1 |
Influences of Sea-land Breezes and Northeastern Monsoon on the Transportation and Dispersion of Air Pollutants over Coastal Region in Southern TaiwanTsai, Hsieh-Hung 11 August 2010 (has links)
This study investigated the influences of sea-land breezes (SLBs) and northeastern monsoon (NEM) on the transportation and dispersion of air pollutants over coastal region in southern Taiwan. The physicochemical properties of particulate matter (PM) was simultaneously sampled and analyzed at both inland and offshore sites during eight intensive sampling periods. This study further used a SURFER 2-D plotting software, a backward trajectory model, a 3-D meteorological model (MM5), and a comprehensive air quality model (CAMx) to simulate surface wind fields and spatial distribution of air pollutants over the coastal region during the intensive sampling periods of SLBs and NEM.
According to the meteorological condition and the synoptic weather patterns of the observation data showed that the SLBs sampling periods commonly occurred the weather patterns were zone of low pressure, pacific high pressure, and west stretch of the pacific high pressure when the main prevailing wind direction were west wind and southwest wind. During the NEM sampling periods, the weather patterns were strong northeastern monsoon, standard northeastern monsoon, and outflow rebound with high pressure in southern Taiwan, and then the main prevailing wind direction were northwest wind and northeast wind. However, during the MIX sampling periods, the weather patterns were outflow circulation of typhoon, weak northeastern monsoon, and outflow rebound with high pressure, while the wind directions didn¡¦t change regular. Thus, at coastal sites, the sea-land breezes induce an inland transport of air pollutants during the daytime and a seaward return of air pollutants at nighttime, causing a recirculation of air pollutants back to inland regions each day during the SLBs sampling periods. During the NEM sampling period was mainly brought from the northeastern wind which transported air pollutants from the northern region to Kaohsiung metropolitan area.
The results of PM concentration and size distribution indicated that the inland sites had a higher fraction of fine particles (PM2.5), whereas the offshore sites had a higher fraction of coarse particles (PM2.5-10). These phenomena were attributed to the fact that marine aerosols are generally abundant in the coarse particles. PM concentration is relatively higher during the NEM sampling periods than during the SLBs and MIX sampling periods. For PM concentration, the order of secondary inorganic aerosols (SIA) was NEM > MIX > SLBs, while the SIA/Ions ratio of PM2.5 were approximately 50% during sampling periods. The [NO3-]/[SO42-] ratios of PM2.5 and PM2.5-10 during the SLBs sampling periods were always lowest than those during the NEM and MIX sampling periods. It is suggested that the PM concentrations during the SLBs sampling periods were highly influenced by stationary sources emissions. The crustal elements indicated that the Al, Ca, Fe, and K contributed major composition of particles. Artificial metals, such as Mg, Pb, V, and Zn were also enriched in the atmospheric PM during the NEM sampling periods. In addition, the higher concentration of Fe was attributed to local anthropogenic emission and weak northeastern monsoon during the MIX sampling periods. Regardless of inland or offshore sites, a high concentration of secondary organic carbon (SOC) during the NEM and MIX sampling periods was consistent with OC/EC ratio higher than 2.2 indicates the potential formation of secondary aerosols. Chloride deficit of PM at inland sites were lost easily for approximately 40.28% during the NEM sampling periods. Moreover, the lowest [Cl-]/[Na+] ratio occurred during the sampling periods when the chloride deficit was relative high at inland sites. According to the results of neutralization ratio (NR), regardless of the periods (SLBs, NEM, and MIX), the particulates of inland and offshore sites were both acid. During the NEM and MIX sampling periods, sulphur oxidation ratio (SOR) of PM2.5 over coastal region in southern Taiwan were above 0.25. It is suggested that the results of SOR during those sampling periods were highly influenced by long transportation. In addition, the nitrogen oxidation ratio was lower influenced than SOR over coastal region in southern Taiwan. It is suggested that air quality of Kaohsiung metropolitan area were influenced by the industrial source emissions.
