the mechanical material properties could be determined and quantified for the specifically manufactured specimens. The sensitivity of the magnetic hysteresis and, determined from that, the relationship between magnetic flux and magnetic field strength were confirmed. It was shown that a consistent change in hysteresis shape from hardened to high temperature tempered material states develops and that this change allows the characterization of different materials without the need to adjust magnetization parameters. Repeatedly, an increase in remanence with decreasing hardness was found for both test approaches. Likewise, a decreasing coercivity and increasing maximum magnetic flux could be detected with decreasing retained austenite content. The investigated correlations should thus contribute to the calibration of comparable measurement systems through the holistic characterized specimens.

This study aimed to explore the feasibility of using airborne acoustic emission in laser beam butt welding for the development of an automated classification system based on neural networks. The focus was on monitoring the formation of joint gaps during the welding process. To simulate various sizes of butt joint gaps, controlled welding experiments were conducted, and the emitted acoustic signals were captured using audible-to-ultrasonic microphones. To implement an automated monitoring system, a method based on short-time Fourier transformation was developed to extract audio features, and a convolutional neural network architecture with data augmentation was utilized. The results demonstrated that this non-destructive and non-invasive approach was highly effective in detecting joint gap formations, achieving an accuracy of 98%. Furthermore, the system exhibited promising potential for the low-latency monitoring of the welding process. The classification accuracy for various gap sizes reached up to 90%, providing valuable insights for characterizing and categorizing joint gaps accurately. Additionally, increasing the quantity of training data with quality annotations could potentially improve the classifier model’s performance further. This suggests that there is room for future enhancements in the study.

With the increasing power and speed of laser welding, in-process monitoring has become even more crucial to ensure process stability and weld quality. Due to its low cost and installation flexibility, acoustic process monitoring is a promising method and has demonstrated its effectiveness. Although its feasibility has been the focus of existing studies, the temporal resolution of acoustic emissions (AE) has not yet been addressed despite its utmost importance for realizing real-time systems. Aiming to provide a benchmark for further development, this study investigates the relationship between duration and informativeness of AE signals during high-power ($3.5$ $\mathrm{k}$$\mathrm{W}$) and high-speed (12 $\mathrm{m}$/$\min$) laser beam butt welding. Specifically, the informativeness of AE signals is evaluated based on the accuracy of detecting and quantifying joint gaps for various time windows of signals, yielding numerical comparison. The obtained results show that signals can be shortened up to a certain point without sacrificing their informativeness, encouraging the optimization of the signal duration. Our results also suggest that large gaps (>$0.3$$\mathrm{m}$$\mathrm{m}$) induce unique signal characteristics in AE, which are clearly identifiable from 1 $\mathrm{m}$$\mathrm{s}$ signal segments, equivalent to $0.2$$\mathrm{m}$$\mathrm{m}$ weld seam.

This study aimed to explore the feasibility of using airborne acoustic emission in laser beam butt welding for the development of an automated classification system based on neural networks. The focus was on monitoring the formation of joint gaps during the welding process. To simulate various sizes of butt joint gaps, controlled welding experiments were conducted, and the emitted acoustic signals were captured using audible to ultrasonic microphones. To implement an automated monitoring system, a method based on short-time Fourier transformation was developed to extract audio features, and a convolutional neural network architecture with data augmentation was utilized. The results demonstrated that this non-destructive and non-invasive approach was highly effective in detecting joint gap formations, achieving an accuracy of 98%. Furthermore, the system exhibited promising potential for low latency monitoring of the welding process. The classification accuracy for various gap sizes reached up to 90%, providing valuable insights for characterizing and categorizing joint gaps accurately. Additionally, increasing the quantity of training data with quality annotations could potentially improve the classifier model's performance further. This suggests that there is room for future enhancements in the study.