Improved Field Production Planning and Cost Optimization Using Predictive Analytical Choke Performance Model and Intelligent Alarms in an Integrated Platform

Abstract

One of the critical aspects of production optimization and planning is to meet the production targets or to meet some operational requirements such as workovers or maintenance activities. This paper demonstrates how an advanced integration in a digital platform, coupled with a predictive analytical model for choke performance and intelligent alarming, can significantly help asset in production planning and cost optimization while accurately regulating the field rates.

First, the bulk of well-test data from corporate databases is integrated into an advanced digital platform with an automated well-test validation workflow. The workflow output provides the choke tuning factors for each test while validating the well-test parameters. The digital platform provided the initial data check to ensure the validated tests with choke tuning factors were processed for further regression analysis. The network model in the digital platform for the entire asset was run for a predefined set of iterations to generate the representative choke tuning factors for each well, based on production test parameters and flow line pressure constraints. The regression analysis output was used to predict the choke sizes for different inflow performance rates and various operating wellhead pressures and vice versa.

The predictive choke analytical model outputs were utilized to predict the choke size for a set of well parameters, such as rates and wellhead pressures, based on historical well performance. The choke sizes predicted could be used to identify preferred wells in an area to be controlled to achieve production targets, minimizing the operational effort and time. The predictive choke model with intelligent alarm feature provided users instantaneous insight into underperforming and overperforming wells, assisting them to take further actions in an effective way. The other intelligent alarms worked in combination to detect lifting problems associated with wells more efficiently, such as the liquid loading intelligent alarm. The predictive model was also valuable for efficient production planning in terms of setting the quarterly well allowable, choke sizes, or performing field capacity tests to meet the business production target on field & well level and to analyze short-term and medium-term forecast cases using an automated reservoir integration workflow in the digital platform. This was helpful in planning ahead of time for future operations and saving a significant amount of time and effort for engineers and the operation team.

This specialized approach of predictive choke performance modeling in a digital platform provided asset a robust tool to plan and optimize their field production while leveraging the power of data-driven digital platforms consisting of closed-loop automated engineering workflows. The accuracy of prediction proved significant cost optimization and proactive planning, where the bulk of data was handled effectively and efficiently to identify production optimization opportunities and field bottlenecks.