Multivariate significance testing and model calibration under uncertainty

The importance of modeling and simulation in the scientific community has drawn interest towards methods for assessing the accuracy and uncertainty associated with such models. This paper addresses the validation and calibration of computer simulations using the thermal challenge problem developed at Sandia National Laboratories for illustration. The objectives of the challenge problem are to use hypothetical experimental data to validate a given model, and then to use the model to make predictions in an untested domain. With regards to assessing the accuracy of the given model (validation), we illustrate the use of Hotelling’s T² statistic for multivariate significance testing, with emphasis on the formulation and interpretation of such an analysis for validation assessment. In order to use the model for prediction, we next employ the Bayesian calibration method introduced by Kennedy and O’Hagan. Towards this end, we discuss how inherent variability can be reconciled with “lack-of-knowledge” and other uncertainties, and we illustrate a procedure that allows probability distribution characterization uncertainty to be included in the overall uncertainty analysis of the Bayesian calibration process.     

CFD Study on Compound Impaction in a Jet-in-Well Impactor

In a classical jet impactor, the jet is assumed to impact on an infinite plate; however, the typical construction of an impactor necessitates that a wall surrounds the jet. If the wall is close to the jet, secondary impaction can take place on the wall. Also, for some impactor designs, e.g., large particle scalping devices in air sampling inlets, the jet is purposefully placed in a well and the overall performance of the impactor is the compound effect of the jet impacting on a collection plate and on the wall of the well. In this study we primarily use CFD to examine the performance of compound round jet impactors, where the ratio of well diameter to jet diameter is from 3 to 15 and the particle size is in the range of 5 to 15 μm AD. The onset of the secondary impaction occurs at a ratio of well diameter to jet diameter of about eight with the collection on the side wall increasing for smaller ratios. Compound impaction can significantly enhance the overall particle collection and can result in a much smaller value of Stk _0.5 . At a ratio of well diameter to jet diameter of 3, the value of Stk _0.5 is 0.07, as compared with a classical circular jet impactor where Stk _0.5 is about 0.24 for a typical set of operational and configuration parameters. Numerical predictions for the jet-in-well impactors are compared with experimental results using liquid particles and good agreement is obtained. Photos taken over a range of experimental conditions verify the strong side wall impaction for certain conditions.     

Wind Tunnel Evaluation of a Rotating-Element Large-Particle Sampler

A large-particle sampler was tested in an environmental wind tunnel to characterize the efficiency as a function of particle size and type (solid or liquid). The sampler, which had been developed by another organization, has two slotted-cylinder collection elements which are rotated through air at a tangential speed of 39 m/s. Data from wind tunnel tests show the efficiency of liquid droplet collection to increase with particle size to approximately 50% at 25 μm and then decrease with a further increase in particle size. It is suggested that the air flow patterns created by the sampler produce this anomalous behavior. Results of tests with solid particles show higher efficiencies than are obtained with liquid droplets of the same size. It appears that solid particles, which rebound from collection sites other than a slot, can subsequently enter the slots. Supporting data were obtained on the performance of a single slotted cylinder that was fixed in the discharge region of a small free-jet tunnel. Those results show the impaction efficiency of all sizes of liquid droplets in the slotted cylinder follow the type of trend expected for a circular cylinder, i.e., a curve in which efficiency monotonically increases with size. However, 41-μ m diameter solid particles are re-entrained in the air stream and show reduced efficiency in comparison with liquid aerosol particles.     

A Circumferential Slot In-Line Virtual Impactor

An In-line Virtual Impactor is presented, which has an application as a pre-separator for sampling inlets, where the device scalps large particles from the aerosol size distribution. Numerical simulation was the principal tool employed in the design process, with physical experiments used to verify computational predictions. Performance investigations were primarily carried out for a configuration that provides a nominal cutpoint particle size of 10 μ m aerodynamic diameter at an inlet flow of 111 L/min and a major flow exhaust of 100 L/min; however, the concept is scalable in terms of both flow rates and cutpoint sizes. An inverted dual cone configuration contained within a tube provides a characteristic circumferential slot of width 2.54 mm (0.100 inches) and a slot length of 239 mm (9.42 inches) at the critical zone. The upper cone causes the flow to accelerate to an average throat velocity of 3.15 m/s, while the lower cone directs the major flow toward the exit port and minimizes recirculation zones that could cause flow instabilities in the major flow region. The cutpoint Stokes number is 0.73; however, the cutpoint can be adjusted by changing the geometrical spacing between the acceleration nozzle exit plane and a flow divider. When the system is operated at the major exhaust flow rate of 100 L/min, the pressure drop is 45 Pa. Good agreement is obtained between numerically predicted and experimentally observed performance.     

