Summary This paper presents a data-driven methodology to predict calcium carbonate (CaCO3)-scale formation and design its inhibition program in petroleum wells. The proposed methodology integrates and adds to the existing principles of production surveillance, chemistry, machine learning (ML), and probability theory in a comprehensive decision workflow to achieve its purpose. The proposed model was applied on a large and representative field sample to verify its results. The method starts by collecting data such as ionic composition, pH, sample-collection/inspection dates, and scale-formation event. Then, collected data are classified or grouped according to production conditions. Calculation of chemical-scale indices is then made using techniques such as water-saturation level, Langelier saturation index (LSI), Ryznar saturation index (RSI), and Puckorius scaling index (PSI). The ML part of the method starts by dividing the data into training and test sets (80 and 20%, respectively). Classification models such as support-vector machine (SVM), K-nearest neighbors (KNN), gradient boosting, gradient-boosting classifier, and decision-tree classifier are all applied on collected data. Prediction results are then classified into a confusion matrix to be used as inputs for the probabilistic inhibition-design model. Finally, a functional-network (FN) tool is used to predict the formation of scale. The scale-inhibition program design uses a probabilistic model that quantifies the uncertainty associated with each ML method. The scale-prediction capability compared with actual inspection is presented into probability equations that are used in the cost model. The expected financial impact associated with applying any of the ML methods is obtained from defining costs for scale removal and scale inhibition. These costs are factored into the probability equations in a manner that presents incurred costs and saved or avoided expenses expected from field application of any given ML model. The forecasted cost model is built on a base-case method (i.e., current situation) to be used as a benchmark and foundation for the new scale-inhibition program. As will be presented in the paper, the results of applying the preceding techniques resulted in a scale-prediction accuracy of 95% and realized threefold cost-savings figures compared with existing programs.
This paper will present a statistical risk based approach to proactively predict casing leaks using Electromagnetic (EM) corrosion logs. The corrosion growth of downhole casing hotspots (areas likely to develop casing damage) is monitored to develop expectancy calculations of a typical well's remaining life. Currently the predominant technology used for casing integrity measurement and monitoring are the EM corrosion logging tools. While this technology has provided a step-change in the ability to measure and monitor corrosion, the findings are not usually conclusive and need to be integrated with other data to make qualitative assessment. This is largely due to the nature of the tool's output where averaging is used to assess metal loss rather than direct measurement of the spot where metal loss is taking place. In other words, 50% average metal loss could mean a failure if one part of the casing is completely gone and the other is intact; or 50% metal loss is distributed evenly across the 360 degree circumference of a casing with no leak. This wide range of possibilities and uncertainty has made it extremely challenging to both interpret and analyze EM corrosion logging data. Establishing consistent criteria to classify the corrosion severity and confidently decide on the need to workover the well or not is a challenge to all field operators worldwide. A probabilistic approach was introduced to improve EM corrosion logs' data interpretation. More than five hundred data points were collected and statistically analyzed to build a probability of failure model as a function of EM average metal thickness loss. These models were used to delineate a dynamic safe window of the average metal loss value across multiple casings.
The increase in water production from mature fields calls for continuous monitoring of wastewater disposal (WWD) system capacities. The three main components of any WWD system are discharge pumps, pipeline's network and disposal wells. Situations may arise when the full system capacity is constrained by the flow performance of one or a combination of these components. Subsequently, produced water volumes are reduced by closing high water cut oil producers. The grave and critical importance of such systems necessitates proactive diagnosing and forecasting of WWD performances.Plant-A injects 45 thousand barrels per day (MBD) of wastewater into five disposal wells. Historical system performance showed an increase in the pump discharge pressure that has nearly reached its maximum operating limit. A variation in injection performance among all the five disposal wells was observed. This variation is attributed to the different completion designs and reservoir inflow performances of the five subject wells. A 14 km flow line of varying sizes connects the disposal wells to two parallel discharge pumps. These pumps are designed to inject the abovementioned rate at a discharge pressure of 800 psi at current conditions.A full system analysis of Plant-A disposal network, utilizing hydraulic simulation techniques, was performed to model the current injection performance. The goal of this comprehensive study of the disposal system was to explore all possible root causes of the discharge pressure increase. In addition, the hydraulic simulator forecasted several feasible and recommended scenarios to eliminate this high discharge pressure.The five disposal well's inflow and outflow performance relationships were generated and calibrated using actual injection data. Network flow analysis was performed to account for both frictional and gravitational pressure losses in pipelines and flow restrictions using sophisticated software. This high accuracy model improved the decision making process and an optimized solution was put into effect based on the simulation results.
Continuous monitoring of wastewater disposal (WWD) systems capacities is necessary given the increase in water production from mature fields. The grave and critical importance of such systems necessitates proactive diagnosing and forecasting of WWD system performances. The precise outcomes of hydraulic simulation have proven reliable in forecasting operational scenarios related to WWD systems. The accurate predictions of hydraulic simulation, coupled with identified time frames and monetary investments, pave the way for accountable and successful managerial decisions.Plant-A injects wastewater into five disposal wells at a given capacity. Historical system performance showed an increase in the pump discharge pressure that has reached near its maximum operating limit. Future production demands entail the need to avail extra injection capacity to this WWD system.A full system analysis of Plant-A disposal network, utilizing hydraulic simulation techniques, was performed to model the current injection performance. The inflow and outflow performance relationships for five disposal wells were generated and calibrated using actual injection data. Network flow analysis was performed to account for both frictional and gravitational pressure losses in pipelines and flow restrictions using sophisticated software. This comprehensive study of the disposal system was aimed at exploring all possible root causes of the discharge pressure increase. In addition, the hydraulic simulator forecasted several feasible and recommended scenarios to eliminate this high discharge pressure. The contemplated scenarios were tapping into a nearby disposal system (either from the pumps suction or through transfer lines to the close by disposal system) and installing horizontal pump systems at the injection wellheads. This paper will present in depth the hydraulic simulation results of the above-mentioned scenarios and highlight their feasibility as viable short-term options. In addition, a detailed account of the associated time periods and respective costs of conceived short-term plans is shown. This high accuracy model improved the decision-making process, and an optimized solution was put into effect based on the simulation results.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.