ALBERT

All Library Books, journals and Electronic Records Telegrafenberg

feed icon rss

Your email was sent successfully. Check your inbox.

An error occurred while sending the email. Please try again.

Proceed reservation?

Export
  • 1
    Electronic Resource
    Electronic Resource
    Oxford, UK : Blackwell Publishing
    Expert systems 21 (2004), S. 0 
    ISSN: 1468-0394
    Source: Blackwell Publishing Journal Backfiles 1879-2005
    Topics: Computer Science
    Notes: Abstract: A key problem of modular neural networks is finding the optimal aggregation of the different subtasks (or modules) of the problem at hand. Functional networks provide a partial solution to this problem, since the inter-module topology is obtained from domain knowledge (functional relationships and symmetries). However, the learning process may be too restrictive in some situations, since the resulting modules (functional units) are assumed to be linear combinations of selected families of functions. In this paper, we present a non-parametric learning approach for functional networks using feedforward neural networks for approximating the functional modules of the resulting architecture; we also introduce a genetic algorithm for finding the optimal intra-module topology (the appropriate balance of neurons for the different modules according to the complexity of their respective tasks). Some benchmark examples from nonlinear time-series prediction are used to illustrate the performance of the algorithm for finding optimal modular network architectures for specific problems.
    Type of Medium: Electronic Resource
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
  • 2
    Publication Date: 2019
    Description: To ensure the formation of a sound shell-core bond interface free of defects between the shell and the core in work rolls used in the finishing stands of hot strip mills, a complete fusion of this interface must be achieved, avoiding excessive mixing of the two components and the formation of hard, fragile microstructures. The shell is made of white cast iron, alloyed with Ni and Cr, and the core is manufactured of grey cast iron spheroidal graphite in a pearlitic matrix. It is thus advisable to inoculate the shell with 0.6 kg/T SiCaMn, as this promotes discontinuity in the carbide network and leads to an increase in the impact toughness of the bond interface. Furthermore, inoculation of the shell with FeSi-La should be avoided, as this inoculant leads to an increase in graphite counts, promoting it with a lamellar morphology at the edge of the bond and hence reducing the impact toughness in this interface. Addition of Mg to the shell has been found to produce an increase in hardness in the regions adjacent to the bond interface.
    Electronic ISSN: 1996-1944
    Topics: Mechanical Engineering, Materials Science, Production Engineering, Mining and Metallurgy, Traffic Engineering, Precision Mechanics
    Published by MDPI
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
  • 3
    Publication Date: 2019
    Description: This paper presents the results obtained regarding the control by manufacturers of the percentage, shape, and size of the precipitated graphite in the working layer of duplex work-rolls used in hot strip mill finishing stands. This working layer is manufactured in a martensitic white cast iron alloyed with Cr and Nb to promote the precipitation of M3C and MC carbides, which provide a high wear resistance. The thermal cycling behavior of this layer also has a decisive influence on its service life. In this context, the percentage of graphite and its morphology play a very important role against said thermal cycling. With the aim of studying their effect on the sphericity of graphite, the analyzed industrial manufacturing factors worth highlighting include the liquidus temperature, the %Si, the use of an FeSi inoculant with traces of Lanthanum, inoculation with different amounts of FeB and SiCaMn, and the addition of Mg. At the periphery of the working layer, it was found that the use of the FeSi inoculant with traces of La promoted an increase in the density of counts of graphite, and that inoculation with FeB and the addition of 0.02% Mg diminished the nodularity of the graphite. Furthermore, throughout the entire thickness of the working layer, an increase in the amount of SiCaMn of up to 0.6 kg/T produced an increase in the size of the graphite particles and a marked improvement in their nodularity.
    Electronic ISSN: 1996-1944
    Topics: Mechanical Engineering, Materials Science, Production Engineering, Mining and Metallurgy, Traffic Engineering, Precision Mechanics
    Published by MDPI
    Location Call Number Expected Availability
    BibTip Others were also interested in ...
Close ⊗
This website uses cookies and the analysis tool Matomo. More information can be found here...