Casing Wear Simulation by Application of Real-Time Data and Predictive Modelling

Casing wear poses a significant safety hazard during drilling and production of hydrocarbons. Failure to maintain integrity due to inaccurate casing wear estimation can cause severe accidents or preclude prospective operations. Current industry practice is to estimate casing wear during the planning phase of the well and subsequently use assumed operational parameters with inherent uncertainties. This paper aims to study how to utilize real-time data to improve the industry standard methodology and evaluate the benefit of the modification. The research was conducted by applying the model on data from a well on the Norwegian continental shelf. There were two main objectives of the research. Firstly, the industry standard approach to casing wear estimation was expanded to include real-time data. Application of real-time data to the industry standard method for estimating casing wear caused a significant difference in results. The approach using real-time data resulted in an estimate of more casing wear compared to the standard approach. Secondly, an algorithm for continuous prediction of casing wear at the end of operation was developed. The predictive algorithm resulted in consistently more accurate estimates in relation to the final value throughout the operation. With variations in input parameters and consecutive casing wear of this magnitude, well integrity cannot be ensured during operation without application of real-time data. The failure to maintain well integrity demonstrates the necessity of the proposed approach.