Currently considerable part of new oil and gas fields put into production, features structurally complex low permeable reservoirs. Under complex Arctic climatic conditions, Yamal in particular, development of gas-condensate deposits seems to be especially difficult and capital intensive task. To increase projects commercial efficiency, productivity of each gas well needs to be maximized. Classic solution here is to drill vertical well (VW) with fracture. Under complicated conditions especially when developing offshore fields, where drilling unit costs are considerably higher than in already developed oil and gas bearing regions of Russia, horizontal or sub-horizontal wells (HW) with multiple fractures can be more efficient. To evaluate process efficiency under conditions of Urengoiskoe gas-condensate field, a high resolution sector reservoir model for a single well was built. The model was used to calculate production parameters for various bottomhole design variants: vertical well with fracture, horizontal well with several transverse or longitudinal fractures. In this sector model, each fracture was explicitly simulated, and in number of models minimum width of a cell containing fracture was 5 to 10 mm, which correlated with real fracture width. Analytical calculations and simulation results, considered in this paper, are indicative for a number of typical effects for gas-condensate wells, which should be given a notice: ▪Generation of condensate bank in bottomhole formation zone, leading to well productivity decline;▪Inertial effects in fracture (Forchheimer effect) caused by high gas filtration rate and also resulting in well productivity decline;▪"Straightening’ of relative permeability curves in the areas with high fluid velocity.

Most of oil and gas companies are building simulation models for its assets to planning reservoir development. However, even in those cases when the models are reliable and well calibrated to the production history, they do not always reflect the interaction between the different parts of a single system "reservoir-well-gathering system-processing facilities". In other cases when hydrodynamic models are the parts of an integrated model (IM), the models become too complex and require a long time of simulation, what mostly is not very convenient. This article provides an example of the IM building for the two formations of one of the largest oil-gas-condensate field in the world. Two large gas-condensate reservoirs are in the pilot stage. Full-field development of these reservoirs will increase hydrocarbon production by 5 times (Figure 1). To develop and optimize production plans and the development of the asset, it was decided to use the integrated model. There were considered different methodologies for constructing the unified model, which combines the reservoir models, models for wells and gathering systems and for processing facilities. Finally the best approach for this project has been selected. The initial compositional hydrodynamic models, which was matched to the production history, have been successfully converted to the Black-Oil models, while giving identical forecasts for gas and condensate and significantly reducing the simulation time. Well models were calibrated to historical data. The formation fluid in the gas gathering network was modelled using a simplified description, while in the models of processing facilities the fluid was modelling with the detailed composition. Despite of the differences in the approaches to the description of PVT properties of gas condensate in different simulators (Eclipse, Gap, Hysys), the developed Integrated Model has demonstrated consistency in the description of fluid PVT-properties. A significant reduction in time of simulation was obtained during the forecast calculations. The results of the Integrated Model were very important for the field development plan optimization, the development of which was previously limited to disparate models of reservoirs and ground infrastructure.

Progress in multi-parallel computers opened up new possibilities in reservoir modelling providing opportunities in compositional simulation to study in detail highly complex physical process that are extremely computational demanding. This is particularly true for probabilistic methods such as the uncertainty quantification and analysis. Experimental design, ED, concept supersedes the conventional deterministic uncertainty analysis and offers new efficient and system approaches to identify and assess the impact of the key geology and engineering risks in strategic reservoir engineering decisions. Use of response surfaces, Monte Carlo stochastic modelling is an alternative to numerous full-scale reservoir simulation and is employed in petroleum industry since long ago [Amudo et al. 2008]. However, although ED has been introduced to reduce computation, it is computational expensive especially in compositional simulation with many degrees of freedom in the uncertainty parameters. This paper shows implementation of ED and the use of multi-parameter sensitivity analysis based on multidimensional Response Surfaces Models (RSM). The method was validated when key risks associated with development of a unique gas-condensate field in Yamal peninsular, Russia, were assessed. This reservoir is characterized by poor permeabilities in the range around 1.0 mD, abnormally high reservoir pressure and temperature (600 atm, up to 115 °C), high potential condensate content (potential C5+ content 280-380 g/m3) and the risk of a heavily compartmentalized some blocks of the field through faulting. The numerous identified uncertainties impacting the development of the gas-condensate reservoir are difficult to address in a traditional reservoir study using deterministic methods and only a probabilistic approach would guarantee to assess the production and recovery potential in all aspects and, moreover, allows informed decisions to be made on the data acquisition – or in general the appraisal strategy – and the development.

The Urengoi field (Novo-Urengoiskiy license area - NU LA) is used in this article as an example to review available knowledge on the Achimov deposits based on seismic, geological and field development data, as well as reservoir-scale geological simulation. A comprehensive approach is required for detailed characterization of the deposit structure to enable forecast of best prospects and differentiated treatment of individual areas of the field as development efforts are planned for the future. Three geological bodies were distinguished at the Urengoi field (NU LA) in the Achimov formation each having its own turbidite fans simulated in the 3D reservoir model.