Refine
Document Type
- Conference Proceeding (23) (remove)
Institute
- Forschungszentrum Energie (5)
- Forschungszentrum Business Informatics (4)
- Forschungszentrum Digital Factory Vorarlberg (4)
- Forschungszentrum Human Centred Technologies (4)
- Josef Ressel Zentrum für Intelligente Thermische Energiesysteme (4)
- Forschungszentrum Mikrotechnik (3)
- Technik | Engineering & Technology (3)
- Department of Computer Science (Ende 2021 aufgelöst; Integration in die übergeordnete OE Technik) (2)
- Forschung (2)
- Gestaltung (1)
Has Fulltext
- yes (23) (remove)
Keywords
- Cloud manufacturing (2)
- Ontologies (2)
- Auction-based production planning (1)
- Business analytics (1)
- Controlling (1)
- Creativity (1)
- Design Thinking (1)
- Digital boardgame (1)
- Digital twin (1)
- Distributed manufacturing (1)
Power plant operators increasingly rely on predictive models to diagnose and monitor their systems. Data-driven prediction models are generally simple and can have high precision, making them superior to physics-based or knowledge-based models, especially for complex systems like thermal power plants. However, the accuracy of data-driven predictions depends on (1) the quality of the dataset, (2) a suitable selection of sensor signals, and (3) an appropriate selection of the training period. In some instances, redundancies and irrelevant sensors may even reduce the prediction quality.
We investigate ideal configurations for predicting the live steam production of a solid fuel-burning thermal power plant in the pulp and paper industry for different modes of operation. To this end, we benchmark four machine learning algorithms on two feature sets and two training sets to predict steam production. Our results indicate that with the best possible configuration, a coefficient of determination of R^2 = 0.95 and a mean absolute error of MAE=1.2 t/h with an average steam production of 35.1 t/h is reached. On average, using a dynamic dataset for training lowers MAE by 32% compared to a static dataset for training. A feature set based on expert knowledge lowers MAE by an additional 32 %, compared to a simple feature set representing the fuel inputs. We can conclude that based on the static training set and the basic feature set, machine learning algorithms can identify long-term changes. When using a dynamic dataset the performance parameters of thermal power plants are predicted with high accuracy and allow for detecting short-term problems.
Flexibility estimation is the first step necessary to incorporate building energy systems into demand side management programs. We extend a known method for temporal flexibility estimation from literature to a real-world residential heat pump system, solely based on historical cloud data. The method proposed relies on robust simplifications and estimates employing process knowledge, energy balances and manufacturer's information. Resulting forced and delayed temporal flexibility, covering both domestic hot water and space heating demands as constraints, allows to derive a flexibility range for the heat pump system. The resulting temporal flexibility lay within the range of 24 minutes and 6 hours for forced and delayed flexibility, respectively. This range provides new insights into the system's behaviour and is the basis for estimating power and energy flexibility - the first step necessary to incorporate building energy systems into demand side management programs.
Bubble column humidifiers (BCHs) are frequently used for the humidification of air in various water treatment applications. A potential but not yet profoundly investigated application of such devices is the treatment of oily wastewater. To evaluate this application, the accumulation of an oil-water emulsion using a BCH is experimentally analyzed. The amount of evaporating water vapor can be evaluated by measuring the humidity ratio of the outlet air. However, humidity measurements are difficult in close to saturated conditions, as the formation of liquid droplets on the sensor impacts the measurement accuracy. We use a heating section after the humidifier, such that no liquid droplets are formed on the sensor. This enables us a more accurate humidity measurement. Two batch measurement runs are conducted with (1) tap water and (2) an oil-water emulsion as the respective liquid phase. The humidity measurement in high humidity conditions is highly accurate with an error margin of below 3 % and can be used to predict the oil concentration of the remaining liquid during operation. The measured humidity ratio corresponds with the removed amount of water vapor for both tap water and the accumulation of an oil-water emulsion. Our measurements show that the residual water content
in the oil-water emulsion is below 4 %.
With the digitalisation, and the increased connectivity between manufacturing systems emerging in this context, manufacturing is shifting towards decentralised, distributed concepts. Still, for manufacturing scenarios manual input or augmentation of data is required at system boundaries. Especially in distributed manufacturing environments, like Cloud Manufacturing (CMfg) systems, constant changes to the available manufacturing resources and products pose challenges for establishing connections between them. We propose a feature-oriented representation of concepts, especially from the manufacturing domain, which serves as the basis for (semi-) automatically linking, e.g., manufacturing resources and products. This linking methodologies, as well as knowledge inferred using it, is then used to support distributed manufacturing, especially in CMfg environments, and enhance product development. The concepts and methodologies are to be evaluated in a real world learning factory.
