Analysis of Drill String Dynamic Behavior
Research output: Thesis › Doctoral Thesis › Research
Vibrations are caused by bit and drill string interaction with formations under certain drilling conditions. They are known as destructive loads and thus are leading to downhole fatigue failures, severe bottom hole assembly and bit wear and may cause wellbore instabilities. Vibrations are affected by different parameters such as weight on bit, rotary speed, mud properties, bottom hole assembly and bit design as well as formation characteristics. During drilling operations, the drill bit interacts with different formation layers whereby each of those formations has usually different mechanical properties. Vibrations are also indirectly affected since the weight on bit and the rotary speed of the drill string are usually optimized against changing formations. Since uniaxial compressive strength is one of the representatives of the formations characteristics, axial vibrations are expected to change with the variations of the uniaxial compressive strength of the formations. In this work the relationship between formation characteristics and drill string vibrations in laboratory scale have been studied. In addition, optimizing weight on bit and rotary speed to enhance the advance rate of drilling operations and manage vibrations has been considered as another main objective of this study. Non-linear models have also been built to establish a relationship between drill string vibration and other parameters. A fully automated laboratory scale drilling rig, the CDC miniRig, was used to conduct experimental tests. A vibration sensor sub attached to the drill string, above the bit, recorded the drill string vibrations during the tests and an additional sensor system recorded the drilling parameters. Numerous uniform and layered concretes cubes as well as different types of uniform and layered rocks were drilled by a double roller cone bit. The uniaxial compressive strength of the concretes and rocks were measured prior to the experiments. The experiments were conducted using different combinations of weight on bit and rotary speed. Optimum drilling parameter windows based on the experimental results of uniform cubes, leading to minimum drill string vibrations and maximum rate of penetration, have been achieved. Therefore monitoring the drill string vibrations can help the driller to enhance the quality of drilling operations and increase the rate of penetration. Performing the layered experimental tests using the optimum drilling parameters obtained from uniform experiments and analysis the measured and calculated data (drill string vibrations and mechanical specific energy) revealed the fact of changes in the axial vibration measurements due to variations of uniaxial compressive strength of the layers. It is concluded that the formations can be recognized in real-time by incorporation of the vibration measurements and allowed to differentiate the individual layers with different strengths. Since linear equations are not sufficient to describe the relation between the drill string vibrations and other parameters, non-linear models like artificial neural networks in combination with sequential forward selection method were used. Different models to estimate the rate of penetration based on the drilling parameters, drill string vibrations and formation characteristics were built. The axial vibrations can be estimated using bit bounce models when they are being provided with drilling parameters and formation characteristics. Measured and calculated data from the uniform experimental tests formed the formation prediction model. Predicted (recognized) formation is the output of the model when it is being fed with the drilling parameters and drill string vibrations.