Publication Server of Albstadt-Sigmaringen University
Not a member yet
662 research outputs found
Sort by
Using AutomationML to Generate Digital Twins of Tooling Machines for the Purpose of Developing Energy Efficient Production Systems
No full textt. The development of energy efficient production systems such as machine tools is a complex process. All specialised departments must work interdisciplinary during the design process in order to achieve an optimal result. In addition to the mechanical aspects, e.g., lightweight construction, the optimization of the Programmable Logic Controller (PLC) programs of tooling machines plays an increasingly important role. By optimizing the programs in terms of energy
efficiency, the energy consumption of a machine can be significantly reduced. However, energy consumption depends on many parameters, so the optimization process is complex and a matter of all engineering disciplines working together. By using Digital Twins of tooling machines, simulations can be used to perform many parameter studies for optimizing energy consumption. However, the generation of
Digital Twins of production systems is very expensive if they are to represent all relevant features of a production system. Through an IT system of networked software programs and using AutomationML as a special data interface these Digital Twins of machine tools can be generated automatically. The article describes the structure and function of this IT system and how it will be efficient within all
involved engineering disciplines
Power Function Error Initialization Can Improve Convergence of Backpropagation Learning in Neural Networks for Classification
No full textSupervised learning corresponds to minimizing a loss or cost function expressing the differences between model predictions yn and the target values tn given by the training data. In neural networks, this means backpropagating error signals through the transposed weight matrixes from the output layer toward the input layer. For this, error signals in the output layer are typically initialized by the difference yn- tn, which is optimal for several commonly used loss functions like cross-entropy or sum of squared errors. Here I evaluate a more general error initialization method using power functions |yn- tn|q for q>0, corresponding to a new family of loss functions that generalize cross-entropy. Surprisingly, experiments on various learning tasks reveal that a proper choice of q can significantly improve the speed and convergence of backpropagation learning, in particular in deep and recurrent neural networks. The results suggest two main reasons for the observed improvements. First, compared to cross-entropy, the new loss functions provide better fits to the distribution of error signals in the output layer and therefore maximize the model's likelihood more efficiently. Second, the new error initialization procedure may often provide a better gradient-to-loss ratio over a broad range of neural output activity, thereby avoiding flat loss landscapes with vanishing gradients
Calorie restriction affects human nucleotide excision repair and the expression of aging-associated genes
Get PDFCalorie restriction (CR) is the only intervention known to influence the lifespan of model organisms like worms, flies, yeast or rodents positively. The impact of CR on model organisms is extensively researched and molecular mechanisms have been hypothesized or even
revealed. However, studies involving human test persons were scarce in the past and mostly dealt with the improvement of health parameters like body weight, insulin sensitivity, blood pressure or cholesterol. These are beyond doubt important parameters, whose improvements
lead to an increased quality of life and a healthier aging. However, one of the most common aging-related diseases threatening a healthy aging is cancer. The emergence of cancer is closely linked to an increase in DNA mutations and impaired DNA repair. Human DNA is constantly exposed to damaging substances, which arise either intrinsically due to the physiologic cell metabolism or extrinsically as a result of the impact of environmental factors. In order to
prevent the formation of DNA mutations out of DNA damages and hence cancer formation, functional DNA repair mechanisms are required to guard the integrity of the genome. Despite the positive effects of CR, complying with the restrictions a CR is accompanied with is rather
difficult for most persons. A convenient approach to circumvent the restrictions of a CR partially are so-called calorie restriction mimetics (CRMs), as they could complement or support the positive effects of CR. Two CRMs researched and discussed at the moment are
spermidine and resveratrol. Both CRMs have been shown to have be positive influences on the occurrence of cardiovascular diseases in humans and thus have been proposed to act CR� mimicking. Since the influence of CR on the human organisms is still not understood completely, this work aims at contributing to a better understanding of the impact of CR on human DNA repair and the mechanisms of aging. Furthermore, this thesis deals with the verification of the similar
effects CR and the CRMs spermidine and resveratrol have on the expression of aging� associated genes. For this purpose, the influence of F.X. Mayr therapy on DNA repair capacity of nucleotide excision repair (NER) and the expression of the aging-associated genes SIRT1,
XPA, SIRT3, SOD2, FOXO3, AMPKα, LC3B and BECN1 was analyzed in human peripheral blood mononuclear cells (PBMCs). The modified host cell reactivation assay (mHCRA) revealed a significant increase in DNA repair capacity due to CR. However, only individuals displaying a low DNA repair capacity before CR experienced an increase, whereas no
alteration was obvious for the individuals displaying a normal DNA repair capacity in the beginning of the study. In the end, DNA repair capacity of the “low” group was elevated to the level of the “normal” group. With regard to the expression of aging-associated genes, a
