No description available.

No description available.

Threshold leak sizes and leak rates were calculated for a number of liquid food products exhibiting a wide range of surface tension and viscosity values. From this data, one can see that mathematically, under typical pressure differentials generated in food packages (less than or equal to ±34.5 kPa), a leak will never start through a 2 μm defect. The calculated leak rates were compared to calculated evaporation rates. The evaporation rate exceeds the leak rate at lower sized microholes (2, and 5 μm diameter) under typical pressure differentials found in food packages. If the liquid, typically aqueous in food products, is evaporating off faster than the leak itself, then there will be solids left behind that could effectively plug the leak. The critical leak size is the size micro-defect that allows microbial penetration into the package. The critical leak size of air-filled defects was found to be 7 μm at all pressures tested. This size is considerably important to food packagers because this is when sterility of the package is lost. Previous leak studies have shown that the critical leak size for liquid-filled defects coincide with the threshold leak size and pressure. If this is in fact true, then air-filled defects should exhibit a larger critical leak size than the liquid-filled defects. In this study, air-filled defects were examined. A bioaerosol exposure chamber was used to test micro-defects, nickel microtubes of known diameters 2, 5, 7, 10, 20, and 50 μm hydraulic diameters, against pressure differentials of 0, -6.9, -13.8, and -34.5 kPa. / Master of Science

Large graph-based datasets are common to many applications because of the additional structure provided to data by graphs. Patterns extracted from graphs must adhere to these structural properties, making them a more complex class of patterns to identify. The role of graph mining is to efficiently extract these patterns and quantify their significance. In this thesis, we focus on two application domains and demonstrate the design of graph mining algorithms in these domains. First, we investigate the use of graph grammar mining as a tool for diagnosing potential memory leaks from Java heap dumps. Memory leaks occur when memory that is no longer in use fails to be reclaimed, resulting in significant slowdowns, exhaustion of available storage, and eventually application crashes. Analyzing the heap dump of a program is a common strategy used in memory leak diagnosis, but our work is the first to employ a graph mining approach to the problem. Memory leaks accumulate in the heap as classes of subgraphs and the allocation paths from which they emanate can be explored to contextualize the leak source. We show that it suffices to mine the dominator tree of the heap dump, which is significantly smaller than the underlying graph. We demonstrate several synthetic as well as real-world examples of heap dumps for which our approach provides more insight into the problem than state-of-the-art tools such as Eclipse's MAT. Second, we study the problem of multipartite graph clustering as an approach to database summarization on an integrated biological database. Construction of such databases has become a common theme in biological research, where heterogeneous data is consolidated into a single, centralized repository that provides a structured forum for data analysis. We present an efficient approximation algorithm for identifying clusters that form multipartite cliques spanning multiple database tables. We show that our algorithm computes a lossless compression of the database by summarizing it into a reduced set of biologically meaningful clusters. Our algorithm is applied to data from C. elegans, but we note its applicability to general relational databases. / Master of Science

