The purpose of this paper is to describe the flow oscillations which occur in the AP600 long term cooling test facility at Oregon State University. The AP600 system is an advanced pressurized water reactor design utilizing passive emergency cooling systems. A few hours after the initiation of a cold leg break, the passive cooling systems inject gravity fed cold water at a rate allowing steam production in the reactor vessel. Steam production in the core causes the pressure in the upper head to increase leading to flow oscillations in all the connecting reactor systems. This paper will show that the oscillations have a definite region of onset and termination for specific conditions in the APEX testing facility. Tests performed at high powers, high elevation breaks, and small break sizes do not exhibit oscillations. The APOS (Advanced Plant Oscillation Simulator) computer code has been developed using a quasi-steady state analysis for flows and a transient analysis for the core node energy balance. The pressure in the reactor head is calculated using a modified perfect gas analysis. For tank liquid inventories, a simple conservation of mass analysis is used to estimate the tank elevations. Simulation logic gleaned from APEX data and photographic evidence have been incorporated into the code to predict termination of the oscillations. Areas which would make the work more complete include a better understanding of two-phase fluid behavior for a top offtake on a pipe, more instrumentation in the core region of the APEX testing facility, and a clearer understanding of fluid conditions in the reactor barrel. Scaling of the oscillations onset and pressure amplitude are relatively straight forward, but termination and period are difficult to scale to the full AP600 plant. Differences in the core power profile and other geometrical differences between the testing facility and the actual plant make the scaling of this phenomenon to the actual plant conditions very difficult. / Graduation date: 1997

Thesis (MScEng)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: Traditionally nuclear reactor power plants have been optimised for electrical power generation only. In the light of the ever-rising cost of dwindling fossil fuel resources as well the global polluting effects and consequences of their usage, the use of nuclear energy for process heating is becoming increasingly attractive. In this study the use of a so-called cogeneration plant in which a nuclear reactor energy source is optimised for the simultaneous production of superheated steam for electrical power generation and process heat is considered and analysed. The process heat superheated steam is generated in a once-through steam generator of heat pipe heat exchanger with intermediate fluid while steam for power generation is generated separately in a once-through helical coil steam generator. A 750 °C, 7 MPa helium cooled HTMR has been conceptually designed to simultaneously provide steam at 540 °C, 13.5 MPa for the power unit and steam at 430 °C, 4 MPa for a coal-to-liquid fuel process. The simulation and dynamic control of such a typical cogeneration plant is considered. In particular, a theoretical model of a typical plant will be simulated with the aim of predicting the transient and dynamic behaviour of the HTMR in order to provide guideline for the control of the plant under various operating conditions. It was found that the simulation model captured the behaviour of the plant reasonably well and it is recommended that it could be used in the detailed design of plant control strategies. It was also found that using a 1500 MW-thermal HTMR the South African contribution to global pollution can be reduced by 1.58%. / AFRIKAANSE OPSOMMING: Tradisioneel is kernkragaanlegte vir slegs elektriese kragopwekking geoptimeer. In die lig van die immer stygende koste van uitputbare fossielbrandstohulpbronne asook die besoedelingsimpak daarvan wêreldwyd, word die gebruik van kernkrag vir prosesverhitting al hoe meer aanlokliker. In hierdie studie word die gebruik van ‘n sogenaamde mede-opwekkingsaanleg waarin ‘n kernkragreaktor-energiebron vir die gelyktydige produksie van oorverhitte stoom vir elektriese kragopwekking en proseshitte oorweeg ontleed word. Die oorvehitte stoom word in ‘n enkeldeurvloei-stoomopwekking van die hittepyp-hitteruiler met tussenvloeistof opgewek en stoom vir kragopwekking word apart in ‘n enkeldeurvloei-spiraalspoel-stoomopwekker opgewek. ‘n 750 °C, 7 MPa heliumverkoelde HTMR is konseptueel ontwerp vir die gelytydige veskaffing van stoom by 540 °C, 13.5 MPa, vir die kragopwekkings eenheid, en stoom by 430 °C, 4 MPa, vir ‘n steenkool-tot-vloeibare (CTL) brandstoff proses. Die simulasie en dinamiese beheer van ‘n tipiese HTMR mede-opwekkingsaanleg word beskou. ‘n die besonder word ‘n teoretiese model van die transiënte en dinamiese gedrag van die aanleg gesimuleer om sodoene riglyne te identifiseer vir die ontwikkeling van dinamiese beheer strategië vir verskillende werkstoestande van die aanleg. Daar was ook gevind dat die simulasie model van die aanleg se gedrag goed nageboots word en dat dit dus gebruik kan word vir beheer strategie doeleindes. Indien so ‘n 1500 MW-termies HTMR gebruik word sal dit die Suid Afrikaanse besoedling met 1.58% sal kan verminder.

