Les systèmes cyber-physiques (CPS) et d’Internet des objets génèrent un volume et une variété des données sans précédant. Le temps que ces données parcourent le réseau dans son chemin vers le cloud, la possibilité de réagir à un événement critique pourrait être tardive. Pour résoudre ce problème, les traitements de données nécessitant une réponse rapide sont faits à proximité d’où les données sont collectées. Ainsi, seuls les résultats du pré-traitement sont envoyées au cloud et la réaction pourrai être déclenché suffisamment rapide pour préserver l’intégrité du système. Ce modèle de calcul est connu comme Fog Computing. Un large spectre d’applications de CPS ont des contraintes temporelle et peuvent être facilement parallélisées en distribuant les calculs sur différents sous-ensembles de données en même temps. Ceci peut permettre d’obtenir un temps de réponse plus court et un temps de creux plus large. Ainsi, on peut réduire la fréquence du processeur et/ou éteindre des parties du processeur afin de réduire la consommation d’énergie. Dans cette thèse, nous nous concentrons sur le problème d'ordonnancement d’un ensemble de taches temps-réels parallèles sur des architectures multi-coeurs dans l’objectif de réduire la consommation d’énergie en respectant toutes les contraintes temporelles. Nous proposons ainsi plusieurs modèles de tâches et des testes d'ordonnançabilité pour résoudre le problème d’allocation des threads aux processeurs. Nous proposons aussi des méthodes qui permettent de sélectionner les fréquences et les états des processeurs. Les modèles proposés peuvent être implantés comme des directives dans la même logique que OpenMP. / Cyber physical systems (CPS) and Internet of Objects (IoT) are generating an unprecedented volume and variety of data that needs to be collected and stored on the cloud before being processed. By the time the data makes its way to the cloud for analysis, the opportunity to trigger a reply might be late. One approach to solve this problem is to analyze the most time-sensitive data at the network edge, close to where it is generated. Thus, only the pre-processed results are sent to the cloud. This computation model is know as *Fog Computing* or *Edge computing*. Critical CPS applications using the fog computing model may have real-time constraints because results must be delivered in a pre-determined time window. Furthermore, in many relevant applications of CPS, the processing can be parallelized by applying the same processing on different sub-sets of data at the same time by the mean parallel programming techniques. This allow to achieve a shorter response time, and then, a larger slack time, which can be used to reduce energy consumption. In this thesis we focus on the problem of scheduling a set of parallel tasks on multicore processors, with the goal of reducing the energy consumption while all deadlines are met. We propose several realistic task models on architectures with identical and heterogeneous cores, and we develop algorithms for allocating threads to processors, select the core frequencies, and perform schedulability analysis. The proposed task models can be realized by using OpenMP-like APIs.

Diesel particulate matter (DPM) is the solid portion of diesel exhaust, which occurs primarily in the submicron range. It is complex in nature, occuring in clusters and agglomerated chains, and with variable composition depending on engine operating conditions, fuel type, equipment maintenance, etc. DPM is an occupational health hazard that has been associated with lung cancer risks and other respiratory issues. Underground miners have some of the highest exposures to DPM, due to work in confined spaces with diesel powered equipment. Large-opening mines present particular concerns because sufficient ventilation is very challenging. In such environments, reliable DPM sampling and monitoring is critical to protecting miner health. Though complex, DPM is made up primarily of elemental (EC) and organic carbon (OC), which can be summed to obtain total carbon (TC). The Mine Safety and Health Administration (MSHA) currently limits personal DPM exposures in metal/non-metal mines to 160 µg/m3 TC on an 8-hour time weighted average. To demonstrate compliance, exposures are monitored by collecting filter samples, which are sent to an outside lab and analyzed using the NIOSH 5040 Standard Method. To support real-time results, and thus more timely decision making, the Airtec handheld DPM monitor was developed. It measures EC, which is generally well correlated with TC, using a laser absorption technique as DPM accumulates on a filter sample. Though intended as a personal monitor, the Airtec has application as an engineering tool. A field study is reported here which demonstrated the usefulness of the Airtec in tracking temporal and spatial trends in DPM. An approach to sensitizing the monitor to allow "spot checking" was also demonstrated. Since DPM in mine environments generally occurs with other airborne particulates, namely dust generated during the mining process, DPM sampling must be done with consideration for analytical interferences. A common approach to dealing with mineral dust interferences is to use size selectors in the sampling train to separate DPM from dust; these devices are generally effective because DPM and dust largely occur in different size ranges. An impactor-type device (DPMI) is currently the industry standard for DPM sampling, but it is designed as a consumable device. Particularly for continuous monitoring applications, the sharp cut cyclone (SCC) has been suggested as a favorable alternative. In another field study reported here, the effect of aging (i.e., loading as an artifact of sampling) on the DPMI and SCC was investigated. Results suggest the effective cut size of the DPMI will be reduced much more rapidly than that of the SCC with aging — though even in a relatively high dust, high DPM environment, the DPMI performs adequately. In a third field study, the possibility of attachment between DPM and respirable dust particles was investigated. Such a phenomenon may have implications for both reliable sampling and health outcomes. Data collected by transmission electron microscope (TEM) on samples collected in the study mine showed that DPM-dust attachment does indeed occur. Moreover, the study results suggest that respirable particulate sampling — as opposed to submicron sampling, which is currently used — may be favorable for ensuring that oversized DPM is not excluded from samples. This strategy may require additional sample preparation to minimize dust interferences, but methods have been previously developed and were demonstrated here. / Master of Science

