HFTor - HfT Open Repository (Hochschule für Technik Stuttgart)
Not a member yet
404 research outputs found
Sort by
Testing of uncovered solar thermal collectors under dynamic conditions and identification of performance parameters - for nocturnal radiative cooling applications
No full textThis paper presents the first part of a research-work conducted at the University of Applied Sciences (HFT-Stuttgart). The aim of the research was to investigate the potential of low-cost renewable energy systems to reduce the energy demand of the building sector in hot and dry areas. Radiative cooling to the night sky represents a low-cost renewable energy source. The dry desert climate conditions promote radiative cooling applications. The system technology adopted in this work is based on uncovered solar thermal collectors integrated into the building's hydronic system. By implementing different control strategies, the same system could be used for cooling as well as for heating applications. This paper focuses on identifying the collector parameters which are required as the coefficients to configure such an unglazed collector for calibrating its mathematical model within the simulation environment. The parameter identification process implies testing the collector for its thermal performance. This paper attempts to provide an insight into the dynamic testing of uncovered solar thermal collectors (absorbers), taking into account their prospective operation at nighttime for radiative cooling applications. In this study, the main parameters characterizing the performance of the absorbers for radiative cooling applications are identified and obtained from standardized testing protocol. For this aim, a number of plastic solar absorbers of different designs were tested on the outdoor test-stand facility at HFT-Stuttgart for the characterization of their thermal performance. The testing process was based on the quasi-dynamic test method of the international standard for solar thermal collectors EN ISO 9806. The test database was then used within a mathematical optimization tool (GenOpt) to determine the optimal parameter settings of each absorber under testing. Those performance parameters were significant to compare the thermal performance of the tested absorbers. The coefficients (identified parameters) were used then to plot the thermal efficiency curves of all absorbers, for both the heating and cooling modes of operation. Based on the intended main scope of the system utilization (heating or cooling), the tested absorbers could be benchmarked. Hence, one of those absorbers was selected to be used in the following simulation phase as was planned in the research-project
An updated literature review of CO2e calculation in road freight transportation
No full textIn recent decades, increasing emphasis is being placed on sustainable freight transport. The aim of this study is to discern how detailed the calculation of CO2e emissions in road transport is. For this, the study conducted a sample review with 53 scientific sources and 121 annual reports. Only 14 out of the 53 scientific sources incorporated emissions that occur as a by-product of energy supply. Not more than 14 of the 121 companies included emissions from transportation activities of both the upstream and downstream processes. Therefore, the study highlights the necessity for new and validated models to support decision-making processes that sustainably reduce CO2e emissions
Working at Your Own Pace: Computer-based Learning for CL.
No full textIntroductory Computational Linguistics (CL) classes are often made up of students from different fields of study, most commonly CL, Linguistics or Computer Science (CS) – all of whom have different expertise and perspectives on the subject. Even among students of a single program, learning speeds and previous experience vary widely.
The teaching method of Computer-based Learning (CBL) was developed specifically to address this type of heterogeneity among students. It aims to combine the advantages of on-line and in-presence teaching: Students do self-paced work on video inputs, feedback tests and practical exercises while the professor is available in the room for questions and discussions. In this way, the method can accommodate different learning speeds as well as different levels of previous knowledge and different study backgrounds.
We demonstrate the method using the CL content of a class on Artificial Intelligence in the CS Bachelor’s program at Hochschule für Technik Stuttgart as an example. We outline the CL content and describe one class in detail. Student and teacher feedback from this class and two more classes taught through CBL show the method’s strengths and its suitedness to all phases of study, be it in the first semester, the final semesters or mid-program
Development of QGIS Plugin for Urban Energy Simulation Using 3D City Model at the City District Level
No full textIn the context of climate change, the increasing demand for energy-efficient buildings and sustainable urban development has become a pressing issue due to the significant proportion of global energy consumption and carbon dioxide (CO2) emissions attributable to the building sector. This requires a concerted effort to reduce its environmental impact, and Geographic Information System (GIS) applications are vital tools for achieving this by optimizing heat supply, calculating costs, analyzing profitability, and balancing CO2 emissions. This study aims to address the challenge of achieving energy efficiency and reducing CO2 emissions in the building sector, specifically at the district level. To this end, the research objective is to develop a QGIS plugin that can simulate urban energy demand at the district level by integrating 2D data with CityGML files and connecting QGIS to SimStadt software via API to visualize the simulated urban energy results in 3D on the Web Globe. The proposed plugin leverages the open-source QGIS tool QField to capture building conditions and connect 2D and 3D data on urban energy simulation. Supplementary to this, this plugin provides up-to-date information on energy demand, consumption, CO2 emissions, building component conditions via updating related tables in the database. Decision-makers can use this comprehensive and user-friendly tool to understand and act on the results, ultimately leading to a CO2-neutral district by 2045. The development of the QGIS plugin represents a significant step towards sustainable urban development and climate change mitigation by utilizing GIS applications for optimizing energy demand and reducing CO2 emissions in the built environment