Orientador: Silvio Roberto Andrietta / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Quimica / Made available in DSpace on 2018-08-06T08:57:55Z (GMT). No. of bitstreams: 1 Porto_LiadeMendonca_D.pdf: 3965643 bytes, checksum: af08099e2fd964cfed91edc4842675e0 (MD5) Previous issue date: 2006 / Resumo: Este trabalho consiste em determinar o modelo que melhor descreve a fermentação alcoólica industrial, em sistemas contínuos de reatores de mistura em série. Para determinação dos parâmetros cinéticos foi desenvolvido um programa utilizando os dados de análises feitas em amostras retiradas de usinas em operação. Dos modelos testados, o Tosseto (2002), Lee; Pagan; Rogers (1983) e Levenspiel (1980), não apresentaram incoerência física nem problema de convergência sendo estes indicados para descrever a cinética da fermentação alcoólica. Os parâmetros cinéticos comuns aos modelos apresentaram valores semelhantes, onde o parâmetro Ks' limitação pelo substrato, foi de 3,0:t 0,4, o valor da concentração limite pelo produto inibidor, Pmáx' foi 92:t 9 e o fator exponencial de inibição pelo produto, YN, foi 5,3:t 0,9. O fator de inibição pelo substrato do modelo Tosseto (2002), Ki, foi 27:t 5 . O parâmetro X máx ' inibição pela massa celular, e YM, fator exponencial desta inibição, para o modelo Lee; Pagan; Rogers (1983) foram 1O0:t 1 e 0,9:t 0,1, respectivamente. A escolha pelo modelo Tosetto (2002) para implementação do programa de simulação do processo foi devido à tentativa de uma maior abrangência na obtenção das constantes cinéticas devido à possibilidade de inibição pelo substrato. A simulação em regime permanente trata-se da resolução equações algébricas dos balanços de massa do sistema e é capaz de dimensionar a etapa de fermentação de novas plantas a serem implantadas, enquanto que a simulação em regime transiente, tem o intuito de avaliar modificações durante a operação da usina e as equações diferenciais obtidas a partir dos balanços de massa devem ser resolvidas por método numérico, que neste caso foi utilizado o Runge-Kutta de quarta ordem / Abstract: This work aimed to determinate a model that best describes the industrial scale alcoholic fermentation in continuous serial mixing reactors. Software was developed in order to establish the kinetics parameters, using analysis data performed on samples from operating plants. The models described by Tosseto (2002), Lee; Pagan; Rogers (1983) and Levenspiel (1980), did not display physical incoherence or even convergence problems that described alcoholic fermentation kinetics. The kinetics parameters shown in all models presented similar values, like K (substrate limitation) was 3,0:t 0,4, Pmáx (limited product concentration) was 92:t 9 and YN (product inhibition power factor) was 5,3:t 0,9. The K (substrate inhibition) studied by Tosseto (2002) was 27:t 5. X máx (cell mass inhibition) and YM (cell mass inhibition factor power) were respectively 1O0:t 1 and 0,9:t 0,1 to Lee; Pagan; Rogers (1983). The kinetics parameters abranger due the substrate inhibition obtained by Tosetto (2002) this model was chosen in order to implement the simulating software. The stady state simulation is about system' s mass balance algebric equations resolution and is capable to dimension the fermentation step in new plants to be implanted while the unstady state simulation have the intention to evaluate modifications during the plant operation and the differential equations obtained from mass balance must be resolved by numerical methods, in this case, was used the fouth order Runge-Kutta / Doutorado / Desenvolvimento de Processos Biotecnologicos / Doutor em Engenharia Química