Drilling penetration rate or rate of penetration (ROP) is one of the bit performance indexes. This paper introduces the experimental set-ups for rock failure by inserts of roller cone bits, experiments involved and a method for processing the experimental data. A new drilling ROP model is developed and presented with verification using several groups of lab data. The model is derived directly based on rock craters fractured by a single insert. The work behavior of each insert is not a simple indentation, but includes both crushing and shearing, which represents the actual movement of inserts on a roller cone bit. Finally, the paper focuses on applications of the model in predicting ROP and rock compressive strength with offset well data from Western Canada. The model can reflect the interaction between rock bit and the rock on the bottom hole, and the field verification results are good. The ROP trends from the model match those from the field wells analyzed. In addition, simulations were carried out to predict the unconfined rock strength of formations with the same model using two sets of offset well data. The predicted rock strength matches the strength from log data well, both in trend and values.
One way of improving the drilling operation efficiency and to reduce drilling cost is to develop simulation and real time analysis tools. These tools can optimize the drilling operation by predicting and comparing drilling performance. Modeling bit performance is a scientific approach to optimizing drilling performance. Drilling penetration rate or rate of penetration (ROP) is one of the most important bit performance indexes. Therefore improvements in ROP will bring increased efficiency to the drilling industry. Several ROP models for roller cone bits have been developed. These models can select best weight on bit and rotary speed to achieve the minimum cost per foot. Among them, the most complete mathematical drilling models being used includes Bourgoyne and Young's model [1], Warren's model [2, 3], and Warren's model modified by Hareland [4, 5]. These models can be applied in ROP prediction and realtime drilling simulation. However, there are in some cases differences between the model predictions and the field observed data. This is often due to the technical complexity of the bit-rock interaction, especially in uncontrollable field conditions. In the late 1990s, Ma developed a computer simulation program on the bitrock interaction [6, 7]. The program reflects the effect of cutting structure on the ROP very well. The drawback of the simulation program is that it is complex and time consuming to run, so it is impossible to use in real-time drilling simulation. This paper shows the development of a simple ROP model to predict the roller cone bit's performance which can be used for real-time simulation purposes. Then the model is modified considering bit wear. The model is a ROP model, and the initial bits simulated are some of the most utilized roller cone insert bits including the IADC 437, 447, 517 and 627 types.
