A New Robust Drilling Rate Model

Abstract Drilling rate remains a major challenge when it comes to planning and drilling workover and development wells. The main mission of a drilling engineer is to design a well optimizing time, cost, economics and safety. Analysis of previously drilled wells operations' records is required to perform optimization techniques to reduce drilling cost for new wells to be drilled. Among the many potential optimizations, Rate of Penetration (ROP) has the most obvious impact on cost effectiveness of a well's drilling, but to ensure optimized rate of penetration, it must be engineered. There are several correlations, methods, designs, models, tools, charts, fields' results, and experimental studies to enhance the well drilling performance. Many of these are effective, but some others are incomplete and not suited for drilling operations as they are based on unproven theory and lack proper experimental data. Proper models must include knowledge of all factors that affect rate of penetration in order to provide a meaningful tool for optimizing the well design. The objective of this work is to develop a new method for optimized drilling rate that will model rate of penetration more accurately and therefore facilitate improving drilling efficiency and cost. The drilling parameters and mud rheological properties in certain hole sections were collected and analyzed first to determine the effect of mud properties and drilling parameters on ROP performance. The parameters and properties that are selected are from the same hole size, formation type and mud type. The relationship between mud rheological properties and ROP was then evaluated to determine how strong it is. This step helps to determine the significance of mud rheological properties on estimating ROP and that will lead to optimization of the drilling operation and reduction in the drilling time. This is the first model combining the drilling fluid properties, drilling parameters, cuttings volume and dogleg severity with rate of penetration optimization simultaneously. The developed model has been compared with using field data during drilling challenging hole sections in several different fields. It has shown high correlation coefficient regression value matching with actual ROP Values with high percentage of accuracy, which is about 90 %. The new model showed the importance of combining mud properties, cuttings volume percent, drilling parameters and dogleg severity since that will make it applicable in any type of wellbore or hole section such as (vertical, deviated and horizontal). The developed model can assist drilling engineers in selecting improved drilling parameters, mud properties, optimized value of cuttings volume or cuttings concentrations in annulus and dogleg severity value by optimizing the drilling rate using the developed model effectively. It can be used in real time operating center to participate for drilling Automation projects.