Re-Centering Teams as Dynamic Multilevel Open Systems: Reflections on the Next Decade of Groups and Teams Research

This work deals with well productivity reduction due to altered formation permeability and other flow restrictions around a vertical wellbore. This near-wellbore formation change can be caused by liquid dropout in a gas condensate reservoir, for example. The corresponding reduction in well productivity can be quantified as an increase in the total skin (St). According to the conventional method, the total skin is given by a sum of individual components: St = Sm + Sp + Sd + Sa, where Sm represents flow restrictions from mechanical damage, Sp from partial completion, Sd from non-Darcy flow effect, and Sa from permeability reduction due to liquid dropout. The key finding of this work is that the total skin with condensate dropout can be much bigger than (Sm+Sp+Sd+Sa) due to nonlinear multiplicative fluid-dynamic interactions among skin components. These interactions have not been fully recognized and accounted for in literature. Two correct methods are presented for the total skin calculation in a multi-layered formation: direct simulation and heuristic analytical approximation. The analytical approximation equation is simple, but very accurate in most cases under the pseudo-steady state condition, when compared to the direct simulation method. The new equation developed here has significant implications in various engineering analyses. The conventional method can grossly overestimate well deliverability, if the total skin value is computed from individual components. Conversely, when an individual component (such as Sm) is back-calculated from the total skin obtained from pressure transient analysis, the conventional method can grossly overestimate Sm. Also, this new equation can be used to predict simulation impacts of different wellbore description and of adjustments made in permeability distributions and skin components during history matching.

Thermoacoustic refrigeration systems have gained increased importance in cryogenic cooling technologies and improvements are needed to increase the efficiency and effectiveness of the current cryogenic refrigeration devices. These improvements in performance require a re-examination of the fundamental acoustic and fluid dynamic interactions in the acoustic resonators that comprise a thermoacoustic refrigerator. A comprehensive research program of the pulse tube thermoacoustic refrigerator (PTR) and arbitrarily shaped, circular cross-section acoustic resonators was undertaken to develop robust computational models to design and predict the transport processes in these systems. This effort was divided into three main focus areas: (a) studying the acoustic and fluid dynamic interactions in consonant and dissonant acoustic resonators, (b) experimentally investigating thermoacoustic refrigeration systems attaining cryogenic levels and (c) computationally studying the transport processes and energy conversion through fluid-solid interactions in thermoacoustic pulse tube refrigeration devices. To investigate acoustic-fluid dynamic interactions in resonators, a high fidelity computational fluid dynamic model was developed and used to simulate the flow, pressure and temperature fields generated in consonant cylindrical and dissonant conical resonators. Excitation of the acoustic resonators produced high-amplitude standing waves in the conical resonator. The generated peak acoustic overpressures exceeded the initial undisturbed pressure by two to three times. The harmonic response in the conical resonator system was observed to be dependent on the piston amplitude. The resultant strong acoustic streaming structures in the cone resonator highlighted its potential over a cylindrical resonator as an efficient mixer. Two pulse tube cryogenic refrigeration (PTR) devices driven by a linear motor (a pressure wave generator) were designed, fabricated and tested. The characterization of the systems over a wide range of operating conditions helped to better understand the factors that govern and affect the performance of the PTR. The operating frequency of the linear motor driving the PTR affected the systems performance the most. Other parameters that resulted in performance variations were the mean operating pressure, the pressure amplitude output from the linear motor, and the geometry of the inertance tube. The effect of the inertance tubes geometry was controlled by a single parameter labeled the inertance. External/ambient conditions affected the performance of the cryocoolers too. To prevent the influence of the ambient conditions on the performance, a vacuum chamber was fabricated to isolate the low temperature regions of the PTR from the variable ambient atmosphere. The experiments provided important information and guidelines for the simulation studies of the PTR that were carried out concurrently. A time-dependent high fidelity computational fluid dynamic model of the entire PTR system was developed to gain a better understanding of internal interactions between the refrigerant fluid and the porous heat-exchangers in its various components and to facilitate better design of PTR systems based on the knowledge gained. The compressible forms of the conservation of mass, momentum and energy equations are solved in the gas and porous media (appropriate estimation of fluid dynamics in heat-exchangers) regions. The heat transfer in the porous regions is governed by a thermal non-equilibrium heat transfer model that calculates a separate gas and solid temperature and accounts for heat transfer between the two. The numerical model was validated using both temporal and quasi-steady state results obtained from the experimental studies. The validated model was applied to study the effects of different operating parameters (frequency, pressure and geometry of the components) on the PTRs performance. The simulations revealed interesting steady-periodic flow patterns that develop in the pulse tube due to the fluctuations caused by the piston and the presence of the inertance tube. Similar to the experiments, the simulations provided important information that help guide the design of efficient PTR systems.

