A new way of reservoir management is dawning at the horizon - intelligent reservoir management utilizing continuous data from intelligent wells and/or smart fields. Even though there are many different buzz words for this new technology, they all lead to the same - managing a reservoir in REAL TIME or close to REAL TIME. Real time usually means to react to an event as it happens or within a short time lag. In the petroleum industry real time is for sure different. This "short time lag" can be hours, days or even weeks, which of course depends highly on the objective itself. Integrating real-time data into a reservoir management work flow and turn the data into value is a complex task. The bottle neck for the data flow right now is the transfer of the real time data - measured with a secondly and minutely time increment and stored on real time server - to the engineers' desktops in a clean and timely useful fashion. This paper will show ways how to provide a continuous (24/7) flow of clean data to the engineers' desktop as a first step for the intelligent reservoir management. It will be shown that the implementation of a smart field rises or falls with the ability to provide the data to the knowledge worker - the petroleum engineer. Since the data is coming into the database, let's say every hour or every other day, the engineer is not able to check this data for discrepancies. Therefore, intelligent reservoir management needs an alarm system to inform the engineers about any under performing or critical condition of a well or the reservoir itself. Another important aspect is the integration of the standard petroleum engineering tools, like Decline Curve Analysis, Material Balance, IPR curves, Reservoir Simulation, etc., into this work process. Now an Inflow Performance Relationship Curve does not only get data every other month, but every other day. This gives the engineer completely new opportunities, e.g. monitoring the permeability impairment over time. Well tests are usually a snapshot in time, but with a continuous surveillance of the reservoir parameters, the development of, e.g., the skin can be followed over time and actions can be taken in time - predictive maintenance. Neural Networks and Genetic Algorithms are other powerful tools in the real time environment, handling such a large amount of data. A Neural Network learns on the gathered data and detects their underlying relationships - the more data, the better. Afterwards, the Neural Networks can be used for predictions (predictive data mining) - for instance predicting sand production. This approach gives the engineer time to react, and prevents the equipment from harm. This work and the methodology it implies, provides a straight forward way of integrating real time data into a reservoir management process and how to gain value from the information provided by a continuous data stream. Introduction The aim of the smart field - the digital oil field of the future - is to automate as many tasks as necessary to achieve an increase in net present value of an asset. It is not only an increase in oil- or gas production, it is also a reduction of costs. For instance, a careful monitoring of the voidage ratio or the static reservoir pressure calculated with a material balance model can be used for a better allocation of the limited water volumes for injection. Many different papers and articles have been published about intelligent wells and smart fields. These papers focus mainly on the hardware aspect than how to use this data and convert them into value. De Jonge and Stundner1 showed a possible way of how to use high frequency data and turn them into economic value by using data mining technology. The big advantage of their approach is the generality so it can be implemented in any type of asset. The SPE "Real-Time Optimization" Technical Interest Group2 (TIG) gave a very good overview of how to use and gain value from high frequency measurements. Saputelli et.al.3 introduced the self-learning reservoir management. This approach is, as well, general enough to be used in any type reservoir to optimize its performance. Of course other papers are published in this area, but they deal with highly focused applications, which are not generally enough to be applied within every kind of reservoir.
TX 75083-3836, U.S.A., fax 01-972-952-9435. AbstractMethods of Artificial Intelligence like Back-Propagation Neural Networks (BPNN) have become popular software tools to predict permeability and porosity from well logs during the last several years.Similar to Multiple-Linear Regression models, Back-Propagation Neural Networks are trained with a set of target values from core measurements.The Self-Organizing Map (SOM) Neural Network method applies an unsupervised training algorithm. Until now this approach has mainly been applied for clustering purposes only, not for predicting reservoir properties.In a new application, SOM technology has been merged with statistical prediction methods to derive the following types of information from well logs and core measurements in one step: 1) Synthetic lithofacies system (clustering) 2) Porosity and permeability (prediction) SOM technology also provides a data visualization tool which allows evaluating relationships between input variables (well logs) and output variables (reservoir properties).SOM models can also be combined with BPNN in order to subdivide the entire set of well log patterns into different lithofacies and run individual BPNN models for each facies.Application of this method showed increase in prediction accuracy and significant timesavings. This paper should be viewed and printed in color.
TX 75083-3836, U.S.A., fax 01-972-952-9435. AbstractCurrent levels of Reservoir Surveillance technology associated with intelligent well completions, such as fibre optics and permanent downhole gauges, create an increasing flow of data.Conventional routine Reservoir Surveillance tools do not help the knowledge worker anymore to cope with highfrequency real-time data. Overloaded with data handling work, the knowledge worker in our industry is not capable to reveal the great potential inherent in this data.A radical different work process and new applications of Data Mining technologies are presented to support the industry's next goal -The Smart Field.A learning Data Mining approach is presented to detect discrepancies from expected trends and patterns. These trends and discrepancies are then translated into business rules to enable the closed-loop control of oil and gas assets.Lessons learned are presented and necessary future developments are identified.
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.