No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Physics-Informed Neural Networks and Capacitance-Resistance Model: Fast and Accurate Oil and Water Production Forecast Using End-to-End Architecture
Abstract
Capacitance-Resistance model (CRM) has been a useful tool for fast production forecasts for decades. The unique combination of simplicity and physics-based nature in this data-driven approach allowed it to stay as an object of scientific interest and get its own place among other types of models capable of giving predictions on flow rates, such as full-scale 3D reservoir models. However, the model simplicity, assumptions, and limitations does not allow wide application of a conventional CRM in complex field cases. A vast majority of studies on CRM are about overcoming its limitations by introducing new coefficients, modifying the analytical form of the equation, etc. Integrating CRM with rapidly developing artificial intelligence (AI) methods seems to be a logical continuation of model evolution.
Recently introduced Physics-Informed Neural Networks (PINN) can preserve CRM's governing equations and coefficients that gives some insights about wells and formations standing out from other popular machine learning and deep learning methods. Moreover, PINN type models give certain flexibility in the choice of architectures – it means that the model architecture can be changed in a way that may assist in solving different problems. Thus, we introduce end-to-end learning of neural networks (NN) while implying some physical constraints. It is intended to overcome one of the major limitations of CRM, which is obtaining predictions for oil and water production rates from total liquid. This way, the additional training of rough approximation fractional flow models that are either not suitable for the case or may require the knowledge of reservoir properties is not needed.
In this work, the well-known concept of Capacitance-Resistance models appears in a new form, which allows performing history matching rather rapidly, achieving robustness and forecasting liquid, oil and water production rates simultaneously. To test this new approach, several datasets (both synthetic and real) were used. The results obtained by PINN are compared to those obtained by a conventional CRM. By conventional we mean the analytical solution, which was modified by our research group to take into consideration common real field cases such as shut-in wells, workover operations, etc. by introducing dynamic characteristic coefficients [1].
Despite the significant progress over the last 50 years in simulating flow problems using numerical discretization of the Navier–Stokes equations (NSE), we still cannot incorporate seamlessly noisy data into existing algorithms, mesh-generation is complex, and we cannot tackle high-dimensional problems governed by parametrized NSE. Moreover, solving inverse flow problems is often prohibitively expensive and requires complex and expensive formulations and new computer codes. Here, we review flow physics-informed learning, integrating seamlessly data and mathematical models, and implement them using physics-informed neural networks (PINNs). We demonstrate the effectiveness of PINNs for inverse problems related to three-dimensional wake flows, supersonic flows, and biomedical flows.
We analyze how artificial intelligence changes a significant part of the energy sector, the oil and gas industry. We focus on the upstream segment as the most capital-intensive part of oil and gas and the segment of enormous uncertainties to tackle. Basing on the analysis of AI application possibilities and the review of existing applications, we outline the most recent trends in developing AI-based tools and identify their effects on accelerating and de-risking processes in the industry. We investigate AI approaches and algorithms, as well as the role and availability of data in the segment. Further, we discuss the main non-technical challenges that prevent the intensive application of artificial intelligence in the oil and gas industry, related to data, people, and new forms of collaboration. We also outline three possible scenarios of how artificial intelligence will develop in the oil and gas industry and how it may change it in the future (in 5, 10, and 20 years).
Water is one of the most important commodities in the world and plays an essential role in the hydrocarbon industry. With increased agricultural production, rapid industrialization, population growth and climate change, the world is facing an extreme water crisis in many regions. Coupled with a surge in energy demand and consumption, this has greatly influenced the hydrocarbon industry. With increasing stress on water resources, it is essential to examine how water is managed within the hydrocarbon industry and devise ways to utilize water more efficiently, especially within water scarce regions such as the Middle East.
Augmented by the recent activities in the oil and gas industry, it can be seen that an economical and efficient hydraulic fracturing job has become crucial for the successful development of unconventional reservoirs. However, exploitation of unconventional reservoirs is heavily water-intensive as compared to conventional reservoirs. In this study, a comprehensive investigation that deals with quantification of changes with respect to the variation in prime contributors within a traditional fracture design process is presented.
To understand the significance of key design parameters, factors that affect productivity within typical Middle Eastern shale gas reservoirs were analyzed through simple constrained cases. Investigations reveal that parameters such as fracture aperture, natural fracture distribution, fracturing fluid viscosity and Young’s modulus are crucial to the overall production and water requirement. Furthermore, an outline for resource management within a traditional fracture design process is presented along with potential challenges for the region. Enhancing existing methodologies and incorporating key parameters highlighted within this study can contribute to the overall value chain. In addition to ultimately assisting in the verification of modern best practices, this investigative approach will create a paradigm for future studies for the region to assist in a simplistic prediction of fracture propagation and associated response to augment water usage.