Introducing the 2014 JALA Ten Honorees

Summary of Defense Advanced Research Projects Agency's (DARPA) explainable artificial intelligence (XAI) program from the program managers' and evaluator's perspective. Defense Advanced Research Projects Agency (DARPA) formulated the explainable artificial intelligence (XAI) program in 2015 with the goal to enable end users to better understand, trust, and effectively manage artificially intelligent systems. In 2017, the 4-year XAI research program began. Now, as XAI comes to an end in 2021, it is time to reflect on what succeeded, what failed, and what was learned. This article summarizes the goals, organization, and research progress of the XAI program. Dramatic success in machine learning has created an explosion of new AI capabilities. Continued advances promise to produce autonomous systems that perceive, learn, decide, and act on their own. These systems offer tremendous benefits, but their effectiveness will be limited by the machine's inability to explain its decisions and actions to human users. This issue is especially important for the United States Department of Defense (DoD), which faces challenges that require the development of more intelligent, autonomous, and reliable systems. XAI will be essential for users to understand, appropriately trust, and effectively manage this emerging generation of artificially intelligent partners. The problem of explainability is, to some extent, the result of AI's success. In the early days of AI, the predominant reasoning methods were logical and symbolic. These early systems reasoned by performing some form of logical inference on (somewhat) human readable symbols. Early systems could generate a trace of their inference steps, which could then become the basis for explanation. As a result, there was significant work on how to make these systems explainable.1-5 Yet, these early AI systems were ineffective; they proved too expensive to build and too brittle against the complexities of the real world. Success in AI came as researchers developed new machine learning techniques that could construct models of the world using their own internal representations (eg, support vectors, random forests, probabilistic models, and neural networks). These new models were much more effective, but necessarily more opaque and less explainable. The year 2015 was an inflection point in the need for XAI. Data analytics and machine learning had just experienced a decade of rapid progress.6 The deep learning revolution had just begun, following the breakthrough ImageNet demonstration in 2012.6, 7 The popular press was alive with animated speculation about Superintelligence8 and the coming AI Apocalypse.9, 10 Everyone wanted to know how to understand, trust, and manage these mysterious, seemingly inscrutable, AI systems. 2015 also saw the emergence of initial ideas for providing explainability. Some researchers were exploring deep learning techniques, such as the use of deconvolutional networks to visualize the layers of convolutional networks.11 Other researchers were pursuing techniques to learn more interpretable models, such as Bayesian Rule Lists.12 Others were developing model-agnostic techniques that could experiment with a machine learning model—as a black box—to infer an approximate, explainable model, such as LIME.13 Yet, others were evaluating the psychological and human-computer interaction aspects of the explanation interface.13, 14 DARPA spent a year surveying researchers, analyzing possible research strategies, and formulating the goals and structure of the program. In August 2016, DARPA released DARPA-BAA-16-53 to call for proposals. The stated goal of explainable artificial intelligence (XAI) was to create a suite of new or modified machine learning techniques that produce explainable models that, when combined with effective explanation techniques, enable end users to understand, appropriately trust, and effectively manage the emerging generation of AI systems. The target of XAI was an end user who depends on decisions or recommendations produced by an AI system, or actions taken by it, and therefore needs to understand the system's rationale. For example, an intelligence analyst who receives recommendations from a big data analytics system needs to understand why it recommended certain activity for further investigation. Similarly, an operator who tasks an autonomous system needs to understand the system's decision-making model to appropriately use it in future missions. The XAI concept was to provide users with explanations that enable them to understand the system's overall strengths and weaknesses; convey an understanding of how it will behave in future/different situations; and perhaps permit users to correct the system's mistakes. The XAI program assumed an inherent tension between machine learning performance (eg, predictive accuracy) and explainability, a concern that was consistent with the research results at the time. Often the highest performing methods (eg, deep learning) were the least explainable and the most explainable (eg, decision trees) were the least accurate. The program hoped to create a portfolio of new machine learning and explanation techniques to provide future practitioners with a wider range of design options covering the performance-explainability trade space. If an application required higher performance, the XAI portfolio would include more explainable, high-performing, deep learning techniques. If an application required more explainability, XAI would include higher performing, interpretable models. The program was organized into three major technical areas (TAs), as illustrated in Figure 1: (a) the development of new XAI machine learning and explanation techniques for generating effective explanations; (b) understanding the psychology of explanation by summarizing, extending and applying psychological theories of explanation; and (c) evaluation of the new XAI techniques in two challenge problem areas: data analytics and autonomy. The original program schedule consisted of two phases: phase 1, Technology Demonstrations (18 months); and phase 2, Comparative Evaluations (30 months). During phase 1, developers were asked to demonstrate their technology against their own test problems. During phase 2, the original plan was to have developers test their technology against one of two common problems (Figure 2) defined by the government evaluator. At the end of phase 2, the developers were expected to contribute prototype software to an open source XAI toolkit. In May 2017, XAI development began. Eleven research teams were selected to develop the Explainable Learners (TA1) and one