During the SLBs sampling period, sea breezes blown in the morning transported the offshore PM10 back to the inland sites in the Kaohsiung metropolitan area. In contrast, the air mass observed during the NEM sampling period was brought to the Kaohsiung metropolitan area mainly by a northerly wind which transported air mass originating in the northern region (i.e. Chiayi counties). The backward trajectory modeling of the MIX sampling periods suggested that the winds were dominated by the Northeastern Monsoon, and as such, likely inhibited the influences of sea-land breezes. Local surface air mass circulation over southern Taiwan obtained from MM5 model influenced by the southwestern monsoon during the SLBs sampling period. The weather in southern Taiwan during the NEM intensive sampling period was occasionally influenced by the Northeastern Monsoon as well as by complex terrain. During the MIX sampling periods, air mass recirculation was frequently observed in the coastal region of southern Taiwan. The results of CAMx model showed that air pollutions were occurred apparent sea breezes in the afternoon and land breezes at night during the SLBs sampling period. During the NEM sampling period, a strong Northeastern Monsoon blew air pollutions from the north to the southern region. We have compared the model simulation with field measured O3 and PM10 concentrations for inland and offshore regions. The order of correlation coefficients of ¬the model simulation and the measurement for O3¬ and PM10 in sampling periods is SLBs>MIX>NEM and NEM>MIX>SLBs, respectively. Overall, the correlation coefficients of the model simulation and the measurement were middle and high correlation.
|
2 |
A Numerical Study of Melt Pool Heat Transfer in the IVR of a PWR / En numerisk studie av smältpoolvärmeöverföring i IVR för en PWRZhao, Yuer January 2021 (has links)
This thesis aims to provide the thermal condition of melt pool convection by CFD simulation, which is important to the assessment of the invessel melt retention (IVR) strategy widely adopted in Generation III pressurized water reactors (PWRs). As a severe accident mitigation measure, the IVR strategy is realized through external cooling of the lower head of a reactor pressure vessel (RPV). To achieve the coolability and retention of the corium pool in the RPV lower head, the heat flux at the outer surface of the vessel should be less than the critical heat flux (CHF) of boiling around the lower head. Under such condition, the integrity of the RPV is guaranteed by the adequate thickness of the unmelted vessel wall. The thesis work starts from the selection and validation of a turbulence model in the CFD computational tool chosen (Fluent). Afterwards a numerical model is set up for estimation of melt pool heat transfer of a reference PWR with the power capacity of 1000 MWe, including a mesh sensitivity study. Based on the numerical model of a twolayer melt pool, four tasks are carried out to investigate the effects of Zr oxidation ratio, Fe content, and radiation emissivity on heat flux profiles, as well as the focus effect under extreme conditions. Selection and validation of the turbulence model are conducted by comparing the simulation results of different turbulence models with the DNS data on the convection of volumetrically heated fluid layer bounded by rigid isothermal horizontal walls at equal temperature. The internal Rayleigh numbers of the flow reach up to 10e6. The comparison shows a good agreement of the SST k-ω turbulence model results with the DNS data. The simulations with the Zr oxidation ratio of 0, 0.2 and 0.5, correspondingly, the oxide layer of 1.389m, 1.467m and 1.580m, and the metal layer of 0.705m, 0.646m and 0.561m in height, show that, the temperature of the oxide layer will increase with Zr oxidation ratio, while the temperature of the metal layer will decrease resulting in more heat transfer through the oxide layer sidewall and less top radiation. Nevertheless, the effect of the Zr oxidation ratio is not pronounced in the range of 00.5. The simulations with the Fe mass of 22t, 33t and 45t, and respective height of the metal layer of 0.462m, 0.568m and 0.646m, show that, the inner metal layer will significantly increase the temperatures of both the metal layer and the oxide layer. The percentage of heat transfer at the oxide layer sidewall will increase to supplement the reduction of that at the metal layer. The simulations with the radiation emissivity of 0.2, 0.35, 0.45 and 0.7 show that, the emissivity below 0.45 has an impact on heat transfer, and the temperatures and sidewall heat flux of both the oxide layer and the metal layer will increase with decreasing emissivity. The impact is negligible when the emissivity is above 0.45. The simulations under the hypothetically extreme conditions with either an adiabatic top boundary or a very thin metal layer show the focusing effect may occur, i.e., the heat flux through the metal sidewall is