Submitted by Fabio Sobreira Campos da Costa (fabio.sobreira@ufpe.br) on 2016-07-04T11:51:54Z No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Dissertação Mestrado Ráurium versão final.pdf: 6757441 bytes, checksum: b3d2956882bd467a5f2fcb592d7f079c (MD5) / Made available in DSpace on 2016-07-04T11:51:54Z (GMT). No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Dissertação Mestrado Ráurium versão final.pdf: 6757441 bytes, checksum: b3d2956882bd467a5f2fcb592d7f079c (MD5) Previous issue date: 2015-10-26 / A costa litorânea do Estado de Pernambuco no Nordeste do Brasil é dotada por várias bacias hidrográficas onde pode-se destacar a bacia do rio Pirapama, que abrange sete municípios e é constituída dos principais mananciais de abastecimento da Região Metropolitana do Recife. O presente estudo apresenta a simulação da disponibilidade hídrica da bacia do Pirapama, e propõe regras de operação para os reservatórios de abastecimento hídrico inclusos na bacia, levando-se em conta a variabilidade climática com ênfase a evitar as condições de racionamento e colapso do reservatório Pirapama frente aos eventos de seca. A partir do modelo hidrológico SMAP foi gerada a série de vazões em 7 sub-bacias da bacia do rio Pirapama abrangendo o período de 1933 a 2014. Essa série de dados foi inserida no modelo de simulação AcquaNet que permitiu elaborar a rede de fluxo representativa da bacia e simular a disponibilidade hídrica nas suas sub-bacias e reservatórios estratégicos. Embasadas em metodologias distintas, foram apresentadas duas propostas de regras de operação para o reservatório Pirapama e uma para o reservatório Sicupema. A primeira proposta, as curvas-guia, consistiu no zoneamento do volume útil do reservatório de acordo com o estado de severidade de seca, sendo os eventos de seca quantificados através do índice de escoamento padronizado SRI (Standardized Runoff Index). A segunda proposta foi a simulação das Curvas de Aversão a Risco – CAR através do AcquaNet, que possibilitou analisar o comportamento do reservatório para várias vazões de retiradas, tendo por base as afluências no biênio mais critico da série simulada. Para o reservatório Sicupema, foram realizadas simulações no AcquaNet, estabelecendo regras de operação através do volume meta. Os resultados mostraram um ajuste satisfatório na calibração e validação do modelo hidrológico, assim como, na análise comparativa dos volumes simulados e monitorados do reservatório Pirapama. O comportamento dos 82 anos simulados no reservatório Pirapama demonstrou a necessidade de se alterar as regras de operação atuais, pois o mesmo esteve por 5 anos e meio nas condições de racionamento ou colapso. As regras de operação propostas, quando aplicadas ao reservatório Pirapama, se mostraram eficientes e eficazes, eliminando o risco de colapso, e reduzindo a condição de racionamento de 39 para 4 e 7 meses, nas curvas-guia e CAR, respectivamente, constituindo numa importante ferramenta de tomada de decisão na gestão do maior sistema de abastecimento do Estado de Pernambuco. / The coastline of the state of Pernambuco in northeastern Brazil is endowed by various river basins where we can highlight the basin of the Pirapama river, covering seven cities and is responsible for most of the supply of the Metropolitan Region of Recife.This study presents the simulation of water availability of Pirapama River basin, located in the coastline of the state of Pernambuco in northeastern Brazil, and proposes operating rules for water supply reservoirs included in the basin, taking into account the variability climate with emphasis on avoiding rationing conditions and collapse of Pirapama reservoir forward to drought events. From the hydrological model was generated series of discharges in 7 sub-basins of Pirapama river basin in period from 1933 to 2014.This data set was inserted into the simulation model AcquaNet which established the network of representative outflow of the basin and simulate water availability in their sub-basins and strategic reservoirs. Based on different methodologies were presented two proposals for operating rules for Pirapama reservoir and one for Sicupema dam. The first proposal, the rules curves, consisted of the zoning of the useful volume of the reservoir according to drought severity state, and drought events quantified by Standardized Runoff Index- SRI. The second proposal was the simulation of Aversion to Risk Curves - CAR through AcquaNet, which made it possible to analyze the reservoir behavior for several withdrawals flows, based on two years of critical inflows simulated series. For Sicupema reservoir simulations were performed in AcquaNet, establishing operating rules by volume goal. The results showed a satisfactory adjustment in the calibration and validation of hydrological model, as well as the comparative analysis of the simulated volumes and monitored the Pirapama reservoir. The behavior of the simulated 82 years in Pirapama reservoir demonstrated the need to change the current operating rules, because it was for 5 years and a half in the rationing conditions or collapse. The rules of operation proposals when applied to Pirapama reservoir, were efficient and effective, eliminating the risk of collapse, and reducing 39 rationing condition for 4 and 7 months, the guide curves and CAR, respectively, constituting an important decision-making tool in managing the largest supply system of the State of Pernambuco.

