Dynamic models of piled foundations

The vibration behavior of piled foundations is an important consideration in fields such as earthquake engineering, construction, machine-foundation design, offshore structures, nuclear energy, and road and rail development. This paper presents a review of the past 40 years' literature on modeling the frequency-dependent behavior of pile foundations. Beginning with the earliest model of a single pile, adapted from those for embedded footings, it charts the development of the four pile-modeling techniques: the "dynamic Winkler-foundation" approach that uses springs to represent the effect of the soil; elasticcontinuum-type formulations involving the analytical solutions for displacements due to a subsurface disk, cylinder, or other element; boundary element methods; and dynamic finite-element formulations with special nonreflecting boundaries. The modeling of pile groups involves accounting for pile-soil-pile interactions, and four such methods exist: interaction factors; complete pile models; the equivalent pier method; and periodic structure theory. Approaches for validating pile models are also explored. Copyright © 2013 by ASME.

Comparisons of operating envelopes for contaminated soil stabilised/solidified with different cementitious binders

This work initiated the development of operating envelopes for stabilised/solidified contaminated soils. The operating envelopes define the range of operating variables for acceptable performance of the treated soils. The study employed a soil spiked with 3,000 mg/kg each of Cd, Cu, Pb, Ni and Zn, and 10,000 mg/kg of diesel. The binders used for treatment involved Portland cement (CEMI), pulverised fuel ash (PFA), ground granulated blast furnace slag (GGBS) and hydrated lime (hlime). The specific binder formulations were CEMI, CEMI/PFA = 1:4, CEMI/GGBS = 1:9 and hlime/GGBS = 1:4. The water contents employed ranged from 13 % to 21 % (dry weight), while binder dosages ranged from 5 % to 20 % (w/w). We monitored the stabilised/solidified soils for up to 84 days using different performance tests. The tests include unconfined compressive strength (UCS), hydraulic conductivity, acid neutralisation capacity (ANC) and pH-dependent leachability of contaminants. The water content range resulted in adequate workability of the mixes but had no significant effect on leachability of contaminants. We produced design charts, representing operating envelopes, from the results generated. The charts establish relationships between water content, binder dosage and UCS; and binder dosage, leachant pH and leachability of contaminants. The work also highlights the strengths and weaknesses of the different binder formulations. © 2013 Springer-Verlag Berlin Heidelberg.

Noise and vibration in brushless doubly fed machine and brushless doubly fed reluctance machine

This study investigates modes of vibration in brushless doubly fed machine and brushless doubly fed reluctance machine due to the interaction of its fundamental magnetic fields, via the bending forces they set up in the back iron. It is shown that the presence of two field components of different pole numbers leads to vibration components in addition to those that would be expected in single field machines such as the induction motor. Formulations for the frequencies and magnitudes of the expected vibration components are given and verified experimentally. It is shown that the strength of the vibration components is highly dependent on the choice of pole numbers in the machine, with some much worse than equivalent induction machines and some very similar. The methodology presented enables designers to determine whether their machines are likely to suffer from vibration problems in advance of construction, and to apply remedies where appropriate. © The Institution of Engineering and Technology 2014.

Effects of buffer layer properties and annealing process on bulk heterojunction morphology and organic solar cell performance

The performance of polymer-fullerene bulk heterojunction (BHJ) solar cells is strongly dependent on the vertical distribution of the donor and acceptor regions within the BHJ layer. In this work, we investigate in detail the effect of the hole transport layer (HTL) physical properties and the thermal annealing on the BHJ morphology and the solar cell performance. For this purpose, we have prepared solar cells with four distinct formulations of poly(3,4- ethylenedioxythiophene) poly(styrenesulfonate) (PEDOT:PSS) buffer layers. The samples were subjected to thermal annealing, applied either before (pre-annealing) or after (post-annealing) the cathode metal deposition. The effect of the HTL and the annealing process on the BHJ ingredient distribution - namely, poly(3-hexylthiophene) (P3HT) and [6,6]-phenyl C61 butyric acid methyl ester (PCBM) - has been studied by spectroscopic ellipsometry and atomic force microscopy. The results revealed P3HT segregation at the top region of the films, which had a detrimental effect on all pre-annealed devices, whereas PCBM was found to accumulate at the bottom interface. This demixing process depends on the PEDOT:PSS surface energy; the more hydrophilic the surface the more profound is the vertical phase separation within the BHJ. At the same time those samples suffer from high recombination losses as evident from the analysis of the J-V measurements obtained in the dark. Our results underline the significant effect of the HTL-active and active-ETL (electron transport layer) interfacial composition that should be taken into account during the optimization of all polymer-fullerene solar cells. © 2012 The Royal Society of Chemistry.

Gaussian processes for POMDP-based dialogue manager optimization

A partially observable Markov decision process (POMDP) has been proposed as a dialog model that enables automatic optimization of the dialog policy and provides robustness to speech understanding errors. Various approximations allow such a model to be used for building real-world dialog systems. However, they require a large number of dialogs to train the dialog policy and hence they typically rely on the availability of a user simulator. They also require significant designer effort to hand-craft the policy representation. We investigate the use of Gaussian processes (GPs) in policy modeling to overcome these problems. We show that GP policy optimization can be implemented for a real world POMDP dialog manager, and in particular: 1) we examine different formulations of a GP policy to minimize variability in the learning process; 2) we find that the use of GP increases the learning rate by an order of magnitude thereby allowing learning by direct interaction with human users; and 3) we demonstrate that designer effort can be substantially reduced by basing the policy directly on the full belief space thereby avoiding ad hoc feature space modeling. Overall, the GP approach represents an important step forward towards fully automatic dialog policy optimization in real world systems. © 2013 IEEE.