Connection and coordination: the interplay between architecture and dynamics in evolved model pattern generators

We undertake a systematic study of the role of neural architecture in shaping the dynamics of evolved model pattern generators for a walking task. First, we consider the minimum number of connections necessary to achieve high performance on this task. Next, we identify architectural motifs associated with high fitness. We then examine how high-fitness architectures differ in their ability to evolve. Finally, we demonstrate the existence of distinct parameter subgroups in some architectures and show that these subgroups are characterized by differences in neuron excitabilities and connection signs.     

The smooth (tractor) operator: insights of knowledge engineering

The design of and training for complex systems requires in-depth understanding of task demands imposed on users. In this project, we used the knowledge engineering approach (Bowles et al., 2004) to assess the task of mowing in a citrus grove. Knowledge engineering is divided into four phases: (1) Establish goals. We defined specific goals based on the stakeholders involved. The main goal was to identify operator demands to support improvement of the system. (2) Create a working model of the system. We reviewed product literature, analyzed the system, and conducted expert interviews. (3) Extract knowledge. We interviewed tractor operators to understand their knowledge base. (4) Structure knowledge. We analyzed and organized operator knowledge to inform project goals. We categorized the information and developed diagrams to display the knowledge effectively. This project illustrates the benefits of knowledge engineering as a qualitative research method to inform technology design and training.     

Mapped infinite elements for three-dimensional multi-region boundary element analysis

A formulation for an infinite boundary element (BE) is presented, which allows the modelling of infinite surfaces. The concept is that the finite surface is mapped to an infinite surface using special mapping functions. Using such mapping functions together with linear and quadratic interpolation for the displacements and the tractions, respectively, the desired decay behaviour can be modelled. The implementation of the proposed infinite elements becomes straightforward, since the Cauchy principal value, as well as the free term, are evaluated for the finite and infinite BEs with exactly the same techniques. The element developed can be used in a multi-region BE analysis of piecewise homogeneous domains, or for domains with joints and faults. The accuracy of the element is tested on some benchmark problems. Finally, a practical application in tunnelling is shown. Copyright © 2004 John Wiley & Sons, Ltd.     

Australia's housing choices: Retrospect and prospect

This article identifies issues around the Auckland Region's strategy for urban intensification and discusses the ensuing policy implications. Medium density housing forms an important element of the strategy for growth management. Implications for its success are primarily drawn from research into a major housing development in Waitakere City. It is argued that for ideological, political and technical reasons the strategy could be easily undermined. The article also suggests that something of a paradigm shift is underway in urban planning.     

Convective condensation in a stator–rotor–stator spinning disc reactor

Centrifugal intensification of condensation heat transfer in the rotor–stator cavities of a stator–rotor–stator spinning disc reactor (srs‐SDR) is studied, as a function of rotational velocity ω, volumetric throughflow rate , and average temperature driving force . For the current range of ω, heat transfer from the vapor bubbles to the condensate liquid is limiting, due to a relatively low gas–liquid interfacial area a_GL. For rad s^?1, a strong increase of a_GL, results in increasing the reactor‐average condensation heat transfer coefficient h_c from 1600 to 5600 W m^?2 K^?1, for condensation of pure dichloromethane vapor. Condensation heat transfer in the srs‐SDR is enhanced by rotation, independent of the vapor velocity. The intensified condensation comes at the cost of relatively high energy dissipation rates, indicating condensation in the srs‐SDR is more suited as a means to supply heat (e.g. in an intensified reactor‐heat exchanger), rather than for bulk cooling purposes. © 2016 American Institute of Chemical Engineers AIChE J, 62: 3784–3796, 2016     

Transformation of Free and Dipeptide‐Bound Glycated Amino Acids by Two Strains of Saccharomyces cerevisiae

The yeast Saccharomyces cerevisiae transforms branched‐chain and aromatic amino acids into higher alcohols in the Ehrlich pathway. During microbiological culturing and industrial fermentations, this yeast is confronted with amino acids modified by reducing sugars in the Maillard reaction (glycation). In order to gain some preliminary insight into the physiological “handling” of glycated amino acids by yeasts, individual Maillard reaction products (MRPs: fructosyllysine, carboxymethyllysine, pyrraline, formyline, maltosine, methylglyoxal‐derived hydroimidazolone) were administered to two strains of S. cerevisiae in a rich medium. Only formyline was converted into the corresponding α‐hydroxy acid, to a small extent (10 %). Dipeptide‐bound pyrraline and maltosine were removed from the medium with concomitant emergence of several metabolites. Pyrraline was mainly converted into the corresponding Ehrlich alcohol (20–60 %) and maltosine into the corresponding α‐hydroxy acid (40–60 %). Five specific metabolites of glycated amino acids were synthesized and characterized. We show for the first time that S. cerevisiae can use glycated amino acids as a nitrogen source and transform them into new metabolites, provided that the substances can be transported across the cell membrane.     

Understanding challenges in the front lines of home health care: A human-systems approach

A human-systems perspective is a fruitful approach to understanding home health care because it emphasizes major individual components of the system – persons, equipment/technology, tasks, and environments – as well as the interaction between these components. The goal of this research was to apply a human-system perspective to consider the capabilities and limitations of the persons, in relation to the demands of the tasks and equipment/technology in home health care. Identification of challenges and mismatches between the person(s) capabilities and the demands of providing care provide guidance for human factors interventions. A qualitative study was conducted with 8 home health Certified Nursing Assistants and 8 home health Registered Nurses interviewed about challenges they encounter in their jobs. A systematic categorization of the challenges the care providers reported was conducted and human factors recommendations were proposed in response, to improve home health. The challenges inform a human-systems model of home health care.     

Spray formation of middle distillates for autothermal reforming

The reforming of diesel and diesel-like fuels plays a central role in the development of fuel cell systems for on-board power supplies. The vaporization of the fuel via a spray formation and the subsequent mixture with water vapor and air determine the quality of the reforming process, as is shown in this paper. By using a high quality nozzle residual hydrocarbons were below 25 ppmV during the reforming of standard diesel. Through the use of a fuel injector in pulsed operation, the load range was able to be increased from 1:1.67 to 1:6. Spray pattern analyses were conducted using a high-speed camera. The formation of the spray pattern lasted 1.5–2 ms. The testing of a fast-closing magnetic valve manufactured by GSR Ventiltechnik was carried out on the autothermal reformer (ATR) type AH2. It exist not any direct influence of the pulsed operation on hydrogen production.     

The virtual production simulation platform: from collaborative distributed simulation to integrated visual analysis

Computational simulations are used for the optimization of production processes in order to significantly reduce the need for costly experimental optimization approaches. Yet individual simulations can rarely describe more than a single production step. A set of simulations has to be coupled to each other to form a contiguous representation of a process chain. Eventually, simulated results have to be analyzed by domain experts to gather insight from the preformed computations. In this paper, we propose an IT infrastructure and software tools that aim at a rather non-intrusive way of coupling resources and domain expert’s knowledge to enable the collaborative setup, execution and analysis of distributed simulation chains. We illustrate the approach in the domain of materials processing. Beyond means originating from the domain of GRID computing for resource management, a data integration component assures semantic data integrity between the simulation steps and stores simulation data in an application independent way. Thus, we can transform this data into native formats for each simulation tool, and finally into a format that allows for contiguous visualizations and an intuitive, comprehensive analysis of complete simulated process chains.