A well-functioning supply chain is a key to success for every business entity. Having an accurate projection on inventory offers a substantial competitive advantage. There are many internal factors like product introductions, distribution network expansion; and external factors such as weather, extreme seasonality, and changes in customer perception or media coverage that affects the performance of the supply chain. In recent years Artificial Intelligence (AI) has been proved to become an extension of our brain, expanding our cognitive abilities to levels that we never thought would be possible. Though many believe AI will replace humans, it is not true, rather it will help us to unleash our true strategic and creative potential. AI consists of a set of computational technologies developed to sense, learn, reason, and act appropriately. With the technological advancement in mobile computing, the capacity to store huge data on the internet, cloud-based machine learning and information processing algorithms etc. AI has been integrated into many sectors of business and been proved to reduce costs, increase revenue, and enhance asset utilization. AI is helping businesses to get almost 100% accurate projection and forecast the customer demand, optimizing their R&D and increase manufacturing with lower cost and higher quality, helping them in the promotion (identifying target customers, demography, defining the price, and designing the right message, etc.) and providing their customers a better experience. These four areas of value creation are extremely important for gaining competitive advantage. Supply-chain leaders use AI-powered technologies to a) make efficient designs to eliminate waste b) real-time monitoring and error-free production and c) facilitate lower process cycle times. These processes are crucial in bringing Innovation faster to the market.
Capacitance resistance models (CRMs) comprise a family of material balance reservoir models that have been applied to primary, secondary and tertiary recovery processes. CRMs predict well flow rates based solely on previously observed production and injection rates, and producers’ bottomhole pressures (BHPs); i.e., a geological model and rock/fluid properties are not required. CRMs can accelerate the learning curve of the geological analysis by providing interwell connectivity maps to corroborate features such as sealing faults and channels, as well as diagnostic plots to determine sweep efficiency and reservoir compartmentalization. Additionally, it is possible to compute oil and water rates by coupling a fractional flow model to CRMs which enables, for example, optimization of injected fluids allocation in mature fields. This literature review covers the spectrum of the CRM theory and conventional reservoir field applications, critically discussing their advantages and limitations, and recommending potential improvements. This review is timely because over the last decade there has been a significant increase in the number of publications in this subject; however, a paper dedicated to summarize them has not yet been presented.
Nanoparticles are usually small enough that they can pass through the porous media without mechanically plugging the pore throats. However, physicochemical interaction between the nanoparticles and the pore walls can cause significant retention of nanoparticles. The objective of this paper is to provide theoretical equations based on DLVO theory to calculate the rate of deposition and release at different temperatures, ionic strengths, and pH values.
DLVO theory is used to understand the interaction between nanoparticles and rock minerals. Electrostatic interaction depends on the zeta potential of nanoparticles and pore surface. In this paper, an equation is developed to calculate zeta potential at different temperatures, ionic strengths, and pH values. The rate of deposition and release of Silica nanoparticles in a sandstone formation, where interaction energy profile has energy barrier, has been derived. To validate the theoretically calculated rates, a numerical model is developed to compare the theoretical calculations with experimental data.
Increasing ionic strength and temperature decreases the energy barrier height and hence increases the rate of deposition. The effect of pH on the rate of deposition depends on the location of environment pH with respect to the isoelectric point of nanoparticles and rock surface. For extreme values of pH, energy barrier exists and rate of deposition is low. However, when the pH of the solution is between the isoelectric points of nanoparticles and rock surface, the energy barrier decreases and the rate of deposition increases. The rate of deposition is time dependent with the rate decreasing as more rock surface is covered by nanoparticles. These theoretically calculated rate values are used in a numerical model of the advection-dispersion equation with source/sink term. Several experimental data have been perfectly matched with the model that validates the theoretical calculations of the rate of deposition.
The new mechanistic model for nanoparticles can be used to determine the fate of nanoparticles in porous media under different conditions of nanoparticle size, temperature, ionic strength, and pH. This model can help to understand the nanoparticles transport in porous media and effectively design nanoparticles fluid for injection into oil and gas reservoirs.
The threat of climate change has elicited divergent climate policy responses from the world's major oil multinationals, splitting the oil industry into two factions. This article analyzes the causes and consequences of this split in the oil industry. First, it demonstrates that oil companies made divergent assessments of the market risks and opportunities related to climate change based on the scientific networks and policy fields in which they were embedded rather than on rational economic criteria. Second, it documents that although the climate policy split in the oil industry has had few effects on oil company operations, it changes the terms of debate over profitable corporate action on climate change, with significant material consequences for climate regulation and patterns of energy production. This analysis contributes to the debate between treadmill of production and ecological modernization theorists by highlighting the midrange processes of contestation shaping the long-term environmental trajectory of capitalism.
Artificial intelligence (AI) was introduced to develop and create “thinking machines” that are capable of mimicking, learning, and replacing human intelligence. Since the late 1970s, AI has shown great promise in improving human decision-making processes and the subsequent productivity in various business endeavors due to its ability to recognise business patterns, learn business phenomena, seek information, and analyse data intelligently. Despite its widespread acceptance as a decision-aid tool, AI has seen limited application in supply chain management (SCM). To fully exploit the potential benefits of AI for SCM, this paper explores various sub-fields of AI that are most suitable for solving practical problems relevant to SCM. In so doing, this paper reviews the past record of success in AI applications to SCM and identifies the most fruitful areas of SCM in which to apply AI.