team was selected to develop the Psychological Models of Explanation. Evaluation was provided by the Naval Research Lab. The following summarizes those developments and the final state of this work at the end of the program. An interim summary of the XAI developments at the end of 2018 is given in Gunning and Aha.15 The program anticipated that researchers would examine the training process, model representations, and, importantly, explanation interfaces. Three general approaches were envisioned for model representations. Interpretable model approaches would seek to develop ML models that were inherently more explainable and more introspectable for machine learning experts. Deep explanation approaches would leverage deep learning or hybrid deep learning approaches to produce explanations in addition to predictions. Finally, model induction techniques would create approximate explainable models from more opaque, black-box models. Explanation interfaces were expected to be a critical element of XAI, connecting a user to the model to enable them to understand and interact with the decision making process. As the research progressed, 11 XAI teams explored a number of machine learning approaches, such as tractable probabilistic models16 and causal models and explanation techniques such as state machines generated by reinforcement learning algorithms,17 Bayesian teaching,18 visual saliency maps,19-24 and network and GAN dissection.24-26 Perhaps the most challenging and most unique contributions came from the combination of machine learning and explanation techniques27 to conduct well-designed psychological experiments to evaluate explanation effectiveness.28-31 As the program progressed, we also gained a more refined understanding of the spectrum of users and development timeline (Figure 3). The program structure anticipated the need for a grounded psychological understanding of explanation. One team was selected to summarize current psychological theories of explanation to assist the XAI developers and the evaluation team. This work began with an extensive literature survey on the psychology of explanation and previous work on explainability in AI.32 Originally, this team was asked to (a) produce a summary of current theories of explanation, (b) develop a computational model of explanation from those theories, and (c) validate the computational model against the evaluation results from the XAI developers. Developing computational models proved to be a bridge too far, but the team did gain a deep understanding of the area and successfully produced descriptive models. These descriptive models were critical to supporting the effective evaluation approaches, which involved carefully designed user studies, carried out in accordance with DoD human subject research guidelines. Figure 4 illustrates a top-level descriptive model of the XAI explanation process. Evaluation was originally envisioned to be based on a common set of problems, within the data analytics and autonomy domains. However, it quickly became clear that it would be more valuable to explore a variety of approaches across a breadth of problem domains. In order to evaluate the performance in the final year of the program, the evaluation team, led by Eric Vorm, PhD, of the US Naval Research Laboratory (NRL), developed an explanation scoring system (ESS). This scoring system provided a quantitative mechanism for assessing the designs of XAI user studies in terms of technical and methodological appropriateness and robustness. The ESS enabled the assessments of multiple elements of each user study, including the task, domain, explanations, explanation interface, users, hypothesis, data collection, and analysis to ensure that each study met the high standards of human subject research. XAI evaluation measures are shown in Figure 5, and include functional measures, learning performance measures, and explanation effectiveness measures. The DARPA XAI program demonstrated definitively the importance of carefully designing user studies in order to accurately evaluate the effectiveness of explanations in ways that directly enhance appropriate use and trust by human users, and appropriately support human-machine teaming. Often times, multiple types of measures (ie, performance, functionality, and explanation effectiveness) will be necessary to evaluate the performance of an XAI algorithm. XAI user study design can be tricky and the DARPA XAI program discovered that the most effective research teams were ones that featured diverse teams with cross-disciplinary expertise (ie, computer science combined with human-computer interaction and/or experimental psychology, etc.). The XAI program explored many approaches, as shown in Table 1. Interactive debugger interface for visualizing poisoned training datasets. Work is applied on the IARPA TrojAI dataset.33 Establishing objective/quantitative criteria to assess value of explanations for ML models34 CNN-based one-shot detector, using network dissection to identify the most salient features41 Explanations produced by heat maps and text explanations42 Human-machine common ground modeling Indoor navigation with a robot (in collaboration with GA Tech) Video Q&A Human-assisted one-shot classification system by identifying the most salient features Three major evaluations were conducted during the program: one during phase 1 and two during phase 2. In order to evaluate the effectiveness of XAI techniques, researchers on the program designed and executed user studies. User studies are still the gold standard for assessing explanations. There were approximately 12 700 participants in user studies carried out by XAI researchers, including approximately 1900 supervised participants, where the individual was guided through the experiment by the research team (eg, in person or on Zoom) and 10 800 unsupervised participants, where the individual self-guided through the experiment and was not actively guided by the research team (eg, Amazon Mechanical Turk). In accordance with policy for all US DoD funded human subjects research, each research protocol was reviewed by a local Institutional Review Board (IRB) and then a DoD human research protection office reviewed the protocol and the local IRB findings. As mentioned earlier, there seemed to be a natural tension between learning performance and explainability. However, throughout the course of the program, we found evidence that explainability can improve performance. From an intuitive perspective, training a system to produce explanations provides additional supervision, via additional loss functions, training data, or other mechanisms, that encourages a system to learn