In the first part of this thesis, GPS data for a year's worth of travel by 255 Seattle households is used to illuminate how plug-in electric vehicles (PEVs) can match household needs. Data from all vehicles in each of these households were analyzed at a disaggregate level primarily to determine whether each household would be able to adopt various types of PEVs without significant issues in meeting travel needs. The results suggest that a battery-electric vehicle (BEV) with 100 miles of all-electric range (AER) should meet the needs of 50% of Seattle's one-vehicle households and the needs of 80% of the multiple-vehicle households, when households charge just once a day and rely on another vehicle or mode just 4 days a year. Moreover, the average one-vehicle Seattle household uses each vehicle 23 miles per day and should be able to electrify close to 80% of its miles, while meeting all its travel needs, using a plug-in hybrid electric vehicle with 40-mile all-electric-range (PHEV40). Households owning two or more vehicles can electrify 50 to 70% of their total household miles using a PHEV40, depending on how they assign the vehicle across drivers each day. Cost comparisons between the average single-vehicle household owning a Chevrolet Cruze versus a Volt PHEV suggest that, when gas prices are $3.50 per gallon and electricity rates are 11.2 ct per kWh, the Volt will save the household $535 per year in energy/fuel costs. Similarly, the Toyota Prius PHEV will provide an annual savings of $538 per year over the Corolla. The results developed in this research provide valuable insights into the role of AER on PEV adoption feasibility and operating cost differences. The second part of this thesis uses detailed travel data from the Seattle metropolitan area to evaluate the effects of built-environment variables on the use of non-motorized (bike + walk) modes of transport. Several model specifications are used to understand and explain non-motorized travel behavior in terms of household, person and built-environment variables. Land-use measures like land-use mix, density, and accessibility indices were also created and incorporated as covariates to appreciate their marginal effects. The models include a count model for household vehicle ownership levels, a binary choice model for the decision to stay within versus departing one's origin zone (i.e., intra- versus inter-zonal trip-making), discrete choice models for destination choices and mode choices, and a zero-inflated negative binomial model for non-motorized trip counts per household. The mode and destination choice models were estimated separately for interzonal and intrazonal trips and for each of three different trip types (home-based work, home-based non-work, and non-home-based), to recognize the distinct behaviors at play when making shorter versus longer trips and different types of trips. This comprehensive set of models highlights how built-environment variables -- like the number and type of intersections present around one's origin and destination, the number of bus stops available within a certain radius, household and jobs densities, parking prices, land use mixing, and walk-based accessibility -- can significantly shape the pattern of one's non-motorized movement. The results underscore the importance of street connectivity (quantified as the number of 3-way and 4-way intersections in a half-mile radius), higher bus stop density, and greater non-motorized access in promoting lower vehicle ownership levels (after controlling for household size, income, neighborhood density and so forth), higher rates of non-motorized trip generation (per day), and higher likelihoods of non-motorized mode choices. Destination choices are also important for mode choices, and local trips lend themselves to more non-motorized options than more distance trips. Intrazonal trip likelihoods rose with higher street connectivity, transit availability, and land use mixing. For example, the results suggest that an increase in the land-use mix index by 10% would increase the probability of choosing to travel within the zone by 12%. As expected destinations with greater population and job numbers (attraction), located closer (to a trip's origin), offering lower parking prices and greater transit availability, were more popular. Interestingly, those with more dead ends (or cul de sacs) attracted fewer trips. Among all built environment variables tested, street structure offered the greatest predictive benefits, alongside jobs and population (densities and counts). For example, a 1-percent increase in the average number of 4-way intersections within a quarter-mile radius of the sampled households is estimated to increase the average household's non-motorized trip generation by 0.36%. A one-standard-deviation increase in the (mean) number of 4-way intersections at the average trip origin is estimated to increase the probabilities of bike and walk modes for interzonal home-based-work trips by 57% and 30%, respectively. In contrast, increasing the number of dead-ends at the origin by one standard deviation is estimated to decrease the probability of biking for both home-based-work and non-work trips by ~30%. These results underscore the importance of network density and connectivity for promoting non-motorized activity. The regional non-motorized travel (NMT) accessibility index ( derived from the logsum of a destination choice model) also offers strong predictive value, with NMT counts rising by by 7% following a 1% increase in this variable -- if the drive alone accessibility index is held constant (along with all other variables, evaluated at their means). Similarly, household vehicle ownership is expected to fall by 0.36% with each percentage point increase in the NMT accessibility index, and walk probabilities rise by 26.9% following a one standard deviation increase in this index at the destination zone. A traveler's socio-economic attributes also have important impacts on NMT choices, with demographics typically serving as much stronger predictors of NMT choices than the built environment. For example, the elasticity of NMT trip generation with respect to a household's vehicle ownership count is estimated to be -0.52. Males and tose with drivers licenses are estimated to have 17% and 39% lower probabilities, respectively, of staying within their origin zone, relative to women and unlicensed adults (ceteris paribus). Non-motorized model choices also exhibit strong sensitivity to age and gender settings. Several of the regional variables developed in this work, and then used in the predictive models, are highly correlated. For example, bus stop and intersection densities are very high in job- and population-dense areas. For example, the correlation co-efficients between the bus stop density and 4-way intersection density is 0.805, between NMT and SOV AIs is 0.830 and between 4-way intersection density and NMT AI is 0.627. As a result, many variables are proxying for and/or competing with each other, as is common in models with many land use covariates, and it is difficult to quantify the exact impact of each of these variables. Nonetheless the models developed here provide valuable insight into the role of several new variables on non-motorized travel choices. Some final case study applications, moving all households to the downtown area (that has high accessibility indices and density), illustrate to what extent these revealed-data-based models will predict shifts toward and away from non-motorized trip-making. It appears that average household vehicle ownership level reduces to 0.57 from 1.89 (a 70% reduction) and average two-day NMT trip generation increases to 5.92 from 0.83 (an increase of more than 6 times). Such ranges are valuable to have in mind, when communities seek to reduce reliance on motorized travel by defining new built-environment contexts. / text