The term diesel particulate matter (DPM) is used to refer to the solid phase of diesel exhaust, which is mainly composed of elemental carbon and organic carbon. DPM is generally in the nano-size range (i.e., 10-1,000 nm). Occupational exposure is a health concern, with effects ranging from minor eye and respiratory system irritation to major cardiovascular and pulmonary diseases. Significant progress has been made in reducing DPM emissions by improving fuels, engines and after-treatment technologies. However, the mining industry, in particular, remains challenged to curb exposures in some operations where relatively many diesel engines are working in confined environments with relatively low airflow. Basic theory and a limited amount of prior research reported in the literature suggest that water sprays may be able to scavenge airborne DPM. The goals of the work presented in this thesis were to build an appropriate laboratory set up and to test the efficacy of micron-scale water (or fog) droplets to remove DPM from an air stream. The general experimental approach was to direct diesel exhaust through a chamber where fog drops are generated, and to measure DPM up- and down-stream of the treatment. Initially, fundamental experiments were conducted to explore the effect of the fog drops on the removal of (electrically neutralized) DPM from a dry exhaust stream. Compared to no treatment (i.e., control) and with the use of a diffusion dryer downstream of the fog treatment, the fog improved DPM removal by about 57% by mass and 45% by number density (versus no treatment). Without the use of the diffusion dryer, improvement in DPM removal was about 19% by mass. Analysis of the results suggests that a likely mechanism for the DPM removal in this experimental system is thermal coagulation between DPM and fog droplets, followed by gravitational settling and/or impaction of the droplets with system components. Further tests using raw exhaust (i.e., neither dried nor neutralized) having a higher DPM number density; shorter residence times; additional fogging devices; and no diffusion dryer downstream of the fog treatment were also carried out. These yielded an average overall improvement in DPM mass removal of about 45% attributed to the fog treatment (versus no treatment). The significant increase in DPM removal in these tests compared to the initial test (i.e., 19% removal by mass) cannot be fully explained by differences in residence time or DPM and fog droplet densities. Increased humidity in the system (due to the undried exhaust) may have allowed for a larger mean droplet size, and therefore might explain more rapid settling of DPM-laden droplets. Another possible contributing factor is ambient surface charge of the DPM, which might perhaps result in more efficient attachment between DPM and fog drops and/or increased deposition loses in the system. / Master of Science

Submitted by Marco Antônio de Ramos Chagas (mchagas@ufv.br) on 2017-06-19T13:31:01Z No. of bitstreams: 1 texto completo.pdf: 374341 bytes, checksum: 3a803e02d833c5ebb271e242c66f17ea (MD5) / Made available in DSpace on 2017-06-19T13:31:01Z (GMT). No. of bitstreams: 1 texto completo.pdf: 374341 bytes, checksum: 3a803e02d833c5ebb271e242c66f17ea (MD5) Previous issue date: 2000-07-14 / A maioria dos alimentos que compõem a ração dos animais em uma propriedade agrícola deve ser desintegrada, picada ou moída, utilizando-se equipamentos próprios, entre eles os denominados DPM (Desintegrador/Picador/Moedor). O objetivo deste trabalho consistiu então na avaliação comparativa do desempenho de cinco marcas comerciais deste tipo de equipamento. Avaliaram-se a demanda de potência, a capacidade de processamento, o consumo específico de energia, a distribuição granulométrica do produto moído pelo DPM e o nível de ruído no posto de trabalho. Durante os testes, também foram observados aspectos relacionados à segurança operacional, ergonomia, vazamentos e contaminação do ambiente com pó do produto em suspensão. Para avaliar o desempenho de cada moedor, o consumo específico foi calculado com cinco velocidades angulares distintas, e com quadro diferentes taxas de processamento. Foram determinados os consumos específicos das quatro peneiras (0,8; 3; 5; e 10 mm) que acompanham os equipamentos. Foram observados uma baixa capacidade de processamento e alto consumo específico de energia para peneira 0,8 mm, recomendada para a produção de fubá, sendo sugerida sua substituição pela de 3 mm, que apresenta distribuição granulométria semelhante à encontrada no fubá comercializado no mercado local. A máquina 4 apresentou o menor consumo específico de energia para peneira 3 mm e as peneiras de 5 e 10 mm não apresentaram diferenças significativas entre si. O nível de ruído medido durante os ensaios em todas as máquinas esteve bem acima do limite de 85 dB(A), recomendado pelo Ministério do Trabalho. As máquinas avaliadas não tinham dispositivos de proteção e paras as correias de acionamento e a bica de alimentação. / Most components of animal ration in a farm must be disintegrated, chopped or crushed, with adequate equipment, among them the ones designated DPM (disintegrator/chopper/grinder). The objective of this work was to compore the performance of five corn grinders. It was evaluate the power demand, processing capacity, specific energy consumption, granulometric distribution of the product output by the machine and noise at work place. During the tests, aspects related to operational safety, ergonomics, leaks and environmental contamination by the product dust in suspension were also observed. The specific consumption was measured in five different angular speeds and with four different processing rates. The specific consumption of four sieves (0.8; 3; 5 and 10 mm) were determined. It was observed a low processing capacity and a high specific consumption of energy for sieve 0.8 mm, recommended for corn flour production, suggesting its replacement to for the 3 mm sieve, which presents a similar granulometric distribution to the one found in the local market commercialized flour. Machine 4 presented the smallest energy specific consumption for sieve 3mm, and sieves 5 and 10 mm were not different to each other. The level of measured noise during the assays in all machines was well above the limit of 85 dB(A), recommended by the Ministrv of Labour. The tested machines did not present protection devices for the driving belts and for the feeding spout. / Dissertação importada do Alexandria

