Effect of rotation on the structure of a convecting mushy layer

Summary Nonlinear natural convection in a mushy layer during solidification of a binary alloy is investigated under a high gravity environment where the rotation axis is inclined to the high gravity vector. Asymptotic and scaling analyses are applied to convective flow within the mushy layer and in vertical chimneys. The main result is that, for some particular moderate rotation range, vertical velocity in the chimneys decreases rapidly with increasing rotation rate and appears to have opposite signs across some rotation dependent vertical level.     

Empirical Test Observations in Client-Server Systems

Formal statistical analysis of defect databases for two commercial client-server products for disparate applications - C-language parsing and geophysical modeling - tests the hypothesis that defects in the graphical client mask more important underlying defects in the computational server. It also quantifies the benefit of continued testing after delivery in terms of reduced software defect densities apparent to end users.     

Classifier Ensembles with a Random Linear Oracle

We propose a combined fusion-selection approach to classifier ensemble design. Each classifier in the ensemble is replaced by a miniensemble of a pair of subclassifiers with a random linear oracle to choose between the two. It is argued that this approach encourages extra diversity in the ensemble while allowing for high accuracy of the individual ensemble members. Experiments were carried out with 35 data sets from UCI and 11 ensemble models. Each ensemble model was examined with and without the oracle. The results showed that all ensemble methods benefited from the new approach, most markedly so random subspace and bagging. A further experiment with seven real medical data sets demonstrates the validity of these findings outside the UCI data collection     

Rival-Model Penalized Self-Organizing Map

As a typical data visualization technique, self-organizing map (SOM) has been extensively applied to data clustering, image analysis, dimension reduction, and so forth. In a conventional adaptive SOM, it needs to choose an appropriate learning rate whose value is monotonically reduced over time to ensure the convergence of the map, meanwhile being kept large enough so that the map is able to gradually learn the data topology. Otherwise, the SOM's performance may seriously deteriorate. In general, it is nontrivial to choose an appropriate monotonically decreasing function for such a learning rate. In this letter, we therefore propose a novel rival-model penalized self-organizing map (RPSOM) learning algorithm that, for each input, adaptively chooses several rivals of the best-matching unit (BMU) and penalizes their associated models, i.e., those parametric real vectors with the same dimension as the input vectors, a little far away from the input. Compared to the existing methods, this RPSOM utilizes a constant learning rate to circumvent the awkward selection of a monotonically decreased function for the learning rate, but still reaches a robust result. The numerical experiments have shown the efficacy of our algorithm     

On the Evolutionary Optimization of Many Conflicting Objectives

This study explores the utility of multiobjective evolutionary algorithms (using standard Pareto ranking and diversity-promoting selection mechanisms) for solving optimization tasks with many conflicting objectives. Optimizer behavior is assessed for a grid of mutation and recombination operator configurations. Performance maps are obtained for the dual aims of proximity to, and distribution across, the optimal tradeoff surface. Performance sweet-spots for both variation operators are observed to contract as the number of objectives is increased. Classical settings for recombination are shown to be suitable for small numbers of objectives but correspond to very poor performance for higher numbers of objectives, even when large population sizes are used. Explanations for this behavior are offered via the concepts of dominance resistance and active diversity promotion.     

Using transistors to linearise biochemistry

A fundamental relationship exists between diffusion characteristics within semiconductors and Nernstian equilibrium in biological systems. In a transistor operating in weak inversion the potential difference between terminals governs electron concentration in an exponential way according to the Boltzmann distribution of charged particles while in a biochemical cell the potential difference across a membrane is governed by ionic concentration in a logarithmic way according to the Nernst equation. These two nonlinear physical phenomena form an interaction that potentially leads to linearisation and subsequent modelling of or interaction with biological systems by integrated semiconductor devices. To demonstrate the authors' hypothesis a silicon transistor-based biosensor is considered. This natural bridge between biochemistry and semiconductor silicon chips will enable the potential mass production of portable biochemical devices for the consumer market     

你是家庭和朋友的IT人吗？

当今,支持的费用是IT部门最大的开支。我们都在想方设法计量和减少这种开销。但是有一种IT支持的费用没有列在预算或收支平衡表上,那就是我们花费在支持朋友和家庭成员解决IT问题上的时间。我们将它称之为＂个人IT＂。     

Impact of Preferential P-Diffusion Along the Grain Boundaries on Fine-Grained Polysilicon Solar Cells

Thin-film polysilicon solar cells are a promising low-cost alternative for bulk silicon solar cells due to their reduced material thickness. Recently, we showed that the use of an amorphous silicon/polycrystalline silicon heterojunction emitter instead of a diffused homojunction emitter led to a boost in the open-circuit voltage by 90 mV. Now, we present a full evidence that shows that this improvement is related to the absence of dopant smearing along the grain boundaries. By using scanning spreading resistance microscopy, we found an enlargement of the junction area by a factor of five in case of a homojunction. The tips of the dopant spikes represent lowly doped areas with an enhanced recombination.