Support vector regression for determining the minimum zone sphericity

Several methods have been investigated to determine the deviation of manufactured spherical parts from ideal geometry. One of the most popular is the least squares technique, which is still widely employed in coordinate measuring machines used by industries. The least squares algorithm is optimal under the assumption that the data set is very large and has the inherent disadvantage of overestimating the minimum tolerance zone, resulting sometimes in the rejection of good parts. In addition, it requires that the data be distributed normally. The support vector regression approach alleviates the necessity for these assumptions. While most fitting algorithms in practice today require that the sampled data accurately represent the surface being inspected, support vector regression provides a generalization over the surface. We describe how the concepts of support vector regression can be applied to the determination of tolerance zones of nonlinear surfaces; to demonstrate the unique potential of support vector machine algorithms in the area of coordinate metrology. In specific, we address part quality inspection of spherical geometries.     

Plug-load densities for energy analysis: K-12 schools

Benchmarking plug-load densities is essential to bypass arbitrary and/or incorrect inputs used in building energy analysis. As more building simulationists play a decision-making role for the design team, they tend to lean on building energy standards and guidelines for preliminary inputs such as plug-load densities. It is necessary for building energy standards and rating systems to implement plug-load density benchmarks to reward design teams in their efforts to reduce plug-load energy use. Using case study buildings, this paper establishes benchmark plug-load densities for K-12 schools under two new categories – classrooms with computers and classrooms without computers. Eighteen K-12 schools including 9 elementary, 2 middle, and 7 high schools are assessed for actual plug-load densities. For the same case study buildings, four existing approaches – NREL, COMNET, ASHRAE 90.1-1989, and Title-24 are evaluated for plug-load densities. Results show under- and over-estimation of plug-load densities over actual densities. The development of benchmark for K-12 schools will pave way for instituting targets for trimming plug-load densities in new and retrofit building projects.     

A computational algorithm for minimizing total variation in imagerestoration

A reliable and efficient computational algorithm for restoring blurred and noisy images is proposed. The restoration process is based on the minimal total variation principle introduced by Rudin et al. For discrete images, the proposed algorithm minimizes a piecewise linear l (1) function (a measure of total variation) subject to a single 2-norm inequality constraint (a measure of data fit). The algorithm starts by finding a feasible point for the inequality constraint using a (partial) conjugate gradient method. This corresponds to a deblurring process. Noise and other artifacts are removed by a subsequent total variation minimization process. The use of the linear l(1) objective function for the total variation measurement leads to a simpler computational algorithm. Both the steepest descent and an affine scaling Newton method are considered to solve this constrained piecewise linear l(1) minimization problem. The resulting algorithm, when viewed as an image restoration and enhancement process, has the feature that it can be used in an adaptive/interactive manner in situations when knowledge of the noise variance is either unavailable or unreliable. Numerical examples are presented to demonstrate the effectiveness of the proposed iterative image restoration and enhancement process.     

Learning networks for tornado detection

In this paper, different types of learning networks, such as artificial neural networks (ANNs), Bayesian neural networks (BNNs), support vector machines (SVMs) and minimax probability machines (MPMs) are applied for tornado detection. The last two approaches utilize kernel methods to address non-linearity of the data in the input space. All methods are applied to detect when tornadoes occur, using variables based on radar derived velocity data and month number. Computational results indicate that BNNs are more accurate for tornado detection over a suite of forecast evaluation indices.     

A modified sorting task to investigate consumer perceptions of extra virgin olive oils

A two-stage sorting task was used to probe Californian consumers’ opinions of twenty-five extra virgin olive oils based on visual assessments of the bottles. The modification of the simple sorting task aimed to encourage consumers to further discriminate products through sub-groupings of products within the groups they had already made. Sorting data were analyzed using a 3-way extension of classical multidimensional scaling called DISTATIS which explores the level of consumer agreement as well as the structure of the extra virgin olive oil products. Consumer agreements were explained by 46% of the variances found in the ways consumers sorted the bottles. Decreased amount of explained variance from the first stage to the second stage of the sorting task was observed in the olive oil product structures. Consumers were asked to describe the characteristics defining each group formed. Consumer comments were analyzed qualitatively prior to statistical analysis and were later used to understand the sorting results. Despite background differences in the usage of olive oil products, the majority of the consumers perceived the product set similarly. The two-stage sorting task allowed subjects to provide more criteria or multidimensional views of their perception of the products.     

Nondestructive evaluation of corrosion damage using electrostatic measurements

We investigate the problem of detecting and assessing, by means of static electrical measurements, damage due to corrosion in a structure. Corrosion damage, which is assumed to occur in an inaccessible part of a specimen, is modelled as material loss. The detection device consists of electrodes which inject DC current and measure voltage potentials in the accessible part of the specimen. The topography of the damaged surface is estimated from the measured data. This research is meant to evaluate if a method based on static electrical measurements has the potential to be developed into a useful nondestructive evaluation tool. We propose computational methods that take the measured data and estimate the unknown damaged surface. The methods are studied in order to understand their properties. Several example calculations from synthetic data are presented. Our findings indicate that such a device has limited resolution. However, it offers several advantages that make it worthwhile to pursue further research.     

Bending collapse of thin-walled beams with ultralight filler: numerical simulation and weight optimization

Summary A study on the bending collapse of thin-walled beams filled with aluminum foam or aluminum honeycomb was carried out. The strengthening effects of ultralight metal fillers were quantified numerically and experimentally. The moment-rotation characteristics of filled sections were derived, and the results were then incorporated with structural optimization technique to develop a methodology for crashworthiness optimization of filled members. The proposed methodology requires relatively simple computations and is suitable for early stage design of crash members. Finally, the optimization problem of filled sections under combined compression/bending loading was formulated and solved. The optimization results showed potentials of significant weight saving and volume reduction by utilizing ultralight metal filler.     

装配式钢框架-减震围护墙单元力学性能分析

为更好发挥减震围护墙对装配式钢框架单元力学性能的有利作用，采用ABAQUS软件建立了13个装配式钢框架-减震围护墙单元有限元模型，研究不同上、下层墙板单元高度比、减震层摩擦系数、墙板材质对框架单元的力学性能的影响及其规律，并给出有关减震围护墙设计的建议。研究结果表明：增加上、下层墙板单元高度比、提高减震层摩擦系数和增大墙板材质密度均可一定程度提高框架的屈服荷载和水平承载力，且不会明显增加框架的刚度。因此，设计时可根据洞口尺寸需求尽量选取较大的上、下层墙板单元高度比，且当开设窗洞时，建议将窗洞设置在上层墙板单元。同时建议选取摩擦系数较大的材料制作减震层，并选用材质密度相对较大的墙板。