Construction and optimization of adaptive well pattern based on reservoir anisotropy and uncertainty

Abstract The design of well pattern is of great significance to oilfield development. In view of the shortcomings of traditional well pattern, this paper proposes a new adaptive well pattern construction method. By using an unstructured meshing algorithm, large-scaled high-quality triangular and quadrangular adaptive well patterns are generated, which can effectively deal with various complex reservoir conditions. Existing wells and irregular boundary conditions are constrained and used to obtain the optimal well spacing. Thereby, a new vector well pattern is constructed by making a vector transformation to the adaptive well pattern. Based on this, a well pattern optimization method is proposed, in which NPV is set to be the objective function and key points as the optimization variables. The well pattern after optimization can perfectly adapt to reservoir geological anisotropy and heterogeneity, recovering more oil with fewer wells. In addition, reservoir uncertainty is also taken into consideration. Reservoir stochastic models are generated, and the expectation and standard deviation of the NPV values are used to construct the objective function. In coping with the huge calculation amount caused by reservoir uncertainty, a fuzzy C-mean clustering algorithm (FCM) & loop optimization method is combined to ensure the accuracy of optimization while reducing inefficient calculation. In the end, two examples are adopted to examine the reliability of the method proposed. The results show that the well pattern can perfectly adapt to anisotropic and uncertain reservoirs, and the optimization method can remarkably reduce the computational amount while ensuring the accuracy of the results.