Selecting the best model for monitoring industrial machines is challenging due to limited data and varying conditions. While deep learning (DL) is often considered ideal, it does not always perform well. We compared DL with traditional methods focused on feature extraction and classification. Our tests on seven datasets revealed that traditional methods are more reliable when machine conditions change, highlighting the importance of simpler, interpretable models for industrial monitoring.

A simple and cost-effective MIP-based sensing system for the colorimetric detection of 2-methoxphenidine (2-MXP) was demonstrated. Unlike typical colorimetric systems that rely on spectrometers, this system utilizes an RGB sensor and an LED. Constructed from readily available commercial components, it can be manufactured in a standard lab environment. The system detects 2-MXP in aqueous solutions at concentrations below 1 mM and can distinguish between concentrations of 1, 0.5, 0.25, and 0.1 mM.

We developed a measurement system to improve sand core production in foundries. Sand cores are essential for creating complex shapes in metal casting. Our system uses ultrasound waves to monitor the curing process, indicating how much of the catalyst remains in the exhaust gas. This real-time monitoring allows manufacturers to optimize the process, reducing waste and enhancing product quality. We tested the system on a lab-scale core-making machine and demonstrated its effectiveness.

Photo-thermal actuation and laser Doppler vibrometry enable fast, contactless characterization of MEMS (microelectromechanical system) resonators already before electrical connections are made. This paper details a tailored setup combining a precision stage, vacuum chamber, laser diode, and vibrometer to analyze silicon MEMS devices on a wafer level. Tests reveal how silicon device layer thickness affects resonance frequency and identify dominant loss mechanisms in different vibrational modes.

Continuous odor monitoring is becoming increasingly important. Therefore, odor monitoring systems are needed that are less expensive than conventional laboratory equipment and do not require trained personnel. Here, a gas chromatography selective odorant measurement sensor array (GC-SOMSA) system is used as part of the development concept. It correlates a sensor response to a mass spectrometer signal and an odor impression. This allows for a fast and inexpensive characterization suitable for many applications.

Some air pollution gases, such as nitrogen dioxide, are toxic at very low concentrations, highlighting the need for extensive research on materials with high sensitivity. Carbon nanotubes are nanomaterials capable of sensing toxic gases in very low concentrations. Their integration into sensors remains challenging, due, in part, to their small dimensions. This work represents a study that tests a method of safely and efficiently localizing carbon nanotubes for their integration into sensor devices.

The gas-sensing performance of the CuO/Cu₂O functionalized with Au NPs towards CO₂, CO, and HC_Mix was investigated by resistance measurement. The type of ligands has an impact on the sensing behaviour of CuO/Cu₂O sensors, and the highest response (39 %) for CO₂ is achieved by Au NPs stabilized with citrate. These differences in behaviour of the films depending on their ligands could be due to insufficient thermal degradation of the ligands.

Ferromagnetic materials change their magnetic properties under load, enabling the implementation of a force sensor. The magnetic field emerging from such a sensor can be measured by secondary sensors to approximate the load acting on the sensor. A test setup simulating a potential application environment is described and its measurement results are presented. Furthermore, relevant magnetic material properties of an exemplarily chosen cold working steel are discussed.