As the field matures and overall field production decline, an accelerated and advanced selection of existing wells for workovers and sidetracks would be critical to meet with the increasing demand for production. Traditionally, an intensive effort is required to identify the right candidates and to ensure the technical and economic success of well interventions, infill-drilling locations and sidetrack locations. An advanced workflow is developed to automate the repetitive and less added value tasks such as data gathering and validation. The data set included historical production performance, static and dynamic reservoir and fluid properties, events and issues encountered within the evaluated wells and regions. The proposed solution allowed an integrated assessment of production enhancement opportunities through various consistent analytic computations, as well as machine learning techniques including Bayesian networks and time-series forecasting models. The automated process generates a comprehensive list of future well interventions including sidetrack candidates, infill-drilling locations and behind casing opportunities with an advanced scoring system and several technical (production performance and reserves) and financial KPIs (i.e. net present value and unit technical cost). Several dashboards built and adjusted with the involvement of various company departments. Lastly, highly ranked opportunities are incorporated into the business plan in accordance to field development targets as well as the availability technical resources (rigs, materials, well availability). The developed solution was tested and validated in a giant mature carbonate field with over 700 well strings in Offshore Abu Dhabi. The solution identified 20 times more feasible opportunities than the typical multidisciplinary team review in 75% less time duration. The automated workflows considered re-evaluating and selecting prime candidates with a reduced risk of failure, therefore improving the technical and economic value by 34%. The workflows are scheduled on daily basis giving a time-dependent assessment and expert monitoring system, which can also notify the operator when problems encountered. Instead of computationally heavy traditional numerical simulation models, the assessment of a large number of well count can be done within hours instead of months. The combination of physics and machine learning based models lead to the development of automated workflows to rank and determine the best candidates via a successful cost optimization and production enhancement.

The process of short-term water and gas flood optimization attempts to increase short term profit, while maximizing long term net present value (NPV) of the field. The characteristics of each production system would dictate how this process is achieved. Fields where the available producer well potential is significantly larger than the production quota could have infinite possible scenarios of production and injection well settings that would satisfy the field and reservoir production targets. But which of these scenarios maximize the long term NPV? This paper explains the framework being implemented in ADNOC to streamline the optimization workflow, which runs both physics and data driven models, honours all constraints, and covers the associated processes from model maintenance, to calculation, execution, and monitoring. The workflows are orchestrated with a series of in-house interconnected digital solutions. This framework has been implemented in 5 production systems undergoing pattern injection of water, gas, and CO2. The associated digital solutions are well adopted by the asset teams. Ability to optimize production and injection together has allowed the asset to focus on increasing injection capacity as the pattern, sector, and reservoir voidage constraints were identified to be the main constraint to production deliverability. The optimization scenario management and associated workflows have shown to deliver a gain of 1-3% of production by synchronizing the reservoir management process with the production operations business rhythm. The solutions have delivered so far more than 150 MM$ in value.

Asset management success is accomplished when the integrated production system is operating close to its intended potential. Continuous awareness of wells and facility conditions are key factor in the realization of designed capacity. In contrast, unknown status and conditions can severely limit production capacity. The rise of instrumentation technologies over the last four decades have created new opportunities to understand well and reservoir behavior. However, despite of being proved as a cost-effective surveillance initiative, remote monitoring is still not adopted in over 60% of oil and gas fields around the world. Understanding the value of data through machine learning techniques is the basis for establishing a robust surveillance strategy. The objective of this paper is to develop a data-driven approach, enabled by Artificial Intelligence (AI) methodologies including machine learning (ML), to find optimal operating envelope for gas-lift wells. The process involves building ML models for generating instantaneous predictions of multiphase flow rates and other quantities of interest, such as GOR, WCT, using real-time sensor data at the surface, historical performance, and sporadic test data. Additionally, forecasting models were developed for generating short-term (30 days) forecast of cumulative oil, water, gas, and liquid production, multiphase flow rates, WCT, GOR, and reservoir pressure. Using time-series forecasting models, a sensitivity analysis was performed to generate short-term well response for a selected number of combinations of choke settings, and gas injection rates. Sensitivity analysis provides 2D maps of well response highlight an operating envelope, which are proposed to be combined with physical and operational constraints to arrive at optimal operating conditions, which may effortlessly add 2.5% net profit from optimum gas-lift alocation. The results of this work show encouraging results, and demonstrate value that AI-enabled methodologies can provide in instrumented wells by enabling automated workflows for virtual metering, production allocation, short-term production forecasting, and deriving optimal operating conditions. The developed AI methodology has tremendous potential of integration in an end-to-end workflow of autonomous well control by utilizing available data to produce easy to update ML models, with little to no human intervention.

