Brittleness assessment of the shale oil reservoir based on neural network method: A case study of the Yanchang Formation, Ordos Basin

Abstract

The Chang 7 oil-bearing layer of Yanchang Formation is rich in shale oil resources. However, the reservoir indicates strong reservoir heterogeneity and has obvious differences in productivity among wells. The brittleness acts as an important factor causing the above phenomenon. Studies on rock mechanical properties and brittleness evaluation can provide technical support for drilling and fracturing design, but current methods have many disadvantages such as difficulty in obtaining parameters. In this study, the neural network method is used to construct the relationship between elastic modulus, Poisson's ratio and logging curve on the basis of measured rock mechanical parameters, construct rock mechanics parameter profile in the whole well section of Chang 7 oil-bearing layer, and finally quantitative evaluate shale oil reservoir brittleness. The results indicate that, (1) the neural network method is an effective method for rock mechanical parameter prediction and brittleness evaluation. The error between the predicted and measured values of elastic modulus and Poisson's ratio is low, generally within 10%; (2) according to the brittleness evaluation results, on the whole, the brittleness index of the Chang 72 reservoir is high, and the brittleness index of the Chang 73 reservoir is low. The results can provide scientific guidance for benefit development of Chang 7 shale oil in the Ordos Basin.