A holistic analysis of sustainability metrics at an urban district scale
No full textThe presented work provides an overview of sustainability dimensions relevant to urban planning at an urban district level along multiple ressource axis. An analysis of innovative urban development projects and a literature review on city district sustainability were coupled with city grading schemes. Interviews with municipal stakeholders allowed to determine the extent to which these dimensions are already integrated into planning practices, and if not, what the barriers to the implementation of sustainable solutions are, and what the requirements for the development of new tools and methods to enable a holistically approach to sustainable planning are. Based to this, current planning practices prioritize a reduction of building energy demand and greenhouse gas (GHG) emissions during the use phase, with limited or no consideration of dimensions such as GHG from building construction, rainwater retention or urban green for optimizing urban microclimates. However, awareness for such dimensions is rising: this can be seen in pilot projects where the scope of assessment was expanded to include for instance aspects of mobility and rainwater management. It furthermore shows that an approach to optimise planning procedures along multiple dimensions is still absent from many urban planning processes, even though mindsets seem to be shifting gradually
Das Gesundheitsangebot „ReMind“ – ein Reminder für mehr Resilienz und Mindfulness im Studium
No full textIm folgenden Artikel steht die Stärkung der Resilienz von Studierenden zur Förderung der psychischen Gesundheit im Vordergrund. In ihrem Artikel Das Gesundheitsangebot ‚ReMind‘ Ein Reminder für mehr Resilienz und Mindfulness im Studium beschreibenKatrin Allmendinger, Joana Julie Scheppe & Klara Ehrmann ein ganzheitliches, hochschulübergreifendes und niederschwelliges Gesundheitsangebot für Studierende in Baden-Württemberg und werben dafür, diese Themen auch in Zukunft hochschulübergreifend anzugehen
Comparative Assessment Of Life Cycle Aspects Of Lightweight And Conventional Structures And Their Integration Into An Educational Approach
No full textThe article provides information on how teaching sustainability can be applied in a practical context. A comparison of the life cycle of two roof variants and their evaluation shows how the teaching concept can be implemented
Testbed-18: Building Energy Data Interoperability Engineering Report
No full textWithin the context of annual global energy consumption, buildings are one of the most significant consumers. Regionally focused, building related analyses are important in arriving at data driven climate change policy that reduces the overall energy consumption of buildings and provides other benefits to society. Policymakers need simple tools that enable them to conduct rapid visualization-driven assessments of how new or existing building related climate change policies will impact the energy and emissions profiles of the building stock within specific geographic regions (i.e. a city, or a province). The component geospatial and building energy datasets needed to build these tools are available globally from many sources, but the datasets exist in different political, data, and expertise domains. The lack of dataset interoperability and lack of communication between each dataset’s stakeholders results in duplication of effort, lost potential for energy savings, and lost opportunities for effective policy tools. Standards that enable interoperability of building energy and geospatial datasets would therefore address key barriers to integrated planning activities and tools for policymakers.
As the transformation to spatial computing continues, a key potential use case for data interoperability is the Canadian Geospatial Data Infrastructure (CGDI). The CGDI is an infrastructure for geospatial information. The CGDI concept is similar to the traditional physical infrastructure that helps Canadians with their everyday lives (e.g., roads, utilities, and telecommunications). The development of RESTful Web APIs, enhancements in the representation of 3D building data, and the growing availability of modeled energy data have allowed the rudimentary extension of some CGDI capabilities to be used within the domain of buildings and energy. Providing a fully functioning standards based energy and utilities component of the CGDI, enabled by interoperability and organizational communication, would go a long way to enabling municipal and utility entities to implement better energy and emissions policy and programs and work towards implementing Urban Digital Twins
Prototype Softmax Cross Entropy: A New Perspective on Softmax Cross Entropy
No full textIn this work, we consider supervised learning for image classification. Inspired by recent results in the field of supervised contrastive learning, we focus on the loss function for the feature encoder. We show that Softmax Cross Entropy (SCE) can be interpreted as a special kind of loss function in contrastive learning with prototypes. This insight provides a completely new perspective on cross entropy, allowing the derivation of a new generalized loss function, called Prototype Softmax Cross Entropy (PSCE), for use in supervised contrastive learning.
We prove both mathematically and experimentally that PSCE is superior to other loss functions in supervised contrastive learning. It only uses fixed prototypes, so no self-organizing part of contrastive learning is required, eliminating the memory bottleneck of previous solutions in supervised contrastive learning. PSCE can also be used equally successfully for both balanced and unbalanced data