larger than that in the oxide layer. But the local high heat flux is flattened by the vessel wall with good heat conductivity. In summary, the simulations demonstrate that, except for the cases under extreme conditions, the heat fluxes of the melt pools in all other cases are significantly lower than the CHF of external cooling of the lower head. Therefore, the safety margin of the IVR strategy of the PWR chosen is seems sufficient. However, due to some limitations (e.g., simplification and assumptions) in the simulation cases and coupling of different influential factors, as indicated by the present study, the precise predictions of heat flux under all scenarios are still difficult. Therefore, the conclusions could not be generalized to the other conditions or other configurations of the molten pools. By discussing the model and simplifications/assumptions adopted in this work, the improvement directions of the numerical model and other perspectives are proposed at the end of the thesis. / Denna avhandling syftar till att tillhandahålla det termiska tillståndet för smältbassängskonvektion genom CFD-simulering, vilket är viktigt för bedömningen av IVR-strategin som allmänt antagits i tryckvattenreaktorer (PWR) i Generation III. Som en åtgärd för att mildra allvarliga olyckor realiseras IVR-strategin genom extern kylning av det nedre huvudet av ett reaktortryckkärl (RPV). För att uppnå kylbarhet och kvarhållning av koriumbassängen i det nedre RPV-huvudet bör värmeflöde vid den yttre ytan av kärlet vara mindre än det kritiska värmeflödet (CHF) som kokar runt det nedre huvudet. Under sådant tillstånd garanteras RPV: s integritet av den osmälta kärlväggens tillräckliga tjocklek. Examensarbetet startar från valet och valideringen av en turbulensmodell i det valda CFD-beräkningsverktyget (Fluent). Därefter sätts en numerisk modell upp för uppskattning av smältbassängens värmeöverföring av en referens PWR med en effektkapacitet på 1000 MWe, inklusive en nätkänslighetsstudie. Baserat på den numeriska modellen för en tvålagers smältbassäng utförs fyra uppgifter för att undersöka effekterna av Zr-oxidationsförhållande, Fe-innehåll och strålningsemissivitet på värmeflödesprofiler, liksom fokuseffekten under extrema förhållanden. Val och validering av turbulensmodellen utförs genom att jämföra simuleringsresultaten för olika turbulensmodeller med DNS-data för konvektionen av volymetriskt uppvärmt fluidskikt avgränsat av styva isoterma horisontella väggar vid lika temperatur. De interna Rayleigh-siffrorna i flödet når upp till 10e6. Jämförelsen visar att SST k-ω turbulensmodellresultaten överensstämmer med DNS-data. Simuleringarna med Zr-oxidationsförhållandet 0, 0,2 och 0,5, motsvarande oxidskiktet på 1,389 m, 1,467 m och 1,580 m, och metallskiktet på 0,705 m, 0,664 m och 0,561 m i höjd, visar att temperaturen av oxidskiktet kommer att öka med Zr-oxidationsförhållandet, medan metallskiktets temperatur kommer att minska vilket resulterar i mer värmeöverföring genom oxidskiktets sidovägg och mindre toppstrålning. Ändå är effekten av Zr-oxidationsförhållandet inte uttalad i intervallet 00,5. Simuleringarna med Fe-massan på 22t, 33t och 45t och respektive höjd av metallskiktet på 0,462m, 0,568m och 0,664m visar att det inre metallskiktet avsevärt kommer att öka temperaturerna för både metallskiktet och oxiden lager. Andelen värmeöverföring vid oxidskiktets sidovägg ökar för att komplettera minskningen av den vid metallskiktet. Simuleringarna med strålningsemissiviteten 0,2, 0,35, 0,45 och 0,7 visar att emissiviteten under 0,45 påverkar värmeöverföringen, och temperaturerna och sidoväggens värmeflöde för både oxidskiktet och metallskiktet kommer att öka med minskande emissivitet. Effekten är försumbar när strålningen är över 0,45. Simuleringarna under de hypotetiskt extrema förhållandena med antingen en adiabatisk övre gräns eller ett mycket tunt metallskikt visar att fokuseringseffekten kan uppstå, dvs. värmeflödet genom metallsidan är större än det i oxidskiktet. Men det lokala höga värmeflödet plattas ut av kärlväggen med god värmeledningsförmåga. Sammanfattningsvis visar simuleringarna att, förutom fall under extrema förhållanden, är värmeflödet från smältpoolerna i alla andra fall betydligt lägre än CHF för extern kylning av nedre huvudet. Därför verkar säkerhetsmarginalen för IVR-strategin för den valda PWR tillräcklig. På grund av vissa begränsningar (t.ex. förenkling och antaganden) i simuleringsfall och koppling av olika inflytelserika faktorer, vilket indikeras av den aktuella studien, är de exakta förutsägelserna av värmeflöde under alla scenarier fortfarande svåra. Därför kunde slutsatserna inte generaliseras till de andra förhållandena eller andra konfigurationer av de smälta poolerna. Genom att diskutera modellen och förenklingar / antaganden som antagits i detta arbete föreslås förbättringsriktningarna för den numeriska modellen och andra perspektiv i slutet av avhandlingen.
|
Page generated in 0.1252 seconds