Abstract

Confined Compressive Strength (CCS) offers a more accurate indication of rock hardness and rock formation drillability, which is useful in optimizing the drilling operation. PhiDrillSim is a machine learning app that predicts the CCS using neural network techniques. The goal is to predict the Confined compressive Strength using the Neural network technique based on the operation dataset, which will be used to train the neural network model. A specific operation dataset will be provided for the training of the model in the app. It will predict the CCS model of the rock formation and compare it with the actual CCS collected. Preliminary results show that the accuracy of the predicted CCS values is consistent with the CCS calculations using the log data method. There are various methods of calculating the CCS of the formation, with varying degrees of cost and difficulty. Furthermore, the accuracy of CCS measurements directly affects the drilling cost of operations, affecting drill-string durability and operational time. This paper explores the accuracy of the PhiDrillSim App in comparison with a log data method in calculating and predicting the formation of CCS.

CCS Description

Rock strength is essential in the drilling process. Confined Compressive Strength (CCS) is a geomechanical rock property that indicates the rock strength when confined to some medium. (Fabian, 1994). Since UCS is widely used, the classification of rock formations based on UCS is readily available. Moreover, the estimation of rock strength classification is based on UCS. CCS then will be related to UCS using the equation proposed by Caicedo et al., 2005.

This content is only available via PDF.
You can access this article if you purchase or spend a download.