No description available.

No description available.

Peer-to-peer (P2P) technology has triggered a wide range of distributed systems beyond simple file-sharing. Distributed XML databases, distributed computing, server-less web publishing and networked resource/service sharing are only a few to name. Despite of the diversity in applications, these systems share a common problem regarding searching and discovery of information. This commonality stems from the transitory nodes population and volatile information content in the participating nodes. In such dynamic environment, users are not expected to have the exact information about the available objects in the system. Rather queries are based on partial information, which requires the search mechanism to be flexible. On the other hand, to scale with network size the search mechanism is required to be bandwidth efficient. Since the advent of P2P technology experts from industry and academia have proposed a number of search techniques - none of which is able to provide satisfactory solution to the conflicting requirements of search efficiency and flexibility. Structured search techniques, mostly Distributed Hash Table (DHT)-based, are bandwidth efficient while semi(un)-structured techniques are flexible. But, neither achieves both ends. This thesis defines the Distributed Pattern Matching (DPM) problem. The DPM problem is to discover a pattern (\ie bit-vector) using any subset of its 1-bits, under the assumption that the patterns are distributed across a large population of networked nodes. Search problem in many distributed systems can be reduced to the DPM problem. This thesis also presents two distinct search mechanisms, named Distributed Pattern Matching System (DPMS) and Plexus, for solving the DPM problem. DPMS is a semi-structured, hierarchical architecture aiming to discover a predefined number of matches by visiting a small number of nodes. Plexus, on the other hand, is a structured search mechanism based on the theory of Error Correcting Code (ECC). The design goal behind Plexus is to discover all the matches by visiting a reasonable number of nodes.

Peer-to-peer (P2P) technology has triggered a wide range of distributed systems beyond simple file-sharing. Distributed XML databases, distributed computing, server-less web publishing and networked resource/service sharing are only a few to name. Despite of the diversity in applications, these systems share a common problem regarding searching and discovery of information. This commonality stems from the transitory nodes population and volatile information content in the participating nodes. In such dynamic environment, users are not expected to have the exact information about the available objects in the system. Rather queries are based on partial information, which requires the search mechanism to be flexible. On the other hand, to scale with network size the search mechanism is required to be bandwidth efficient. Since the advent of P2P technology experts from industry and academia have proposed a number of search techniques - none of which is able to provide satisfactory solution to the conflicting requirements of search efficiency and flexibility. Structured search techniques, mostly Distributed Hash Table (DHT)-based, are bandwidth efficient while semi(un)-structured techniques are flexible. But, neither achieves both ends. This thesis defines the Distributed Pattern Matching (DPM) problem. The DPM problem is to discover a pattern (\ie bit-vector) using any subset of its 1-bits, under the assumption that the patterns are distributed across a large population of networked nodes. Search problem in many distributed systems can be reduced to the DPM problem. This thesis also presents two distinct search mechanisms, named Distributed Pattern Matching System (DPMS) and Plexus, for solving the DPM problem. DPMS is a semi-structured, hierarchical architecture aiming to discover a predefined number of matches by visiting a small number of nodes. Plexus, on the other hand, is a structured search mechanism based on the theory of Error Correcting Code (ECC). The design goal behind Plexus is to discover all the matches by visiting a reasonable number of nodes.

Purpose of my master’s thesis is an initial analysis of the current status of the selected company, which follows the basic theoretical background of the issue. Essential part of thesis is developing draft of backup and archive system in a small business environment, which will be applied in practice with respect to the maximum economic and qualitative aspects of the system.

Diplomová práce se zabývá prouděním v hydrocyklónu. Hydrocyklón je separační stroj, kterým slouží k oddělování pevných částice z kapaliny (nejčastěji z vody). Díky tangenciálnímu vstupu do zařízení nastává silné vířivé proudění, které způsobuje přisávání vzduchu přes horní a dolní výtok. Toto vzduchové jádro hraje důležitou roli v separaci částic. Výpočty jsou provedeny pomoci CFD.