more effective representations of the world. While this may not be true in all cases and significant work remains to characterize when explainable techniques will be more performant, it provides hope that future XAI systems can be more performant than current systems while meeting user needs for explanations. There currently is no universal solution to XAI. As discussed earlier, different user types require different types of explanations. This is no different from what we face interacting with other humans. Consider, for example, a doctor needing to explain a diagnosis to a fellow doctor, a patient, or a medical review board. Perhaps future XAI systems will be able to automatically calibrate and communicate explanations to a specific user within a large range of user types, but that is still significantly beyond the current state of the art. One of the challenges in developing XAI is measuring the effectiveness of an explanation. DARPA's XAI effort has helped develop foundational technology in this area, but much more needs to be done, including drawing more from the human factors and psychology communities. Measures of explanation effectiveness need to be well established, well understood, and easily implemented by the developer community in order for effective explanations to become a core capability of ML systems. UC Berkeley's result21 demonstrating that advisability, the ability for an AI system to take advice from a user, improves user trust beyond explanations is intriguing. Certainly, users will likely prefer systems where they can quickly correct the behavior of a system in the same ways that humans can provide feedback to each other. Such advisable AI systems that can both produce and consume explanations will be key to enabling closer collaborations between humans and AI systems. Close collaboration is required across multiple disciplines including computer science, machine learning, artificial intelligence, human factors, and psychology, among others, in order to effectively develop XAI techniques. This can be particularly challenging, as researchers tend to focus on a single domain and often need to be pushed to work across domains. Perhaps in the future a XAI-specific research discipline will be created at the intersection of multiple current disciplines. Toward this end, we have worked to create an XAI Toolkit, which collects the various program artifacts (eg, code, papers, reports, etc.) and lessons learned from the 4-year DARPA XAI program into a central, publicly accessible location (https://xaitk.org/).48 We believe the toolkit will be of broad interest to anyone who deploys AI capabilities in operational settings and needs to validate, characterize, and trust AI performance across a wide range of real-world conditions and application areas. Today, we have a more nuanced, less dramatic, and, perhaps, more accurate understanding of AI than we had in 2015. We certainly have a more accurate understanding of the possibilities and the limitations of deep learning. The AI apocalypse has faded from an imminent danger to a distant curiosity. Similarly, the XAI program has produced a more nuanced, less dramatic, and, perhaps, more accurate understanding of XAI. The program certainly acted as a catalyst to stimulate XAI research (both inside and outside of the program). The results have produced a more nuanced understanding of XAI uses and users, the psychology of XAI, the challenges of measuring explanation effectiveness, as well as producing a new portfolio of XAI ML and HCI techniques. There is certainly more work to be done, especially as new AI techniques are developed that will continue to need explanation. XAI will continue as an active research area for some time. The authors believe that the XAI program has made a significant contribution by providing the foundation to launch that endeavor. David Gunning (now retired) is a three-time DARPA program manager, who created and managed the XAI program from its inception in 2016 to its mid-point in 2019. His portfolio of DARPA research programs made significant contributions to the development of AI over the past 25 years. He led the Personalized Assistant that Learns (PAL) program, which produced the technologies behind Apple's Siri. His Command Post of the Future (CPoF) program was later adopted by the US Army as their Command and Control system for use during the Iraq and Afghanistan conflicts. Between DARPA tours, David served in senior positions at Facebook AI, Palo Alto Research Center, Vulcan Inc, Cycorp and co-founded SET Corp. Eric Vorm, PhD, is a cognitive systems engineer and serves as the Deputy Director for the Laboratory for Autonomous Systems Research at the US Naval Research Laboratory in Washington, DC. Dr Vorm led the evaluation team for the DARPA Explainable AI program, and led the development of the first comprehensive criteria for the evaluation of explanations generated by machine learning. Dr Vorm's research focuses on the design of intelligent systems to achieve ideal human-machine teaming, with special emphasis on the role of transparency and explainability in supporting appropriate trust, safety, and reliability in high-risk, time-sensitive operational domains. Jennifer Yunyan Wang, PhD, is a computational neuroscientist with a special focus on AI. As Systems, Engineering and technical Assistance (SETA) contractor to DARPA, she provided technical support and expertise to several programs including XAI, L2M, GARD, and AIE RED. After finishing postdoctoral fellowships in experimental neuroscience at Johns Hopkins University and the Food and Drug Administration, Jennifer joined Quantitative Scientific Solutions in 2018 as a consultant for government R&D and think tanks including IARPA and Center for Security and Emerging Technology. Matt Turek, PhD, joined DARPA's Information Innovation Office (I2O) as a program manager in July 2018 and took over as program manager of the XAI program in 2019. His portfolio also includes the Media Forensics (MediFor), Semantic Forensics (SemaFor), and Machine Common Sense (MCS) programs, as well as the Reverse Engineering of Deceptions (RED) AI Exploration. His research interests include computer vision, machine learning, artificial intelligence, and their application to problems with significant societal impact. Prior to his position at DARPA, Turek led a team at Kitware Inc developing computer vision technologies including large scale behavior recognition and modeling, object detection and tracking, activity recognition, normalcy modeling, and anomaly detection. Data sharing is not applicable to this article as no new data were created or analyzed in this editorial.

