A computerised model for predicting time to failure (survival analysis) of components for example of an electrical submersible pump.

摘要

A statistical methodology is disclosed to provide time-to-event estimates for oilfield equipment utilising Kaplan-Meier and Cox modelling approaches, Lunn-McNeil analysis of competing risks and Akaike Information Criterion (AIC) to select the best approximation model. A method according to the present invention extracts unbiased information from equipment performance data and considers parameters interactions without recourse to data thinning. The analysis explicitly accounts for items of equipment that are still operational at the time of analysis. A method according to the present invention may also be utilised to apply survival analysis to any oilfield equipment components where time-to-event information has been recorded. The method of the present invention allows comparative reckoning between different components present in the system comprising several or many individual components and allows survival analysis of these components either individually or simultaneously, i.e., in the presence of other components.