significant increase in SIRT1, AMPKα and BECN1 expression was verified as consequence of CR. Just like DNA repair capacity, the increase in mRNA expression was solely attributed to individuals with a low DNA repair capacity in the beginning. Furthermore, DNA repair capacity as well as mRNA expression were subject to donor variability. The CRMs spermidine and resveratrol were used to treat human PBMCs ex vivo and subsequently mRNA expression of the aging-associated genes SIRT1, SIRT3, SOD2 and FOXO3 was analyzed. Both CRMs significantly induced mRNA expression of SIRT3, SOD2 and FOXO3, whereas only spermidine induced SIRT1 expression significantly. The CRMs seem to particularly have an influence on anti-oxidative properties. Furthermore, permidine increased NER capacity in human fibroblasts. Taken together, CR influences DNA repair capacity of NER positively, thus providing a better protection of DNA damages and their consequences. The increase in mRNA expression of aging-associated genes also hint at an improved DNA repair and autophagy as well as at better defense mechanisms against oxidative stress. Since not every person benefits from CR, CRMs might be a way to boost CR effects, but they surely cannot replace the holistic approach of a CR. In a wider sense the research conducted within the scope of this thesis contributes to a better characterization of the influence of CR and CRMs on human health and
thus to finding ways to a healthier aging. Simultaneously, the results highlight the importance of studies involving humans in order to better characterize influences of CR and CRMs
Federated Knowledge Base Debugging in DL-Lite A
Get PDFDue to the continuously growing amount of data the federation of different and distributed data sources gained increasing attention. In order to tackle the challenge of federating heterogeneous sources a variety of approaches has been proposed. Especially in the context of the Semantic Web the application of Description Logics is one of the preferred methods to model federated knowledge based on a well-defined syntax and semantics. However, the more data are available from
heterogeneous sources, the higher the risk is of inconsistency – a serious obstacle for performing reasoning tasks and query answering over a federated knowledge base. Given a single knowledge base the process of knowledge base debugging comprising the identification and resolution of conflicting statements have been widely studied while the consideration of federated settings integrating a network
of loosely coupled data sources (such as LOD sources) has mostly been neglected.
In this thesis we tackle the challenging problem of debugging federated knowledge bases and focus on a lightweight Description Logic language, called DL-LiteA, that is aimed at applications requiring efficient and scalable reasoning.
After introducing formal foundations such as Description Logics and Semantic Web technologies we clarify the motivating context of this work and discuss the general problem of information integration based on Description Logics. The main part of this thesis is subdivided into three subjects. First, we discuss the specific characteristics of federated knowledge bases and provide an appropriate approach for detecting and explaining contradictive statements in a federated
DL-LiteA knowledge base. Second, we study the representation of the identified conflicts and their relationships as a conflict graph and propose an approach for repair generation based on majority voting and statistical evidences. Third, in order to provide an alternative way for handling inconsistency in federated DL-LiteA knowledge bases we propose an automated approach for assessing adequate trust
values (i.e., probabilities) at different levels of granularity by leveraging probabilistic inference over a graphical model.
In the last part of this thesis, we evaluate the previously developed algorithms against a set of large distributed LOD sources. In the course of discussing the experimental results, it turns out that the proposed approaches are sufficient, efficient and scalable with respect to real-world scenarios. Moreover, due to the exploitation of the federated structure in our algorithms it further becomes apparent that the number of identified wrong statements, the quality of the generated repair as well as the fineness of the assessed trust values profit from an increasing number of integrated sources
Situational States Influence on Team Workload Demands in Cyber Defense Exercice
No full textCyber operations are increasingly automated processes that can occur
at computational speed with the intent of reducing, or denying time for good decision making or time to ground communication between human agents. There is a lack of performance measures and metrics in cyber operation settings. One potential setting describing human performance could be emotional stability under stress. Measures of higher individual affective variability indicate more emotional adaptability and allows for measuring individuals as dynamic systems. Previous
research in other security-sensitive high-stake situations has shown that individuals with less emotional adaptability display maladaptive behaviors while individuals with more emotional adaptability can adapt more efficiently to changing situa� tions, show more confidence in their own abilities and skills, and display better performance. We hypothesized that measurements of affective variability during a
cyber defense exercise will be associated with team workload demands. Data was collected from 13 cadets during the Norwegian Defence Cyber Academy’s annual Cyber Defense Exercise. Three indicators of individual affective variability were measured daily with the Self-Assessment Manikin and compared to scores on the Team Workload Questionnaire. We found that affective variability was negatively
associated with team workload demands. Participants with higher affective vari� ability, as measured by the Self-Assessment Manikin, will impose less workload demands on the team, which can lead to better outcomes. This is the first study to assess how individual emotional adaptability affects team dynamics in a cyber defense setting. Future research should include variable measurements as they
may have better explanatory power for performance measurements
Purely Inertial Navigation with a Low-Cost MEMS Sensor Array
No full textThis paper examines the position precision of purely inertial navigation using an array of redundant, low-cost MEMS sensors. A carefully designed IMU is used to perform navigation experiments and to analyze the benefits of a sensor array over a single sensor in practice. As our experimental results show, navigation can be improved significantly by calibrating the IMU device regarding scale factors, offsets and cross-axis sensitivity. By comparing predicted navigation error and experimental results it is shown that gyroscope angle random walk and bias instability are dominant and therefore can be used to estimate navigation performance. The latter improves roughly by a factor of √14 when using an array of 14 devices instead of a single one.
A Kalman Filter with motion constraints minimizes the error
when estimating positions.
Index Terms—MEMS inertial sensors, low-cost, gyroscope,
accelerometer, navigation, redundancy, array, motion constraints,
IMU (inertial measurement unit), Kalman Filte