The interior of heavy goods vehicles (HGVs) differs from passenger cars. Both the steering wheel and the occupant are positioned differently in a HGV and increases the risk of steering wheel rim impacts. Such impact scenarios are relatively unexplored compared to passenger car safety studies that are more prevalent within the field of injury biomechanics. The idea with using human body models (HBMs) is to complement current crash test dummies with biomechanical data. Furthermore, the biofidelity of a crash dummy for loading similar to a steering wheel rimimpact is relatively unstudied and especially to different rib levels. Therefore, the aim with this thesis was to evaluate HGV occupant thoracic response between THUMS v4.0 and Hybrid III (H3) during steering wheel rim impacts with respect to different rib levels (level 1-2, 3-4, 6-7, 7-8, 9-10) with regards to ribs, aorta, liver, and spleen. To the author’s best knowledge, use of local injury risk functions for thoracic injuries is fairly rare compared to the predominant usage of global injury criteria that mainly predicts the most commonthoracic injury risk, i.e. rib fractures. Therefore, local injury criteria using experimental test datahave been developed for the ribs and the organs. The measured parameters were chest deflectionand steering wheel to thorax contact force on a global level, whilst 1st principal Green-Lagrangestrains was assessed for the rib and the organ injury risk. The material models for the liver and the spleen were remodelled using an Ogden material model based on experimental stress-strain data to account for hyperelasticity. Rate-dependency was included by iteration of viscoelastic parameters. The contact modelling of the organs was changed from a sliding contact to a tied contact to minimize unrealistic contact separations during impact. The results support previous findings that H3 needs additional instrumentation to accurately register chest deflection for rib levels beyond its current range, namely at ribs 1-2, 7-8, and 9-10. For THUMS, the chest deflection were within reasonable values for the applied velocities, but there were no definite injury risk. Fact is, the global injury criteria might overpredict the AIS3 injury risk (rib fractures) for rib level 1-2, 7-8, and 9-10. The rib strains could not be correlated with the measured chest deflections. This was explained by the unique localized loading characterized by pure steering wheel rim impact that mainly affected the sternum and the rib cartilage while minimizing rib deformation. The organ strains indicate some risk of rupture where the spleen deforms the most at rib levels 3-4 and 6-7, and the liver and the aorta at rib levels 6-7 and 7-8. This study provides a framework for complementing H3 with THUMS for HGV occupant safety with emphasis on the importance of using local injury criteria for functional injury prediction, i.e. prediction of injury risk using parameters directly related to rib fracture or organ rupture. Local injury criteria are thus a powerful safety assessment tool as it is independent on exterior loading such as airbag, steering wheel hub, or seat belt loading. It was noticed that global injury criteria with very localized impacts such as rim impacts have not been studied and will affect rib fracture risk differently than what has been studied using airbag or seat belt restraints. However, improvements are needed to accurately predict thoracic injury risk at a material level by finding more data for the local injury risk functions. Conclusively, it is clear that Hybrid III has insufficient instrumentation and is in need of upgrades to register chest deflections at multiple rib levels. Furthermore, the following are needed: better understanding of global injury criteria specific for HGV occupant safety evaluation, more data for age-dependent (ribs) and rate-dependent (organs) injury risk functions, a tiebreak contact with tangential sliding for better organ kinematics during impacts, and improving the