Business Analytics zählt zu den Zukunftsthemen im Controlling. In der Controllinglehre spielt Analytics bisher aber nur eine untergeordnete Rolle. Der Beitrag beschreibt ein innovatives Lehrprojekt, das Studierende im Masterstudium Accounting, Controlling & Finance an der FH Vorarlberg befähigt, controllingrelevante Fragestellungen im Kontext von Business Analytics eigenständig zu beantworten. Gleichzeitig erlernen die Studierenden den Umgang mit der Open-Source-Software R.
The increasing digitalisation of daily routines confronts people with frequent privacy decisions. However, obscure data processing often leads to tedious decision-making and results in unreflective choices that unduly compromise privacy. Serious Games could be applied to encourage teenagers and young adults to make more thoughtful privacy decisions. Creating a Serious Game (SG) that promotes privacy awareness while maintaining an engaging gameplay requires, however, a carefully balanced game concept. This study explores the benefits of an online role-playing boardgame as a co-designing activity for creating SGs about privacy. In a between-subjects trial, student groups and educator/researcher groups were taking the roles of player, teacher, researcher and designer to co-design a balanced privacy SG concept. Using predefined design proposal cards or creating their own, students and educators played the online boardgame during a video conference session to generate game ideas, resolve potential conflicts and balance the different SG aspects. The comparative results of the present study indicate that students and educators alike perceive support from role-playing when ideating and balancing SG concepts and are happy with their playfully co-designed game concepts. Implications for supporting SG design with role-playing in remote collaboration scenarios are conclusively synthesised.
Introducing 3D sub-micrometer technologies based on polymers opened new possibilities of design and fabrication of photonic devices and components in 3D arrangement. 3D laser lithography is direct writing process based on two photon polymerization exhibiting high accuracy and versatility, where numerous resists and even polymer ceramic mixtures can be used. We present design and simulation of polymer based photonic components with a focus on arrayed waveguide gratings (AWG) based on optical multiplexers/demultiplexers and optical splitters. All optical components were designed for 1550 nm operating wavelength, applying two commercial photonics tools. This study creates a basis for the design of optical components in 3D arrangement, which will be fabricated by 3D laser lithography.
Immersive educational spaces
(2023)
"If only we had had such opportunities to grasp history like this when I was young" – words by an almost 80-year-old woman holding an iPad on which both, the buildings in the background and a tower in the form of a virtual 3D object, appear within reach. To "grasp" history - what an apt use of this action-oriented word for an augmented reality application built on considerations of thinking and acting in history. This telling image emerged during the first test run of the app i.appear which will be the focus of this article's considerations on the use of immersive learning environments. The application i.appear has been used in the city of Dornbirn (Austria) for a year now to teach historical content through location-based augmented reality and other interactive and multimedia technologies. After a brief description of the potential of such applications, the epistemological structure of the hosting app i.appear and its functionality will be outlined. This article will focus on the “Baroque Master Builders” tour of the hosting app that was created and tested as part of the current research.
One goal of the project described in this paper is to create learning algorithms for machines and robots that lack a precise virtual controller for correct simulations. Using a digital twin approach, the developed mixed reality application aims for an overlay of a virtual robot model with the real world counterpart using Microsoft HoloLens 2 smart glasses. The application should help users to have an inside look into the results of the learning algorithm and therefore supervise and improve those results. The main focus of this paper is the visual representation of the digital twin on the smart glasses. One of the challenges is the level of abstraction and specific use of shaders (program code defining material attributes) to help the user differentiating between virtual and real objects. Therefore different presentation methods are described and evaluated. Study results with 48 persons show that the most abstract representation (wireframe) scores lowest, whereas a half-transparent model works best.
The utilization of lasers in dentistry expands greatly in recent years. For instance, fs-lasers are effective for both drilling and caries prevention, while cw-lasers are useful for adhesive hardening. A cutting-edge application of lasers in dentistry is the debonding of veneers. While there are pre-existing tools for this purpose, there is still potential for improvement. Initial efforts to investigate laser assisted debonding mechanisms with measurements of the optical and mechanical properties of teeth and prosthetic ceramics are presented. Preliminary tests conducted with a laser system used for debonding that is commercially available showed differences in the output power set at the systems console to that at specified distances from the handpiece. Furthermore, the optical properties of the samples (human teeth and ceramics) were characterised. The optical properties of the ceramics should closely resemble those of teeth in terms of look and feel, but they also influence the laser assisted debonding technique and thus must be taken into account. In addition first attempts were performed to investigate the mechanical properties of the samples by means of pump-probe-elastography under a microscope. By analyzing the sample surface up to 20 ns after a fs-laser pulse impact, pressure and shock waves could be detected, which can be utilized to determine the elastic constants of specific materials. Together such investigations are needed to shape the basis for a purely optical approach of debonding of veneers utilizing acoustic waves.
