This paper proposes a novel method for the grouping of rock types for the successful estimation of the in-situ absolute permeability accurately and for the smooth interpolation of relative permeability endpoints which are inevitably required for reservoir simulation models. The method consists of the grouping reservoir rocks based on the concept of electrical facies in accordance with their response to the wireline log attributes. The results showed excellent matching of the estimated absolute permeability and relative permeability endpoints to the core data prior to the application of the hydraulic unit method and multi-variate regression on each data group. The individual electrical facies showed the unique petrophysical characteristics each other, and their distribution frequency in each sedimentary zonation exhibited distinguishing feature possibly reflecting the sedimentary environment. The proposed approach lead to the great success on the description of the carbonate reservoir, of which petrophysical characteristic is so complicated that accurate enough estimation of the pointwise petrophysical properties could not be obtained by simple application of the ordinary hydraulic unit method nor multi-variate regression. The methodology is totally general and is considered to be applicable for a variety of reservoirs.
The reservoir characterization is becoming more important since the result of the reservoir simulation is highly dependent on the understanding of the reservoir. One of the multi-disciplinarily approaches of reservoir characterization work for the reservoir simulation of a Middle Eastern carbonate reservoir has been presented in this article. A series of rock types were generated which are meaningful for geologists and reservoir engineers, using thin section description, hydraulic flow units technique and pore throat size distribution data. Huge database was created and analyzed by geologists and reservoir engineers. Then the rock type distribution was generated by 3D visualization software package.
Quantitative assessment of production uncertainties is the goal of applying probabilistic methods to dynamic reservoir modeling. Sophisticated geostatistical algorithm that generates equiprobable alternatives of reservoir image benefits only when they are transformed into stochastic production forecast or economic indices through flow simulations. However, simulating multi-phase displacement processes on statistically meaningful number of realizations, each of which is represented by millions of pixels requires a large CPU's and is usually unrealistic. Therefore, the industry has been making efforts to compress the whole work to a size that can be handled with commonly accessible computing facilities. Attempts have been made at any stage of the evaluation process. Ranking is a method which reduces the number of realizations to be flow-simulated. It can reduce the total computational cost without loosing resolution for describing flow when a ‘good’ ranking index exists. Streamline/streamtube simulations that map one-dimensional displacement profiles onto streamtubes significantly reduce computational time when problems are simplified. Implementation of gravity, capillary and diffusivity effects is actively researched and application to three-dimensional problems is rapidly brought into practice. Upscaling is an attempt to represent single- or multi-phase flow on sub-grid scale reservoir heterogeneities with effective properties of coarser simulation grids. The robustness of boundary conditions imposed on isolated coarse grids for calculating effective permeability is yet to be established. Effective relative permeabilities calculated from the solution of a fine-grid simulation is specifically called pseudofunctions. The ambiguity of averaging criteria for calculating pseudofunctions was only recently settled. Any method that reduces computational cost is, in return, making assumptions in flow modeling. One has to make choice among alternative methods according to the physics dominantly affecting the system of interest. A sufficient understanding on the flow mechanism is the bottom line for exploring the acceleration of flow simulations and any other aspect of reservoir engineering.
Permeability is one of important factors for a reservoir simulation model because flow in a model is affected a lot by this parameter. Whether “reservoir characterization” method or conventional reservoir evaluating method would be taken to construct a simulation model, permeability distribution along the wells is important, because the distribution is used as a base to predict field-wide permeability distribution. In this study, two permeability prediction methods were evaluated for estimating uncored well permeability from wire line well logs. One is method using artificial neural network (ANN), which is suitable to analyze complicated non-linear problem. The other is multiple linear regression analysis (MLRA) method. Also classical empirical porosity-permeability relationship method was evaluated to compare the accuracy or validity of ANN and MLRA methods. ANN showed little improvement and MLRA showed no improvement, compared to conventional porosity-permeability relationship result. Especially, better results were seen when all data were fed in to the ANN system, assuming that no geological information was provided. However, even ANN, which showed the best result, is not recommendable method for this reservoir because of lower quality than required. In conclusion, these trials showed difficulty of estimating uncored well permeability in this reservoir because of poor log quality and the nature of this complicated carbonate reservoir.
Due to complicated nature of the fluvial sandstone reservoir, management of the Snorre Oil Field, located in Norwegian Northern North Sea, is quite challenging. The field has been mainly exploited under water injection scheme for several years, while WAG (water alternating gas) injection seems also promising in view of increased recovery as well as a gas disposal solution. However, the ongoing WAG pilot test indicates the difficulty in forecasting the movement of injected gas and evaluating WAG efficiency. The key uncertainty will be in the reservoir description and in some of the reservoir engineering parameters. A pilot test of 4D seismic survey has been initiated in Snorre Field, where the movement of gas front is to be monitored as time-lapse seismic response. Petroacoustic study and elastic modelling was performed to investigate the technical feasibility of monitoring the gas front under reservoir conditions in Snorre. The study indicated that gas in the reservoir could be monitored as significant response and also showed that low-noise survey and adequate procssing would be essential. A pilot survey started in 1996 and sea bottom cables have been employed as the receiver to secure accuracy and resolution of the time-lapse data. Only a base survey has been completed so far and further shooting will be done in 1997 after injecting sufficient gas to be monitored. The pilot survey is being performed to reduce the reservoir uncertainty and to establish a tool to monitor WAG process in the field. Aggressive reservoir management is required in Snorre Field to exploit the reservoir in an optimal way. The ongoing 4D seismic survey pilot test is expected to have an important contribution to the reservoir management in the field.
Reservoir history matching which has been done through trial and error by engineers is viewed from the standpoint of the dynamic inverse problem, in which automatic history matching is performed as a nonlinear optimization process. The basic concepts, algorithms and example results are briefly introduced, and various obstacles for practical use of it are pointed out. In addition, responding to discussions in the symposium, our ground of research is explained. Lastly, a recommendation is made concerning the systematic field operation that might be useful to improve identifiability of reservoir properties through well performances.
Fission track fading in apatite has been measured in samples from the Shin-Yahiko-1 well, which is situated in a very young sedimentary basin in Japan. Fission track densities were too low to detect confined fission tracks, and consequently semi-track lengths were measured directly instead. Mean lengths and frequency distribution of semi-tracks decrease clearly and progressively with increasing temperature and depth below 4, 400m, and the fading zone of semi-tracks spans an interval of less than 1, 500m, and 40°C in temperature. The smooth trend of fission track fading in the Shin Yahiko-1 well suggests a relatively simple subsidence burial history, which is concordant with the burial history roughly estimated from ages of planktonic foraminifers. From the Shin Yahiko-1 well data, fission track fading over a a very short effective heating time (0.05Ma) delineates a zone on the vitrinite reflectance nomograph. GC/MS analysis suggests that the oil produced from the Shin Yahiko-1 originated from the highly matured Teradomari Formation, which occurs at a depth greater than 4, 500m and roughly corresponds to the top of apatite fission track fading zone. Semi-track length measurement is useful as a supplementary thermal indicator even for very young basins, especially in igneous basement rocks where vitrinites are absent.