Production forecast is an important part of field development planning using for business planning and economic evaluation of an oil and gas field. The conventional approach to production forecasting includes a bunch of methods from decline curve analysis to hydrodynamic modeling; however, these methods have certain limitations of use. The objective of the study is to investigate the possibility of application AI/ML methods in classification of wells by their historical behavior, and to use this information to predict performance of existing and future wells. The core method of the tool is K-means clustering based on wells production profiles. There are several parameters which values have been taken at the certain time-steps, for example, oil rate, watercut, GOR, flowing bottom-hole pressure, etc. Each parameter's monthly record is combined with reservoir properties to define the input vector for each well. These vectors are assigned to a specific cluster number. The defined clusters can then be assigned to wells with shorter production history in order to predict their future performance. The proposed using pattern recognition of production is useful for the identification of possible neighbor wells influence and tunes production prediction. The method was tested on an oil field in Abu Dhabi with over 50 wells. The results have shown that there is a good predictability for tested dataset. The obtained clusters identified the production performance in existing wells and the production forecast in the nearest future. The method introduces a new approach to wells clustering based of production profiles only. The approach is tested on Abu Dhabi field first time and practical implementation of the method is showcased. The limitation is that it is applicable to wells that have long production history with stable reservoir performance.

Production and Injection rate target optimization plays an important role in waterflooded field management in order to ensure hydrocarbon recovery. In line with ADNOC Digital transformation and waterflood excellence initiatives CRM and Optimization technology has been progressed to maximize opportunities in oil recovery increase. The optimization means that producing well delivers a maximum amount of oil with minimal water production along with maintaining proper Voidage Replacement Ratio (VRR) to support reservoir pressure. To reach such goal, the optimization procedure needs to run multiple rate scenarios to calculate the objective function value. The conventional way is to perform multiple runs on simulation model, which can be very time-consuming. The data driven approach described in this paper suggests faster and convenient methodology to solve this problem. The process applied to this approach consists of data preparation/ data cleansing stage, CRM (Capacitance Resistance Model) and optimization procedure based on the objective function with a penalty to imbalanced VRR at the pattern level. The CRM algorithm can calculate fraction of injection distributed from each injecting well to connected producing wells at any timestep. These calculated injection allocation factors are considered in the rate optimization procedure in order to define optimal injection and production rates along with balancing of VRR at the pattern level. The method also considers 3-phase flow across wells and reservoir intervals. The objective function calculates overall profit from oil production, costs for water and gas handling, and the penalty for the production injection difference at the producing well level. At the end, the output of this optimization process is to recommend production and injection rates targets for each well and short term forecast of the production based on fractional flow model. The data driven approach shows quite good efficiency in terms of time and efforts, the injection allocation factors based on CRM model are comparatively same as it is calculated in streamline simulation model but with better granularity at the pattern level. The optimization procedure works quite fast, and the results have shown decrease of water production rate and increase of recovery factor due to maintaining VRR close to the target level. The data driven approach described in the paper implements a new way to apply CRM in fields with waterflooding and gas injection with the enhancement of managing 3-phase flow. The in-house developed optimization function and its implementation is a novel approach in terms of practical application to the fields in Abu Dhabi area.