Open Multiagent Systems, in which heterogeneous agents interact with each other and organize themselves into Virtual Organizations, demand infrastructures supporting these features. In these systems, dynamic and complex interactions between agents may arise. Interaction Protocols allow the definition of communication patterns. However in open systems, dynamic and complex interactions may also require these patterns be modified at execution time. We propose a support for modelling complex, concurrent and dynamic interactions between agents in terms of conversations. Conversations between agents follow predefined Interaction Protocols that can be dynamically modified without restarting the system. This support is provided at agent level and is integrated into the Magentix Multiagent Platform.

An open thermodynamic system exchanges energy or information with its environment. The general systems models of an open thermodynamic system have been obtained in this paper. These models are the non-integrable open system and the non-ergodic open system. These models are obtained by presenting the open system dynamics as unpredictable evolution. Evolution in both models is set as an autonomous action which is the observable projection of the open system dynamics. Unpredictability is presented as ambiguity in the case of the non-integrable open system, and as indeterminacy—in the case of the non-ergodic open system. It has been shown that the unpredictable evolution determines an encoding for the non-integrable open system, and a monochromatic subgraph—for the non-ergodic open system. The redundancy and entropy for both models have been introduced by using these combinatorial mechanisms. The application of the two open system models in artificial intelligence and in the functional diagnostics of dynamic systems has been discussed

