Shor, RomanKantzas, ApostolosPrasad, Mandyam Rangayyan Prathik2024-08-202024-08-202024-08-14Prasad, M. R. P. (2024). A novel data science approach to estimate a borehole caliper log utilizing tripping data (Master's thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca.https://hdl.handle.net/1880/119422During drilling operations, monitoring the borehole integrity is crucial. Borehole irregularities, such as constrictions, ledges, and breakouts, can lead to significant non-productive time and present challenges when casing and cementing the well, adversely affecting operational efficiency. These issues are particularly problematic during tripping – the process of moving the drill string in and out of the borehole. However, monitoring rig data during trips presents an opportunity to monitor the same problems. Early identification of such dysfunctions allows proactive measures to mitigate their impact on operations. This technique shows promise for estimating borehole quality in advance of wireline caliper measurements while also allowing for deeper estimation in geothermal wells where current caliper tools cannot log due to high temperatures. This study introduces a novel forward/backward modelling strategy that utilizes resistance signatures in conjunction with drill string configurations to identify and quantify borehole dysfunctions, thereby enhancing operational efficiency. This approach is based on the analysis of overpull, calculated by subtracting the expected hook load (HL) calculated from a soft string drill string model from the real-time measurements of HL. Two signals were used for analysis: the Bottom Hole Assembly (BHA) configuration signal and a synthetic resistance signal. The BHA signal represents the outer diameter (OD) of the BHA, whereas the synthetic resistance signal represents the borehole gauge vs. depth. A synthetic overpull signal can be obtained by convolving the BHA signal with the resistance profile at each depth. In the reverse problem, the real overpull signal may be deconvolved with the BHA signal to estimate the resistance signal. The model was applied to actual overpull data from a drilling operation. First, the resistance signal was manually estimated to enable the overpull signal to match the field data. There was a good match between the resistance signal and the measured four-arm caliper log. A backward model was then applied to calculate the resistance signal. For shallow depths, the model is capable of detecting changes in the caliper log; however, for deeper depths, drill string dynamics confound the signal, presenting an opportunity to perform further work.enUniversity of Calgary graduate students retain copyright ownership and moral rights for their thesis. You may use this material in any way that is permitted by the Copyright Act or through licensing that has been assigned to the document. For uses that are not allowable under copyright legislation or licensing, you are required to seek permission.Engineering--Chemicalmaster thesis