Reference design

Abstract: Robustness is an important requirement for almost all kinds of products. This article shows how evolutionary algorithms can be applied for robust design based on the approach of Taguchi. To achieve a better understanding of the consequences of this approach, we first present some analytical results gained from a toy problem. As a nontrivial industrial application we consider the design of multilayer optical coatings (MOCs) most frequently used for optical filters. An evolutionary algorithm based on a parallel diffusion model and extended for mixed-integer optimization was able to compete with or even outperform traditional methods of robust MOC design. With respect to chromaticity, the MOC designs found by the evolutionary algorithm are substantially more robust to parameter variations than a reference design and therefore perform much better in the average case. In most cases, however, this advantage has to be paid for by a reduction in the average reflectance. The robust design approach outlined in this paper should be easily adopted to other application domains.

Abstract: Motivation: Despite theoretical arguments that so-called 'loop designs' for two-channel DNA microarray experiments are more efficient, biologists continue to use 'reference designs'. We describe two sets of microarray experiments with RNA from two different biological systems (TPA-stimulated mammalian cells and Streptomyces coelicolor). In each case, both a loop and a reference design were used with the same RNA preparations with the aim of studying their relative efficiency. Results: The results of these experiments show that (1) the loop design attains a much higher precision than the reference design, (2) multiplicative spot effects are a large source of variability, and if they are not accounted for in the mathematical model, for example, by taking log-ratios or including spot effects, then the model will perform poorly. The first result is reinforced by a simulation study. Practical recommendations are given on how simple loop designs can be extended to more realistic experimental designs and how standard statistical methods allow the experimentalist to use and interpret the results from loop designs in practice. Availability: The data and R code are available at http://exgen.ma.umist.ac.uk Contact: veronica.vinciotti@brunel.ac.uk

Abstract: In solid particle separation process, cyclone separators are commonly used due to their simplicity and low-cost manufacture. A reference design S1 and two novel designs C4, N4 of four-inlet cyclone separator have been investigated computationally using the Reynolds Stress Model for turbulent flow and the Discrete Phase Model for tracking the particles. The numerical model has been validated and confirmed to be a reasonable good candidate for the investigation. The results show that, the design S1 provides the lowest average Euler number (4.415) and the design C4 provides the smallest average cut-off diameter (1.399 μ m ). As comparing the performances using weighted sum method, both two novel designs provide the better performance than the reference design S1. The innovative design N4 delivers the best performance which is 1 % and 0.1 % better in performance than S1 and C4, respectively.