There is scarce research on how working teams adjust themselves to external conditions such as time. It is crucial to understand how teams deal with such circumstances so as to provide resources for them. We present a diary study, which focuses on the complete life spans of collaboration of four research teams. Each team of two was assigned to choose a research topic and prepare a presentation on an assigned date. The findings suggest that groups' interaction, collaboration and communication media choice were triggered by members' perception of time and research deadlines. This poster proposes that timing and teams' dynamic interaction and collaboration are inter-dependent, and that teams prefer rich media to compensate for the time constraints. Implications for theory, research, and practice are drawn.

The Dean and Betty Gallo Prostate Cancer Center (GPCC) was established with the goal of eradicating prostate cancer and improving the lives of men at risk for the disease through research, treatment, education and prevention. GPCC was founded in the memory of Dean Gallo, a beloved New Jersey Congressman who died tragically of prostate cancer diagnosed at an advanced stage. GPCC unites a team of outstanding researchers and clinicians who are committed to high-quality basic research, translation of innovative research to the clinic, exceptional patient care, and improving public education and awareness of prostate cancer. GPCC is a center of excellence of The Cancer Institute of New Jersey, which is the only NCI-designated comprehensive cancer center in the state. GPCC efforts are now integrated well as part of our Prostate Program at CINJ, in which Dr. Robert DiPaola and Dr. Cory Abate-Shen are co-leaders. The Prostate Program unites 19 investigators from 10 academic departments who have broad and complementary expertise in prostate cancer research. The overall goal and unifying theme is to elucidate basic mechanisms of prostate growth and oncogenesis, with the ultimate goal of promoting new and effective strategies for the eradication of prostate cancer. Members' wide range of research interests collectively optimize the chances of providing new insights into normal prostate biology and unraveling the molecular pathophysiology of prostate cancer. Cell culture and powerful animal models developed by program members recapitulate the various stages of prostate cancer progression, including prostatic intraepithelial neoplasia, adenocarcinoma, androgen-independence, invasion and metastases. These models promise to further strengthen an already robust program of investigator-initiated therapeutic clinical trials, including studies adopted by national cooperative groups. Efforts to translate laboratory results into clinical studies of early detection and chemoprevention are underway. The specific goals of this program are: (1) To investigate the molecular mechanisms underlying normal prostate growth and differentiation and elucidate the molecular mechanisms underlying prostate oncogenesis. (2) To build on fundamental knowledge to develop effective therapeutic approaches for the treatment of prostate cancer. (3) To improve the control of prostate cancer through early detection, chemoprevention, and outreach and education. This new disease-based program is structured to improve interdisciplinary interactions and translational results. Already, through the dynamic leadership of Drs. Cory Abate-Shen and Robert DiPaola, new investigators were attracted to the field, new collaborations engendered, and numerous investigator-initiated trials implemented. Progress in GPCC and the program overall has been outstanding. The Center has success in uniting investigators with broad and complementary expertise in prostate cancer research. The overall goal and unifying theme is to elucidate basic mechanisms of prostate growth and oncogenesis, with the ultimate goal of promoting new and effective strategies for the eradication of prostate cancer in patients and populations at risk. Members wide range of research interests collectively optimize the chances of providing new insights into normal prostate biology and unraveling the molecular pathophysiology of prostate cancer. Studies in cell culture and powerful animal models developed recapitulate the various stages of prostate cancer progression, including prostatic intraepithelial neoplasia, adenocarcinoma, androgen-independence, invasion and metastases. These models promise to further strengthen an already robust program of investigator-initiated therapeutic clinical trials, including studies adopted by national cooperative groups. Efforts to translate laboratory results into clinical studies of early detection and chemoprevention are underway.

