Natural fractures are the main producibility factor in the weathered granite reservoirs (basement rock) and volcanic-rock reservoirs. Production practices show that these reservoirs could have high production rate, but the difference of well productivity between single wells is obvious. These reservoirs have complex natural fractures oriented at medium-high angles, which could bring high complexity and heterogeneity to the reservoirs, adding anisotropy to reservoir permeability. It is very hard to effectively simulate complex fractures in naturally fractured reservoirs and study the applicability of different well type and well pattern by using common reservoir simulators. A fast EDFM (Embedded Discrete Fracture Model) method was put forward for production simulation of complex fractures in naturally fractured reservoirs. The EDFM processor combining commercial reservoir simulators (ECLIPSE or CMG) is fully integrated to forecast production performance of the weathered granite reservoir. With a new set of EDFM formulations, the non-neighboring connections (NNCs) in the EDFM are converted into regular connections in traditional reservoir simulators, and the NNCs factors are linked with gridblock permeabilities. So complex dynamic behaviors of natural fractures can be captured, which can maintain the accuracy of DFMs (discrete fracture models) and keep the efficiency offered by structured gridding. In this paper, a 3D model with complex natural fractures was built to model the performance of different well types and well patterns. The results show that wells with higher density of natural fractures produce higher oil production, and horizontal wells with higher density of natural fractures have larger oil production than vertical wells because horizontal wells have a larger contact area than vertical wells. What’s more, heterogeneity and anisotropy have a great effect on well pattern and well type, which need to be studied carefully in the oilfield development.

Lost circulation during drilling wells is very detrimental since it greatly increases the non-productive time and operational cost, also seriously lead to wellbore instability, pipe sticking, blow out, etc.. However, in the process of drilling wells, geological characteristics and operational drilling parameters all may have impacts to the lost circulation. This makes the establishment of the relations between the lost circulation and drilling factors very challenging. In this paper, we tested five different kernel function (linear, quadratic, cubic, medium Gaussian and fine Gaussian) derived support vector regression (SVR) models and four-layer artificial neural network (ANN). By combining their accuracy and time efficiency, the ANN is regarded as the optimal predictor of lost circulation. By training ANN using different combination of drilling features, we concluded that depth, torque, hanging weight, displacement, entrance density and export density are the key factors to accurate predict the lost circulation. The corresponding trained ANN network can achieve 99.2% accuracy and evaluate whether a drilling feature vector corresponds to lost circulation or not in milliseconds.

No abstract

No abstract

No abstract