[pt] Empresas que operam dutos possuem como uma de suas premissas básicas a segurança operacional. Dentro desse quesito, alguns fatores podem levar a acidentes com possibilidade de danos materiais, ambientais e pessoais, tais como: corrosão interna e externa, escavações acidentais, operações indevidas levando o duto a pressões excessivas, assim como ações de terceiros cujo objetivo seria o furto de produtos refinados, gasolina, diesel, álcool, entre outros. Esta dissertação está alinhada com necessidades reais de empresas relacionadas ao transporte de produtos líquidos por dutos, desenvolvendo sistemas que possam vir a ser utilizados nestas empresas. Foram desenvolvidos sistemas de detecção e localização de vazamentos a serem utilizados em um centro de controle operacional de oleodutos, investigando métodos não convencionais, além dos mencionados pela norma API 1130. Foram desenvolvidos sistemas de detecção de vazamento por balanço de massa, balanço de volume, lógica fuzzy e redes neurais artificiais. Para localização de vazamentos, foram testados os métodos por redes neurais, pela estimativa da velocidade sônica e do gradiente hidráulico. Os produtos utilizados foram gasolina, diesel e óleo combustível. No quesito detecção de vazamento, o sistema baseado em redes neurais detectou vazamentos simulados, porém também indicou indevidamente. O sistema baseado em lógica fuzzy apresentou bons resultados, indicando vazamentos corretamente sem falsos positivos, interpretando corretamente os fenômenos inerentes à operação de dutos. O sistema de balanço de massa também apresentou bons resultados, ou seja, não gerou alarmes falsos, detectando corretamente os vazamentos simulados, inclusive para detecção de vazamento com duto parado e pressurizado. Para avaliar sistemas de detecção de vazamento é comum realizar testes de campo que podem ter um custo alto e levar muito tempo para realizar. Um método para realizar testes a um custo inferior deve ser desenvolvido e uma proposta está sendo mostrada nesta dissertação. / [en] Companies in the business of pipeline operations have as a basic assumption, operational security. Concerning this issue, there are some factors that could lead to accidents with material, environment and personal damage possibilities such as: internal and external corrosion, accidental excavations, improper operation that could submit the pipe to high pressures and third party interventions interested in commodities theft. This dissertation is aligned with pipeline companies real demands through the development of systems that could be used by these companies. Leak detection systems provided with leak location capabilities were developed to be used in liquid pipeline control centers, exploring non-conventional methods, besides the ones mentioned by API 1130. The leak detection systems developed were: mass balance, volume balance, fuzzy logic and neural networks. For the leak location systems the systems tested were: sonic velocity, hydraulic gradient and artificial neural networks. The products used were gasoline, diesel and fuel oil. On the issue of leak detection, the system based on neural networks detected simulated leakages, although there were some false indications. The system based on fuzzy logic presented good results, giving correct leak indications without any indication of false alarms, interpreting correctly the phenomena due to pipeline usual operations. The mass balance system has also presented good results, not generating false alarms, but detecting simulated leaks even with the pipeline in shut-in condition. To evaluate leak detection systems is common to conduct field tests that can be costly and take a long time to accomplish. A method for testing at a lower cost should be developed and a proposal is being shown in this work.

One of the best opportunities to reduce pipeline accidents and subsequent product loss comes from implementing better pipeline condition assessment and fault detection systems. Transient analysis model based condition assessment is the most promising technique because pressure transients propagate entire system interacting with the pipe and any devices in the system. Transient measurements embody a large amount of information about the physical characteristics of the system. The performance of this technique has its difficulties because a highly accurate transient model is required. Real systems have numerous uncertainties and flow system components that presents a major challenge in the development of precise transient analysis models. To improve transient modelling for the performance of condition assessment, this research undertakes a comprehensive investigation into the transient behaviour of distributed and various local energy loss system components in water and gas pipelines. The dynamic behaviours that have been investigated in this research are the effect of unsteady wall resistance, viscoelasticity effects of polymer pipe, and local energy loss elements including leakages, entrapped air pockets, orifices, and blockages during unsteady pipe flow conditions. The dynamic characteristics of these system components are modelled based on the conservative solution scheme using the governing equations in their conservative form. Use of the conservative form of the equations improves the sensitivity and applicability of transient analysis in both liquid and gas pipeline systems. The numerical model results are compared to laboratory experiments in water and gas pipelines to observe the interaction between transient pressure wave and system components and to verify the proposed models. / http://proxy.library.adelaide.edu.au/login?url= http://library.adelaide.edu.au/cgi-bin/Pwebrecon.cgi?BBID=1337145 / Thesis( Ph.D.) -- University of Adelaide, School of Civil, Environmental and Mining Engineering 2008

Submitted by Claudinei Pracidelli (cpracide@ipen.br) on 2016-07-01T11:08:37Z No. of bitstreams: 0 / Made available in DSpace on 2016-07-01T11:08:37Z (GMT). No. of bitstreams: 0 / Tese (Doutorado em Tecnologia Nuclear) / IPEN/T / Instituto de Pesquisas Energeticas e Nucleares - IPEN-CNEN/SP

Submitted by Claudinei Pracidelli (cpracide@ipen.br) on 2016-07-01T11:08:37Z No. of bitstreams: 0 / Made available in DSpace on 2016-07-01T11:08:37Z (GMT). No. of bitstreams: 0 / Tese (Doutorado em Tecnologia Nuclear) / IPEN/T / Instituto de Pesquisas Energeticas e Nucleares - IPEN-CNEN/SP