Industrial demand side management has shown significant potential to increase the efficiency of industrial energy systems via flexibility management by model-driven optimization methods. We propose a grey-box model of an industrial food processing plant. The model relies on physical and process knowledge and mass and energy balances. The model parameters are estimated using a predictive error method. Optimization methods are applied to separately reduce the total energy consumption, total energy costs and the peak electricity demand of the plant. A viable potential for demand side management in the plant is identified by increasing the energy efficiency, shifting cooling power to low price periods or by peak load reduction.
Grey Box models provide an important approach for control analysis in the Heating, Ventilation and Air Conditioning (HVAC) sector. Grey Box models consist of physical models where parameters are estimated from data. Due to the vast amount of component models that can be found in literature, the question arises, which component models perform best on a given system or dataset? This question is investigated systematically using a test case system with real operational data. The test case system consists of a HVAC system containing an energy recovery unit (ER), a heating coil (HC) and a cooling coil (CC). For each component, several suitable model variants from the literature are adapted appropriately and implemented. Four model variants are implemented for the ER and five model variants each for the HC and CC. Further, three global optimization algorithms and four local optimization algorithms to solve the nonlinear least squares system identification are implemented, leading to a total of 700 combinations. The comparison of all variants shows that the global optimization algorithms do not provide significantly better solutions. Their runtimes are significantly higher. Analysis of the models shows a dependency of the model accuracy on the number of total parameters.
During two studies the influence of technologies on sleep were analyzed. The first one is about the effect of light on the circadian rhythm and as a consequence on sleep quality of persons in a vegetative state. The second one, which is still running, surveys the influence of several technologies on the sleep of elderly people living in a nursing home.
A trend from centralized to decentralized production is emerging in the manufacturing domain leading to new and innovative approaches for long-established production methods. A technology supporting this trend is Cloud Manufacturing, which adapts technologies and concepts known from cloud computing to the manufacturing domain. A core aspect of Cloud Manufacturing is representing knowledge about manufacturing, e.g., machine capabilities, in a suitable form. This knowledge representation should be flexible and adaptable so that it fits across various manufacturing domains, but, at the same time, should also be specific and exhaustive. We identify three core capabilities that such a platform has to support, i.e., the product, the process and the production.We propose representing this knowledge in semantically specified knowledge graphs, essentially creating three through features interconnected ontologies each representing a facet of manufacturing. Finally, we present an exemplary implementation of a Cloud Manufacturing platform using this representation and its advantages.
Varying mindsets in Design Thinking. Why they change during the process and how to nudge them
(2019)
Stress testing is part of today’s bank risk management and often required by the governing regulatory authority. Performing such a stress test with stress scenarios derived from a distribution, instead of pre-defined expert scenarios, results in a systematic approach in which new severe scenarios can be discovered. The required scenario distribution is obtained from historical time series via a Vector-Autoregressive time series model. The worst-case search, i.e. finding the scenario yielding the most severe situation for the bank, can be stated as an optimization problem. The problem itself is a constrained optimization problem in a high-dimensional search space. The constraints are the box constraints on the scenario variables and the plausibility of a scenario.
The latter is expressed by an elliptic constraint. As the evaluation of the stress scenarios is performed with a simulation tool, the optimization problem can be seen as black-box optimization problem. Evolution Strategy, a well-known optimizer for black-box problems, is applied here. The necessary adaptations to the algorithm are explained and a set of different algorithm design choices are investigated. It is shown that a simple box constraint handling method, i.e. setting variables which violate a box constraint to the respective boundary of the feasible domain, in combination with a repair of implausible scenarios provides good results.
In dieser Arbeit wird Supervised Learning verwendet, um die Zuverlässigkeit von Schweißverbindungen zu evaluieren.
Um die Schweißqualität zu bestimmen, wurden End of Life Tests durchgeführt. Für die statistische Auswertung und Vorhersage der zu erwartenden Lebensdauer, wurden die Daten basierend auf einer logarithmischen Normalverteilung und mit einer multivariablen linearen Regression modelliert. Um die signifikanten Einflussfaktoren zu identifizieren, wurde eine schrittweise Regression genutzt. Die Ergebnisse zeigen, dass das entwickelte Modell die Zuverlässigkeit und Lebensdauer der Schweißverbindung akkurat abbildet und präzise Vorhersagen liefern kann.