05 Fakultät Informatik, Elektrotechnik und Informationstechnik

Permanent URI for this collectionhttps://elib.uni-stuttgart.de/handle/11682/6

Browse

Filters

2017 - 20204

Settings

Subject: search.filters.subject.670Subject: search.filters.subject.620Subject: search.filters.subject.004

Search Results

Now showing 1 - 4 of 4
  • Thumbnail Image
    ItemOpen Access
    Maschinelles Lernen für intelligente Automatisierungssysteme mit dezentraler Datenhaltung am Anwendungsfall Predictive Maintenance
    (2019) Maschler, Benjamin; Jazdi, Nasser; Weyrich, Michael
    Für eine hohe Ergebnisqualität sind Machine Learning Algorithmen auf eine breite Datenbasis angewiesen. Studien zeigen jedoch, dass viele Unternehmen nicht bereit sind, ihre Daten mit anderen Unternehmen, beispielsweise in Form einer gemeinsamen Daten-Cloud, zu teilen. Ziel sollte es daher sein, effizientes maschinelles Lernen mit einer dezentralen Datenhaltung, die den Verbleib vertraulicher Daten im jeweiligen Ursprungs-Unternehmen ermöglicht, zu ermöglichen. In diesem Artikel wird diesbezüglich ein neuartiges Konzept vorgestellt und hinsichtlich seiner Potentiale für intelligente Automatisierungssysteme am Beispiel des Anwendungsfalls Predictive Maintenance analysiert. Die Umsetzbarkeit des Konzepts unter Nutzung verschiedener bestehender Ansätze wird diskutiert, bevor schließlich auf potentielle Mehrwerte für Anlagenbetreiber sowie -hersteller unter besonderer Berücksichtigung der Perspektive kleiner und mittlerer Unternehmen eingegangen wird.
  • Thumbnail Image
    ItemOpen Access
    Deep learning based soft sensors for industrial machinery
    (2020) Maschler, Benjamin; Ganssloser, Sören; Hablizel, Andreas; Weyrich, Michael
    A multitude of high quality, high-resolution data is a cornerstone of the digital services associated with Industry 4.0. However, a great fraction of industrial machinery in use today features only a bare minimum of sensors and retrofitting new ones is expensive if possible at all. Instead, already existing sensors’ data streams could be utilized to virtually ‘measure’ new parameters. In this paper, a deep learning based virtual sensor for estimating a combustion parameter on a large gas engine using only the rotational speed as input is developed and evaluated. The evaluation focusses on the influence of data preprocessing compared to network type and structure regarding the estimation quality.
  • Thumbnail Image
    ItemOpen Access
    3D printing-as-a-service for collaborative engineering
    (2017) Baumann, Felix W.; Roller, Dieter (Univ.-Prof. Hon.-Prof. Dr.)
    3D printing or Additive Manufacturing (AM) are utilised as umbrella terms to denote a variety of technologies to manufacture or create a physical object based on a digital model. Commonly, these technologies create the objects by adding, fusing or melting a raw material in a layer-wise fashion. Apart from the 3D printer itself, no specialised tools are required to create almost any shape or form imaginable and designable. The possibilities of these technologies of these technologies are plentiful and cover the ability to manufacture every object, rapidly, locally and cost-efficiently without wasted resources and material. Objects can be created to specific forms to perform as perfectly fitting functions without consideration of the assembly process. To further the advance the availability and applicability of 3D printing, this thesis identifies the problems that currently exist and attempts to solve them. During the 3D printing process, data (i. e., files) must be converted from their original representation, e. g., CAD file, to the machine instructions for a specific 3D printer. During this process, information is lost, and other information is added. Traceability is lacking in 3D printing. The actual 3D printing can require a long period of time to complete, during which errors can occur. In 3D printing, these errors are often non-recoverable or reversible, which results in wasted material and time. In addition to the lack of closed-loop control systems for 3D printers, careful planning and preparation are required to avoid these costly misprints. 3D printers are usually located remotely from users, due to health and safety considerations, special placement requirements or out of comfort. Remotely placed equipment is impractical to monitor in person; however, such monitoring is essential. Especially considering the proneness of 3D printing to errors and the implications of this as described previously. Utilisation of 3D printers is an issue, especially with expensive 3D printers. As there are a number of differing 3D printing technologies available, having the required 3D printer, might be problematic. 3D printers are equipped with a variety of interfaces, depending on the make and model. These differing interfaces, both hard- and software, hinder the integration of different 3D printers into consistent systems. There exists no proper and complete ontology or resource description schema or mechanism that covers all the different 3D printing technologies. Such a resource description mechanism is essential for the automated scheduling in services or systems. In 3D printing services the selection and matching of appropriate and suitable 3D printers is essential, as not all 3D printing technologies are able to perform on all materials or are able to create certain object features, such as thin walls or hollow forms. The need for companies to sell digital models for AM will increase in scenarios where replacement or customised parts are 3D printed by consumers at home or in local manufacturing centres. Furthermore, requirements to safeguard these digital models will increase to avoid a repetition of the problems from the music industry, e. g., Napster. Replication and ‘theft’ of these models are uncontrollable in the current situation. In a service oriented deployment, or in scenarios where the utilisation is high, estimations of the 3D printing time are required to be available. Common 3D printing time estimations are inaccurate, which hinder the application of scheduling. The complete and comprehensive understanding of the complexity of an object is discordant, especially in the domain of AM. This understanding is required to both support the design of objects for AM and match appropriate manufacturing resources to certain objects. Quality in AM and FDM have been incompletely researched. The quality in general is increased with maturity of the technology; however, research on the quality achievable with consumer-grade 3D printers is lacking. Furthermore, cost-sensitive measurement methods for quality assessment are expandable. This thesis presents the structured design and implementation of a 3D printing service with associated contributions that provide solutions to particular problems present in the AM domain. The 3D printing service is the overarching component of this thesis and provides the platform for the other contributions with the intention to establish an online, cloud-based 3D printing service for use in end-user and professional settings with a focus on collaboration and cooperation.
  • Thumbnail Image
    ItemOpen Access
    User-friendly, requirement-based assistance for production workforce using an asset administration shell design
    (2020) Al Assadi, Anwar; Fries, Christian; Fechter, Manuel; Maschler, Benjamin; Ewert, Daniel; Schnauffer, Hans-Georg; Zürn, Michael; Reichenbach, Matthias
    Future production methods like cyber physical production systems (CPPS), flexibly linked assembly structures and the matrix production are characterized by highly flexible and reconfigurable cyber physical work cells. This leads to frequent job changes and shifting work environments. The resulting complexity within production increases the risk of process failures and therefore requires longer job qualification times for workers, challenging the overall efficiency of production. During operation, cyber physical work cells generate data, which are specific to the individual process and worker. Based on the asset administration shell for Industry 4.0, this paper develops an administration shell for the production workforce, which contains personal data (e.g. qualification level, language skills, machine access, preferred display and interaction settings). Using worker and process specific data as well as personal data, allows supporting, training and instating workers according to their individual capabilities. This matching of machine requirements and worker skills serves to optimize the allocation of workers to workstations regarding the ergonomic workplace setup and the machine efficiency. This paper concludes with a user-friendly, intuitive design approach for a personalized machine user interface. The presented use-cases are developed and tested at the ARENA2036 (Active Research Environment for the Next Generation of Automobiles) research campus.