biofidelity of the material models using data from tissue level experiments. / Förarmiljön i lastbilar gentemot personbilar är annorlunda, i detta kontext med avseende på främst ratt- och förarposition som ökar risken för islag med rattkransen för lastbilsförare. Sådana islag är relativt outforskat jämfört med passiv säkerhet för personbilar inom skadebiomekaniken. Tanken bakom användning av humanmodeller är att komplettera nuvarande krockdockor med biomekanisk information. Dessutom är biofideliteten hos en krockdocka vid rattislag relativt okänt, speciellt vid olika revbensnivåer. Därför är målet med detta examensarbete att undersöka thoraxresponsen hos en lastbilsförare genom att använda THUMS v4.0 och Hybrid III (H3) under rattislag med avseende på revbensnivåer (nivå 1-2, 3-4, 6-7, 7-8, och 9-10) och revben, aorta, lever, och mjälte. Enligt författaren verkar användning av lokala riskfunktioner för thoraxskador relativt ostuderat jämfört med den övervägande användningen av globala riskfunktioner som huvudsakligen förutser den mest vanligt förekommande thoraxskadan, nämligen revbensfrakturer. Därför har lokala riskfunktioner skapats för revben och organ, baserat på experimentell data. Uppmätta parametrar var bröstinträngning och kontaktkraft mellan ratt och thorax på global nivå, medan första Green-Lagrange huvudtöjningen användes för att evaluera skaderisken för revben och organ. Materialmodeller för lever och mjälte ommodellerades baserat på experimentell spänning-töjningsdata med Ogdens materialmodell för att ta hänsyn till hyperelasticitet. Töjningshastighetsberoendet inkluderades genom att iterera fram viskoelastiska parametrar. Kontaktmodellering av organ gjordes genom att ändra från glidande kontakt till en låsande kontakt för att minimera orealistisk kontaktseparation under islagsfallen. Resultaten stödjer tidigare studier där H3 visat sig behöva ytterligare givare för att noggrannt kunna registrera bröstinträngning vid olika revbensnivåer bortom dess nuvarande räckvidd, nämligen vid revben 1-2, 7-8, och 9-10. Uppmätt bröstinträngning i THUMS var rimliga för hastighetsfallen men gav inte någon definitiv risk för skada. Faktum är att de globala riskfunktionerna kan överskatta AIS3 risken vid revben 1-2, 7-8, och 9-10. Revbenstöjningarna kunde inte korreleras med bröstinträngningarna. Detta kunde förklaras genom de unika lastfallen som karakteriseras av rena rattislag som främst påverkar sternum och revbensbrosk som i sin tur minimerar deformation av revben. Organtöjningarna indikerar på någon risk för ruptur där mjälten deformerar som mest vid revben 3-4 och 6-7, medan för både levern och aortan sker det vid revben 6-7 och 7-8. Denna studie presenterar ett sätt att komplettera H3 med THUMS inom passiv säkerhet för lastbilsförare med fokus på lokala riskfunktioner för funktionell skadeprediktering dvs. prediktering av skaderisken med hjälp av parametrar som är direkt relaterat till revbensfraktur eller organruptur. Lokala riskfunktioner utgör en kraftfull säkerhetsbedömning som är oberoende av externa lastfall som t.ex. airbag, rattcentrum, eller bälteslast. I denna studie noterades det att de globala riskkriterierna inte har undersökts med väldigt lokala islag som rattislagen utgör och kommer därför att påverka risken för revbensfraktur annorlunda gentemot vad som har studerat, t.ex. airbag eller bältelast. Däremot behövs det mer data för de lokala riskkriterierna för att kunna prediktera thoraxskaderisken med ökad noggrannhet. Avslutningsvis, det är tydligt att Hybrid III har otillräckligt med givare och behöver förbättras för att kunna registrera bröstinträngning vid flera revbensnivåer. Vidare behövs följande: bättre förståelse för globala riskfunktioner anpassat inom passiv säkerhet för lastbilsförare, mer data för åldersberoende (revben) och töjningshastighetsberoende (organ) riskfunktioner, en ”tiebreak” kontakt med tangientiell glidning för bättre organkinematik, och ökad biofidelitet av materialmodeller genom att använda data från vävnadsexperiment.