Chapter 8 - Coevolution of Tumor Cells and Their Microenvironment: “Niche Construction in Cancer”

: We held a symposium, in which questions arising from the interaction of fluid dynamics, applied mathematics and dynamical systems theory in the environment, in flows relevant to aircraft and in model flows, were addressed. This was a relatively small symposium, of fewer than 50 people, so that active and informal discussion could take place. We intended to publish these discussions with the proceedings. The Symposium was entitled Fluid Mechanics and the Environment: Dynamical Approaches. There were twenty-three invited papers, and only two posters.

CITATION: Balvanera, P., et al. 2017. Key features for more successful place-based sustainability research on social-ecological systems : a Programme on Ecosystem Change and Society (PECS) perspective. Ecology and Society, 22(1):14, doi:10.5751/ES-08826-220114.

One of the biggest challenges of engineering is enable computational solutions that reduce processing time and provide more accurate numerical solutions. Proposals with several approaches that explore new ways of solving such problems or improve existing solutions emerge. One of the biggest areas dedicated to propose such improvements is the parallel and high performance computing. Techniques that improve the processing time, more efficient algorithms and faster computers open up new horizons allowing to perform tasks that were previously unfeasible or would take too long to complete. We can point out, among several areas of interest, Fluid Dynamics and Interaction Fluid-Structure. In this work it is developed a parallel computing architecture in order to solve numerical problems more efficiently, compared to sequential architecture (e.g. Fluid Dynamics and Fluid-Structure Interaction problems) and it is also possible to extend this architecture to solve different problems (e.g. Structural problems). The objective is to develop an efficient computational algorithm in scientific programming language C ++, based on previous work carried out in Computational Mechanics Laboratory (CML) at Polytechnic School at University of So Paulo, and later with the developed architecture, execute and investigate Fluid Dynamics and Fluid-Structure Interaction problems with the aid of CML computers. A sensitivity analysis is executed for different problems in order to assess the best combination of elements quantity and speedup, and then a perfomance comparison. Using only one GPU, we could get a 10 times speedup compared to a sequential software, using the Finite Element with Immersed Boundary Method and a direct solver (PARDISO).

: The significant results of a joint research effort investigating the fundamental fluid dynamic mechanisms and interactions within high-speed separated flows are presented in detail. The results have been obtained through primary emphasis on experimental investigations of missile and projectile base flow-related configurations. The objectives of the research program focus on understanding the component mechanisms and interactions which establish and maintain high-speed separated flow regions. The experimental efforts have considered the development and use of state-of-the-art laser Doppler velocimeter (LDV) and particle image velocimeter (PIV) systems for experiments with axisymmetric and planar, two-dimensional models in subsonic and supersonic flows. The LDV experiments have yielded high quality, well documented mean and turbulence velocity data for a variety of high-speed separated flows including the near-wake region behind a cylindrical afterbody in supersonic flow. The PIV experiments have studied the effect of a base cavity in a two-dimensional, subsonic base flow and the mechanism of drag reduction for this configuration. Another experimental study has considered the interaction occurring when a supersonic stream is separated by means of a second stream impinging the first at an angle (plume-induced separation). The results of these various studies have been carefully documented in a series of journal articles, conference proceedings papers, and theses. The fun text of the papers and thesis abstracts are included as appendices of this report. Separated flow, Transonic flow, Particle image velocimetry, Base flow, Supersonic flow, Subsonic flow, Laser Doppler velocimetry,

: This report describes the equipment purchased under ARO Research Instrumentation Grant No. DAAH04-94-G-0386. This equipment has been used to support the ARO research program entitled, 'Fluid Dynamic Mechanisms and Interactions with Separated Flows' (ARO Grant No. DAAH04-93-G-0226). In particular, instrumentation has been purchased to obtain measurements and flow visualizations using schlieren and shadowgraph photography, laser Doppler velocimetry, particle image velocimetry, and Mie/Rayleigh scattering. The specific pieces of equipment, vendors, and purchase prices are detailed.

Proximal Stenosis Is Associated with Rupture Status in Middle Cerebral Artery Aneurysms.