Among the many acid-base concepts, the theory of Usanovich is one of the least known despite the most general scope including almost all chemical reaction types and even redox chemistry. Published 1939 in a Soviet journal in Russian language, it gained little immediate attention, and was later criticized mainly as being too broad in scope. Although several articles recently remembered Usanovich and his acid–base theory, one major inconsistency again was overseen: the electron is put in a row along with anions. Chemical history probably correctly puts this concept aside, also because it added little explanation capabilities beyond the elaborated considerations of the simultaneously published acid–base theory of Gilbert N. Lewis which was later refined by Pearson (hard and soft acids and bases, “HSAB”). A modified version of the core of Usanovich' concept is finally discussed. It combines the classic protic and aprotic acid–base concepts on the foundations of Lewis’ and Pearsons ideas.

This study explores the influence of uncertain boundary climate conditions on the hygrothermal performance of an internally insulated historic masonry wall using numerical simulations. The research compares diverse internal and external climate data sources to evaluate their reliability. A pre-validated hygrothermal simulation model serves as the benchmark for comparing simulated data with actual monitoring data. An array of climate data sources, including adaptive indoor climate models defined in the EN 15026 and UNI EN ISO 13788 standards, Typical Meteorological Years, and ground weather station data are considered. The core assessment parameters are temperature and relative humidity values beneath the insulation. Unexpectedly, the findings reveal that external climate conditions have a minor influence on the simulation results. Conversely, internal climate conditions significantly impact the outcomes, causing substantial variations. These implications underline the criticality of selecting an appropriate indoor climate model and moisture load class. The incorrect choice can lead to substantial errors, with peak relative humidity values predicted by the models varying in a range greater than 15 percentage points of relative humidity. In conclusion, the study reveals that utilizing Typical Meteorological Years and adaptive indoor climate models still yields excellent results, despite the inherent uncertainties. Moreover, this study emphasizes the importance of carefully selecting suitable indoor climate models to enhance the accuracy of hygrothermal simulations in historic buildings and underlines the need for future research focused on developing more precise guidelines for identifying the correct moisture load classes.

Learning videos enjoy great popularity in a digitalized world, especially since their use is usually possible regardless of time and location. Learners use this advantage mainly in self-study. Supervision, as for example in classroom teaching, is rather difficult and learners are usually left to their own devices when learning obstacles arise. However, not treating learning obstacles can have serious consequences, ranging from a gradual loss of the learners’ motivation to the termination of the learning project. Consequently, learning obstacles must be identified and treated in order to support an efficient learning process. Fortunately, a digital learning environment opens up many opportunities to support learners automatically. This paper explains an approach to identify potential learning obstacles in video learning based on indirect feedback. The first part of the approach to removing learning obstacles in learning videos is based on an analysis of learners' click interaction within a video to identify potential problem areas. Building on this, the second part provides first thematically relevant ad hoc video recommendations for the potentially identified learning obstacle. In order to verify whether there is actually a learning obstacle, the third part explicitly induces learners to give indirect or direct feedback on whether the recommendations have helped them and, consequently, whether they have removed an actual learning obstacle.

Teleoperation is a fundamental feature for the development of autonomous vehicles. Not only offers it the possibility of an efficient fallback solution for emerging situations that the autonomous vehicle cannot resolve itself but also it potentially offers new business models for transportation service providers. Fundamental to these considerations, however, is the performance of the communication network and the resulting parameters such as latency and bandwidth. These quantities have a critical impact on the ability of a human to perceive the environment, assess the situation and remotely control the vehicle safely and reliably. In this paper, a framework for simulating and evaluating the impact of network parameters on the performance of a remote human driver is presented. Using an example scenario of a shuttle bus route, the effects of latency for the teleoperation use case is evaluated. The local network and the impact on the human driver are modeled in SUMO (Simulation of Urban MObility) and OMNEST. The latency network simulation is validated against real measurement data. The car-following model Extended Intelligent Driver Model (EIDM) and its human driver model are extended so that latency effects can be quantified. The proposed approach has multiple advantages, on the one hand, it allows to evaluate the network suitability for the teleoperation use case. On the other hand, it limit the number of variables to be evaluated in e.g. user studies.

On-site current-voltage (IV) measurements will play an essential role in the online monitoring of PV systems. However, challenging measurement conditions like inconsistent irradiance levels on PV arrays (e.g., due to local shading) can distort IV curves, leading to inaccurate characterizations. By accurately detecting deformed IV curves, the reliability of both on-site and remote IV measurements is significantly enhanced. For this purpose, several classifiers were evaluated using 4104 manually labeled IV measurements on a mc-Si-PV array. Machine learning tech-niques perform much better than a traditional rule-based filter, with accuracy above 99 %. A deep Autoencoder was employed to reduce IV measurements into a set of 7 features, which encoded the shape of the curves into a low dimen-sionality. The IV-Autoencoder improved the classification of IV curves, yielding better results than a feature reduction with Principal Component Analysis. The proposed classifiers are able to sort out on-site IV measurements under un-satisfactory environmental conditions, benefiting the online monitoring of PV systems. It may also be used as an indi-cator for faulty PV strings.

Hochschulen sind engagierte Demonstratorinnen der Transformation (HRK, 2018). Im Sinne des Whole Institution Approach nachhaltigen Handelns an Hochschulen müssen sie in der Lehre innovative Lehr-Lern-Arrangements erarbeiten, um Zukunftsgestalter:innen auszubilden. Die Formate dienen neben der Sensibilisierung für transformatives Handeln auch der Wissensvermittlung und dem Fördern praktischen Handelns. In einem Lehr-Lern-Festival („Impact 23“) an der Hochschule Coburg erarbeiteten entsprechend Studierende praktische Lösungen für regionale Auftraggeber:innen, um den Herausforderungen der Transformation zu begegnen. Der folgende Beitrag schildert die Erfahrungen aus „Impact 23“ und leitet vielfältige Handlungsempfehlungen für die Hochschullehre ab.