Converting data to actionable information through continuous oil production monitoring is a fundamental part of any production optimization strategy. The development of Intelligent Field technology has remarkably contributed to the upgrading of production surveillance framework and provided an extended access to real-time data. This same technology is still in its infancy when it comes to multiphase mass metering and field practicality issues. As for conventional fields where the unavailability of continuous data flow is not considered out of norm, the high uncertainty in oil production rate estimation and allocation is very well expected. The main source of this uncertainty is the reliance on sporadic welltest data and empirical multiphase flow correlations to allocate liquid production rate.

Critical and subcritical multiphase flow choke performance is predicted using well-known correlations that are based on specific datasets characterized by a specific field or hydrocarbon type. Case studies where those correlations are matched with different production data and used later to predict the choke performance are present in the literature. Yet, the oil industry is faced with many challenges because of the limited accuracy of those predictions. The complexity of multiphase flow behavior and the irregularities in operational conditions can explain such low capability of those correlations particularly on field data.

Artificial intelligence (AI) tools and techniques for so-called artificial neural networks, fuzzy logic and functional networks were employed to develop data-driven oil flow rate computational models for both critical and subcritical flow conditions. These AI models were trained and tested exploiting 595 production rate tests from 31 different wells. The prediction results showed a strong correlation with actual field data and promised a reliable tool/methodology to estimate oil flow rate as a function of operational conditions and choke size. This paper presents an engineering look at the inclusion of AI data-driven models in the production surveillance system to enhance welltest data validation and reduce the uncertainties in production allocation.

You can access this article if you purchase or spend a download.