How to Organise Mud Logging Data For Modelling Rate of Penetration

Abstract

ABSTRACT ROP (Rate of Penetration) modelling can provide a quantitative tool to reduce drilling costs by means of bit selection and determining the optimal combination of mechanical and hydraulic operating parameters. It is recognised that it would be desirable to model ROP using mud logging and electric wireline logging data because such data closely represent the drilling conditions and formation properties respectively. However, this approach has long been diminished by the apparent poor quality of the data available. This paper examines the nature of logging data and describes the relationships between ROP and its related variables. We propose an approach to improving ROP modelling using field data. As a result, a number of empirical ROP models have been developed using routinely available mud and wireline logging data. The models are shown to be physically valid under a wide range of conditions. An appreciation of the ranges of conditions under which the model will apply is a critical consideration given the varying contribution of factors affecting ROP under different drilling conditions. The validity and robustness of the new approach are demonstrated by applying it to the logging data from three Poly-crystalline diamond compact (PDC) bit runs in the North Sea area. The method holds the potential to be fully automated for routine application.