Parameter Back-analysis And Optimization Method Study On Low Permeability Reservoir Fracture Prediction

The distribution characteristics of stress fields and reservoir fractures of the low permeability reservoir play an important role in the hydrocarbon migration, well patterns arrangement and adjustment, optimization design of drilling strings and such relevant fields. However, the rocks complicated formations and properties cause large ambiguity and a lot of limitations to the recognition of the rocks inherent attribute, and as a result the research of the reservoir stresses and fractures is still at the exploratory stage.This dissertation mainly focused on the hot issues and difficult questions in the study of rocks property parameters, reservoir stress distributions and fracture characteristics, learn from the predecessors research results, and make use of modern mathematics, mechanics, and computer knowledge to propose an actual analysis and prediction technique to the stress field and reservoir fracture distribution suitable for low-permeability oil and gas reservoirs, by means of theoretical deduction, numerical simulation, case study, corresponding software development and so on.First of all, established an effective model for the acquisition of rocks parameters based on the relationship between rocks acoustic characteristics and mechanical properties, calculated the rocks property parameters in the light of core test data and optimization back analysis technique utilizing large number of conventional logging information, studied the rocks stochastic and fuzzy attribute, and obtained its statistical eigenvalues of rocks elastic modulus and Poisson's ratio.Secondly, derived mathematical and FEM models for the reservoir stress analysis and prediction, set the magnitudes and azimuths of some key wells in Jiangsu Oilfield as the constraints, made use of multi-objective optimization techniques to obtain the distribution regulations of principal stresses, pore pressures and fracture stresses. Fit the related curves of the principal stresses versus the depths of Jiangsu Oilfield, whose results indicated that the accuracy can meet the engineering requirement.Thirdly, carried out a research on the reservoir rock stress state functions and the failure surface equations expressed by multi-strength parameters, put forward a multi-parameter model for the analysis and prediction of reservoir fractures, which can make up effectively for many defects of conventional criteria. At the same time, in order to consider the influence of rocks original defects, the dislocation theory was introduced to the reservoir fractures research, and a novel model based on dislocation theory was presented, which provided a feasible and effective method for reservoir fractures prediction. Furthermore, a stochastic finite element analysis using the Monte Carlo method was introduced into our research to study the effect of the probabilistic rocks parameters to the reservoir fractures. All our case study results show that the models in this dissertation can give out a better prediction of the fractures, and reflect the distribution of the fractures'density.Finally, a corresponding software named "Reservoir Stress and Fracture Analysis system (RSFA)" was designed, which provided an effective way for the theoretical research results applying to the actual oil production comprehensively and conveniently, and make a simple implementation of the transformation from the theoretical approach to the practical application of engineering.