Made available in DSpace on 2014-10-09T12:35:14Z (GMT). No. of bitstreams: 0 / Made available in DSpace on 2014-10-09T13:58:01Z (GMT). No. of bitstreams: 0 / Dissertação (Mestrado) / IPEN/D / Instituto de Pesquisas Energeticas e Nucleares - IPEN-CNEN/SP

A space camera is a promising candidate to address the non-stop rising interest for astrophysics research in the Compton regime. The MeVCube mission is intended to be launched in 2022, hosting an on-board Compton Camera. To better support the development of the instrument in this early stage, a series of feasibility studies to assess two potential launch orbits were performed. The studies were composed by a series of mission analysis simulations which permitted the characterisation of the orbital environments for the two orbital options. Several sources of background noise to the instrument were identified. The population of trapped protons and trapped electrons were simulated for the periods of Solar Minimum and Solar Maximum, as well as the levels of Galactic Cosmic Ray (GCR) flux. The performance of trade-off studies concluded that an equatorial orbit is more preferable for reducing the influence of background noise. To better estimate the environment effects at the equatorial orbit, the number of particles which can penetrate the detector shielding were simulated. The next step was to perform a series secondary studies whose aim were to simulate the induced current on the electrodes, produced by the interactions occurring within the detector. The actualisation of these simulations required the study of photon interaction with matter, the various Cadmium-Zink-Telluride (CZT) types and the how they operate, and the use of a sophisticated software to perform the appropriate simulations. COMSOL, which allows the method of FEA, was chosen as the tool to perform the simulations. The geometry of the detector voxel was primarily designed in SIEMENS NX. The geometry was inserted into COMSOL, where a number of iterations were performed to finalise the appropriate mesh size, which ensured an accurate representation of the Electric field and the Weighting potential within the detector voxel. The induced current on the electrodes was decided to be calculated via MATLAB. As a verification step it was thought useful to firstly plot the weighting potential of the three electrodes under test; the chosen anode pixel, the steering grid and the cathode. The process revealed a series of numerical errors, most likely introduced by the type of mesh chosen or by the data manipulation process via MATLAB. Significant reduction of the numerical errors would lead to more accurate values for the induced current. Unfortunately, due to time constraints this was a task that was not completed. Solving this problem would be optimal for future studies with MATLAB, as the induced current on the electrodes can be correctly calculated based on charge transport within the detector bulk. / MeVCube, DESY

<p dir="ltr">The genetic perturbations caused by spaceflight on biological systems tend to have a system-wide effect which is often difficult to deconvolute into individual signals with specific points of origin. Single cell multi-omic data can provide a profile of the perturbational effects but does not necessarily indicate the initial point of interference within a network. The objective of this project is to take advantage of large scale and genome-wide perturbational or Perturb-Seq datasets by using them to pre-train a generalist machine learning model that is capable of predicting the effects of unseen perturbations in new data. Perturb-Seq datasets are large libraries of single cell RNA sequencing data collected from CRISPR knock out screens in cell culture. The advent of generative machine learning algorithms, particularly transformers, make it an ideal time to re-assess large scale data libraries in order to grasp cell and even organism-wide genomic expression motifs. By tailoring an algorithm to learn the downstream effects of the genetic perturbations, we present a pre-trained generalist model capable of predicting the effects of multiple perturbations in combination, locating points of origin for perturbation in new datasets, predicting the effects of known perturbations in new datasets, and annotation of large-scale network motifs. We demonstrate the utility of this model by identifying key perturbational signatures in RNA sequencing data from spaceflown biological samples from the NASA Open Science Data Repository.</p>