Iskanje po repozitoriju

A+ | A- | | SLO | ENG

Opis: Modelling Dynamic Systems with Artificial Neural Networks and Related Methods can be used as a textbook for the field it covers or as an introductory textbook for more advanced literature in the field. The textbook comprises introduction to artificial neural networks, identification of linear and nonlinear dynamic systems, control with artificial neural networks, local-model networks and blended multiple-model systems, design of gain-scheduling control and identification of nonlinear systems with Gaussian processes. It is intended for undergraduate students, especially at the postgraduate level, who have sufficient knowledge in dynamic systems, as well as for professionals who wish to familiarise themselves with the concepts and views described. The textbook is not intended to be a detailed theoretically based description of the subject, but rather an overview of the field of identification of dynamic systems with neural networks and related methods from the perspective of systems theory and, in particular, its application. The work is intended to inform the reader about the views on this subject, which are related but treated very differently.
Ključne besede: umetne nevronske mreže, dinamični sistemi, nevronske mreže
Objavljeno: 03.03.2023; Ogledov: 136; Prenosov: 10
Polno besedilo (0,00 KB)
Gradivo ima več datotek! Več...

Ključne besede: titanium dioxide nanotubes, hematite, oxygen vacancies, photoelectrochemical degradation of dye, ibuprofen, chemical transformation, glycerol, vanillyl alcohol, dissertations
Objavljeno: 01.03.2023; Ogledov: 106; Prenosov: 0
Polno besedilo (5,71 MB)

Objavljeno: 28.02.2023; Ogledov: 116; Prenosov: 0
Polno besedilo (11,97 MB)

Ključne besede: nanobodies, adhirons, phage display technology
Objavljeno: 27.02.2023; Ogledov: 107; Prenosov: 0
Polno besedilo (575,21 KB)

Opis: Strong electric fields exist between the electric double layer and charged surfaces. These fields impact molecular structures and chemistry at interfaces. We have developed a transparent electrode with infrared plasmonic enhancement sufficient to measure FTIR and two-dimensional infrared spectra at submonolayer coverages on the surface to which a voltage can be applied. Our device consists of an infrared transparent substrate, a 10–20 nm layer of conductive indium tin oxide (ITO), an electrically resistive layer of 3–5 nm Al2O3, and a 3 nm layer of nonconductive plasmonic gold. The materials and thicknesses are set to maximize the surface number density of the monolayer molecules, electrical conductivity, and plasmonic enhancement while minimizing background signals and avoiding Fano line shape distortions. The design was optimized by iteratively characterizing the material roughness and thickness with atomic force microscopy and electron microscopy and by monitoring the plasmon resonance enhancement with spectroscopy. The design is robust to repeated fabrication. This new electrode is tested on nitrile functional groups using a monolayer of 4-mercaptobenzonitrile as well as on CO and CC stretching modes using 4-mercaptobenzoic acid methyl ester. A voltage-dependent Stark shift is observed on both monolayers. We also observe that the transition dipole strength of the CN mode scales linearly with the applied voltage, providing a second way of measuring the surface electric field strength. We anticipate that this cell will enable many new voltage-dependent infrared experiments under applied voltages.
Ključne besede: two-dimensional infrared spectroscopy, infrared transparent substrate, voltage-dependent infrared experiments
Objavljeno: 24.02.2023; Ogledov: 118; Prenosov: 0
Polno besedilo (6,07 MB)