For a natural gas importing country, &ldquo / take or pay&rdquo / approach creates problems since the demand for natural gas varies during the year and the excess amount of natural gas should be stored. In this study, an underground gas storage project is evaluated in a depleted gas Field M. After gathering all necessary reservoir, fluid, production and pressure data, the data were adapted to computer language, which was used in a commercial simulator software (IMEX) that is the CMG&rsquo / s (Computer Modelling Group) new generation adoptive simulator, to reach the history matching. The history matching which consists of the 4 year of production of the gas reservoir is the first step of this study. The simulation program was able to accomplish a good history match with the given parameters of the reservoir. Using the history match as a base, five different scenarios were created and forecast the injection and withdrawal performance of the reservoir. These scenarios includes 5 newly drilled horizontal wells which were used in combinations with the existing wells. With a predetermined injection rate of 13 MMcf/D was set for all the wells and among the 5 scenarios, 5 horizontal &ndash / 6 vertical injectors &amp / 5 horizontal - 6 vertical producers is the most successful in handling the gas inventory and the time it takes for a gas injection and production period. After the determination of the well configuration, the optimum injection rate for the entire field was obtained and found to be 130 MMcf/D by running different injection rates for all wells and then for only horizontal wells different injection rates were applied with a constant injection rate of 130 MMcf/d for vertical wells. Then it has been found that it is better to apply the 5th scenario which includes 5 horizontal &ndash / 6 vertical injectors &amp / 5 horizontal - 6 vertical producers having an injection rate of 130 MMcf/d for horizontal and vertical wells. Since within the 5th scenario, changing the injection rate to 1.3 Bcf/d and 13 Bcf/d, did not effect and change the average reservoir pressure significantly, it is best to carry out the project with the optimum injection rate which is 130 MMcf/d. The total gas produced untill 2012 is 394 BCF and the gas injected is 340 BCF where the maximum average reservoir pressure was recovered and set into a new value of 1881 psi by injection and cushion gas pressure as 1371 psi by withdrawal. If 5th scenario is compared with the others, there is an increase in injection and production performance about 90%.

L'objet de cette thèse est l'étude des possibles mécanismes élastiques expliquant l'amplification du bruit ambiant au droit de certains réservoirs multi-phasiques. Trois jeux de données sont traités. La signature spectrale observée d'un réservoir de vapeur géothermique est différente de celle d'un stockage de gaz. Dans une approche empirique, un algorithme de classification permet d'extraire et de cartographier les anomalies que l'on présume liées au réservoir. Un travail de modélisation est effectué pour tenter d'expliquer les anomalies mesurées. Dans les données réelles, une forte présence de modes supérieurs d'ondes de Rayleigh est détectée. On modélise numériquement en 2D la propagation de ces modes à travers un réservoir placé au sein d'une structure géologique réaliste. La réponse simulée du réservoir se révèle trop faible par rapport aux observations de terrain. Néanmoins, on parvient à inverser les faibles perturbations d'amplitude synthétiques pour la position du réservoir, dans des modèles de référence simples. Cette méthode pourrait être utilisable pour l'imagerie à partir de faibles variations d'amplitudes dans le cadre du monitoring. Pour ce qui est de fortes anomalies observées sur le terrain , il est à noter que les effets visco-élastiques, les effets 3D, et les effets liés à un éventuel champ incident diffus n'ont pas été pris en compte dans la modélisation. Ainsi ce travail n'exclut pas la possibilité de telles anomalies liées à la présence d'un réservoir. / This PhD work investigates the possible elastic mechanisms behind the ambient noise amplification above multi-phase fluid reservoirs. Three datasets are analysed above different reservoirs. The observed spectral signature is different in the gas storage and geothermal contexts. A non-supervised algorithm for amplitude spectrum classification is developed, allowing to extract and map the relevant attributes of a multi-phase fluid presence. As a first modelling step, a wavefield characterisation methodology is applied to determine the composition of the ambient noise. It reveals the presence of strong Rayleigh overtones. Numerical 2D elastic modelling is used to simulate the propagation of overtones across a reservoir within a realistic geological structure. The modelled reservoir response is too small compared to the real data. However, the small amplitude perturbations arising in the numerical simulations are successfully inverted for the position of the reservoir, in simple background models. The developed method could in theory be used for imaging small time-lapse amplitude variations (monitoring), despite the obstacles remaining to be overcome before a real-data application. Neither visco-elastic nor 3D effects are adressed. Thus this work does not exclude the possibility of strong reservoir-specific spectral anomalies.

The diploma thesis is concerned with the issue of investments decision and capital investment planning. The aim of the thesis is to evaluate a possible investment in the underground gas storage Rožná. The thesis is divided into three parts. The first theoretical part deals with the characterization and terms defining in the field of investments decision. It also mentions the methods how to evaluate the effectiveness of investments decision. The second theoretical part discusses the strategic analysis of gas industry with the emphasis on using of the underground gas storages in the Czech Republic and in other countries within the European Union. The third part analyzes the effectiveness evaluation of the investment while applying the terms of both theoretical parts.