Orientadores: Sandra Lúcia da Cruz, Ana Maria Frattini Fileti / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Química / Made available in DSpace on 2018-08-20T02:02:28Z (GMT). No. of bitstreams: 1 Sousa_ElisangelaOrlandide_D.pdf: 7481778 bytes, checksum: c6a75eef8a6e332ab1da367733280fd8 (MD5) Previous issue date: 2012 / Resumo: Redes de tubulação são conhecidas como sistemas complexos de dutos deste a antiguidade e são utilizadas no transporte de líquidos e gases a longas distâncias. Um pequeno vazamento em uma tubulação pode provocar grandes perdas de produtos e sérios danos ao meio ambiente até serem detectados. Com o propósito de rastrear estes tipos de vazamentos, este trabalho tem como objetivo propor metodologias para a detecção de vazamentos em uma tubulação, de natureza rígida ou flexível, baseando-se no método acústico e na análise de transientes de pressão gerados a partir do vazamento, visando à localização e a determinação da magnitude dos vazamentos através da utilização de redes neurais artificiais. As metodologias propostas se destacam por não apresentarem impactos prejudiciais ao meio ambiente. Variações de transientes de pressão e o ruído sonoro gerados pela ocorrência de vazamentos foram detectados e analisados a partir de experimentos realizados em uma tubulação de ferro galvanizado de 60 m e uma tubulação flexível de 100 m de comprimento operando com escoamento contínuo de gás (ar), em várias condições de operação. O vazamento foi provocado em mais que uma posição ao longo das duas tubulações e utilizaram-se orifícios de diâmetros diferentes. Os transientes de pressão e os ruídos sonoros foram captados por um transdutor de pressão e um microfone, respectivamente, ambos instalados em um vaso de pressão, localizado no início da tubulação e acoplados a uma placa de aquisição de dados em um microcomputador. O sinal gerado pelo microfone foi amplificado e passou por um banco de filtros passa faixa sendo transformado em três sinais com amplitudes independentes, cada uma com uma faixa de freqüência específica de 1kHz, 5kHz e 9kHz. O programa de aquisição de dados foi escrito em linguagem C para ler e processar os dados. Os dados resultantes dos experimentos mostraram que foi possível detectar vazamentos, para todos os orifícios utilizados, baseado nos transientes de pressão e no método acústico. A dinâmica desses dados foi utilizada como entrada para o modelo neural para localizar e determinar a magnitude dos vazamentos, simultaneamente. O método de Levenberg-Marquardt com Regularização Bayesiana foi utilizado no treinamento dos modelos neurais. Os resultados apresentados pelo modelo neural desenvolvido indicaram com sucesso ao mesmo tempo a localização e a magnitude dos vazamentos / Abstract: Pipeline networks are complex systems of ducts used nowadays for gas and chemical products transporting through long distances. They frequently cross highly populated regions, water supplies or natural reserves. Even small leaks in pipelines can lead to great losses of products and serious damages to the environment before it could be detected. With the purpose to track these leaks, this work developed a technique to detection of leaks in pipelines, of rigid or flexible nature, based on acoustic method and on analysis of pressure transients generated by leak occurrence, in order to localization and determination the magnitude of leaks by using neural artificial networks. The methodologies proposed are notated for not having impacts on the environment. Variations of pressure transients and the noise generated by leakage will be detected and analyzed in a 60m-galvanized iron pipe and in a 100m-flexible pipe operating with continuous flow of gas (air) under various operating conditions. Leakages were provoked in many positions along the two pipes and used hole of distinct magnitudes. The pressure transients and the audible noise was captured by the pressure transducer and the microphone, respective, both installed inside the pressure vessel connected to a data acquisition system at a computer. The signal generated by the microphone was amplified and also passed through a bank of band pass filters being transformed into three signals with independent amplitude, each one with a band of specific frequency of 1 kHz, 5 kHz and 9 kHz. The data acquisition software was written in C language to read and process all data. The experimental results showed that it is possible to detect leaks in pipelines, for all holes, based on a pressure transient and on acoustic methods. The dynamics of these data in time is used as input to the neural model to location and determine of the leaks magnitude, simultaneously. The method chosen for training the neural networks was the Levenberg-Marquardt with Bayesian Regularization. The results of neural models indicated successfully in the same time the location and the magnitude of the leaks / Doutorado / Sistemas de Processos Quimicos e Informatica / Doutor em Engenharia Química

No description available.