| dc.description.abstract | During drilling operations the mud pit volume varies quite a lot. These variations are unavoidable due to different short term factors such as tripping in and out of the hole and pump rate changes, and different long term factors such as heat exchange in the well and removal of cuttings. The mud pit can also operate as kick indicator on the rig, but due to the constant variations in the pit level, this is challenging. By a better understanding of the factors controlling the changes in the mud pit level, it might be possible to make a reliable model to spot kick at an early stage. This report has briefly discussed some of the different factors that affect the mud level in short term and long term. These factors include mud pump, viscous flow, ballooning, drilling fluid, cuttings and tripping.
A model was also presented for the estimation of the mud pit level during various drilling operations. The model uses a logarithmic equation for the estimation of the mud pump shut off period, and mud pump flow rate along with rate of penetration and block position for the other estimations.
The model was tested for one set of drilling data from a land rig. Since data were missing, it was run 7 times for different hole size, initial cuttings concentration and mud pump factors. The outcome from the model showed the different drilling operations in a fair manner, and especially the drilling phase showed accurate behavior. A larger hole size resulted in larger reductions in the mud pit volume. Higher initial cuttings concentration resulted in bigger changes in the mud pit. And higher mud pump shut off factor resulted in higher estimated values for the calculation of the mud pump shut off period.
The model still got severe limitations, and especially the mud pump shut off estimation has room for improvements. Another limitation of the model is the absent of drilling data from different wells in different environments. The model is also based on a vertical well, which is insufficient since most wells are deviated. | nb_NO |