Optimization Studies on a Slit Virtual Impactor

This study reports the results of numerical investigations performed on a proposed slit virtual impactor configuration. Simulations were performed using the computational fluid dynamics program CFX-4.4. A detailed numerical investigation was carried out to determine the critical geometrical parameters of the proposed configuration that would optimize the performance for the anticipated operation conditions (specified Reynolds number and major-to-minor flow split ratio). A detailed sensitivity analysis was performed on the optimized configuration to determine its performance characteristics (impactor efficiency and wall loss curves) for (a) operation at different major-to-minor flow split ratios for a constant Reynolds number and (b) operation at different Reynolds numbers but at the same major-to-minor flow split ratio. Finally, the effect of modifying the radius of curvature of the receiving section of the optimized configuration on the impactor performance was studied for the anticipated operation conditions, and the results are reported.     

The impact of motivation on temporal comparisons: Coping with traumatic events by perceiving personal growth.

Four studies were conducted to investigate the impact of self-enhancement motivation on the temporal comparisons of victims of stressful life events. Study 1 revealed that (a) victims were more likely than acquaintances of victims to report greater improvement in their personal attributes after traumatic life events than after mild negative life events and (b) victims perceived improvement by derogating their pre-event attributes. In Studies 2 and 3, an experimental approach was used to study the impact of threatening experiences on perceptions of personal growth, and similar findings were obtained. Study 4 confirmed that threatening self-relevant feelings play a causal role in prompting self-enhancing temporal comparisons. Taken together, the findings of these studies support the view that perceptions of personal improvement reflect, at least in part, motivated illusions that are designed to help people cope with threatening life experiences. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Changes in thymic function with age and during the treatment of HIV infection

The thymus represents the major site of the production and generation of T cells expressing alphabeta-type T-cell antigen receptors. Age-related involution may affect the ability of the thymus to reconstitute T cells expressing CD4 cell-surface antigens that are lost during HIV infection; this effect has been seen after chemotherapy and bone-marrow transplantation. Adult HIV-infected patients treated with highly active antiretroviral therapy (HAART) show a progressive increase in their number of naive CD4-positive T cells. These cells could arise through expansion of existing naive T cells in the periphery or through thymic production of new naive T cells. Here we quantify thymic output by measuring the excisional DNA products of TCR-gene rearrangement. We find that, although thymic function declines with age, substantial output is maintained into late adulthood. HIV infection leads to a decrease in thymic function that can be measured in the peripheral blood and lymphoid tissues. In adults treated with HAART, there is a rapid and sustained increase in thymic output in most subjects. These results indicate that the adult thymus can contribute to immune reconstitution following HAART.     

Sensorimotor rhythm-based brain-computer interface (BCI): feature selection by regression improves performance.

People can learn to control electroencephalogram (EEG) features consisting of sensorimotor rhythm amplitudes and can use this control to move a cursor in one or two dimensions to a target on a screen. In the standard one-dimensional application, the cursor moves horizontally from left to right at a fixed rate while vertical cursor movement is continuously controlled by sensorimotor rhythm amplitude. The right edge of the screen is divided among 2-6 targets, and the user's goal is to control vertical cursor movement so that the cursor hits the correct target when it reaches the right edge. Up to the present, vertical cursor movement has been a linear function of amplitude in a specific frequency band [i.e., 8-12 Hz (mu) or 18-26 Hz (beta)] over left and/or right sensorimotor cortex. The present study evaluated the effect of controlling cursor movement with a weighted combination of these amplitudes in which the weights were determined by an regression algorithm on the basis of the user's past performance. Analyses of data obtained from a representative set of trained users indicated that weighted combinations of sensorimotor rhythm amplitudes could support cursor control significantly superior to that provided by a single feature. Inclusion of an interaction term further improved performance. Subsequent online testing of the regression algorithm confirmed the improved performance predicted by the offline analyses. The results demonstrate the substantial value for brain-computer interface applications of simple multivariate linear algorithms. In contrast to many classification algorithms, such linear algorithms can easily incorporate multiple signal features, can readily adapt to changes in the user's control of these features, and can accommodate additional targets without major modifications.