Proposal
This paper presents a new procedure to quantify communication between vertical wells in a reservoir based on fluctuations in production and injection rates. The proposed procedure uses a nonlinear signal processing model to provide information about preferential transmissibility trends and the presence of flow barriers.
Previous work used a steady-state (purely resistive) model of interwell communication. Data in that work often had to be filtered to account for compressibility effects and time lags. Even though it was often successful, the filtering required subjective judgment as to the goodness of the interpretation. This work uses a more complicated model that includes capacitance (compressibility) as well as resistive (transmissibility) effects.
The procedure was tested on rates obtained from a numerical flow simulator. It was then applied to a short timescale data set from an Argentinean field and a large-scale data set from a North Sea field. The simulation results and field applications show that the connectivity between wells is described by model coefficients (weights) that are consistent with known geology, the distance between wells and their relative positions. The developed procedure provides parameters that explicitly indicate the attenuation and time lag between injector and producer pairs in a field without filtering. The new procedure provides a better insight into the well-to-well connectivities for both fields than the purely resistive model.
The new procedure has several additional advantages. It can be applied to fields in which wells are shut-in frequently or for long periods of time,allows for application to fields where the rates have a remnant of primary production, andhas the capability to incorporate bottom hole pressure data (if available) to enhance the investigation about well connectivity.
Health, Safety and Environment (HSE) operations can be greatly enhanced by reducing the uncertainty inherent in quantifying human activity. One way to do this is by using artificial intelligence (AI) techniques, in particular natural language processing (NLP), on HSE data to identify hidden trends, uncover and quantify textual data and, ultimately create a smart system that can alert users to risk before that risk is realized. This allows an organization to better target their resources, such as safety coaches, more effectively to prevent an adverse event, protecting vital equipment and potentially saving lives. One important aspect inherent in this process is the need to establish trust in the AI system among the users. This is especially the case during data collection, system rollout and user adoption.
In order to realize the goal of an AI driven HSE workflow, a system was implemented which allowed for a) the easy collection, structuring and preprocessing of data associated with the performance of management observations, and for b) the development and implementation of a robust set of AI tools that allowed the users to enhance their existing workflows to better be able to identify, quantify and address HSE risk. This created a new set of leading indicators for HSE awareness. While management observations were the first set of data considered as they were the most robust data set, the adaptable system was constructed in such a way that work orders, audits, and near misses could be easily incorporated in the future.
This study is part of a pilot project with Petroleum Development Oman (PDO) for IHTIMAM, which is a process that creates a safety partnership between the workforce and management that continually focuses everyone's attention and actions on their own and others daily safety ‘behaviour’. It was unique and innovative, because it took an auto ethnographic approach in the analysis of the AI, drawing on a team with multidisciplinary expertise.
A novel, comprehensive approach for HSE data collection and development of industry specific AI models to is presented. In addition, a novel approach was identified to leverage user adoption and use case in terms of both method and application.
Traditionally, a significant amount of time is invested in producing the most optimal drilling schedule to deliver the targets considering various constraints and changing priorities. This paper demonstrates how Artificial Intelligence (AI) can generate an optimal Rig Schedule while improving on conventional planning techniques. This includes adding additional value through increase in production, freeing up assets and reduction in fuel consumption, driving cost reductions further enabling supply-chain debottlenecking.
This paper presents a real-world application and solution of the general set of optimization problems such as the Knapsack Algorithm and Vehicle Routing Problems (VRP) for optimizing rig mobilization with added real-world complexities and constraints. A dynamic system allows more than just scheduling against demand/supply as it also self-calibrates the schedule through new real-time requests and real-time situation analysis based on location, availability, and other relevant constraints.
Based on in-house case studies conducted and compared with traditional approaches for rig scheduling and optimization, the presented solution can reportedly achieve a 99% reduction in time needed for generating key results. Compared to conventional drilling scheduling methodologies, there are no or minimal white spaces for the resource allocation strategies presented by the AI solution with a potential reduction in the asset utilization (with a reduction of 5%) along with being able to reduce total distance traveled and the fuel burned (carbon emissions) assuming standard mobilization patterns based on historical data, with a reduction ranging between 11-24% as a minimum depending on the scenarios selected.
This case study provides a novel approach to the scheduling of rigs that leverages artificial intelligence for complex fleet and schedule management that provides an opportunity to generate best plans to meet KPI's with significant reduction in assets required and fuel burned (energy efficiency) during mobilization; but also provides a higher level of input into operations and could in future provide real time input into operational activity plans minimizing overall costs and input to streamline supply chain from layers of conservatism.
We disclose a new-age field-scale production forecast model that handles complex treatment of wellbores during their life cycle. Predictive production models have been an object of increased interest and research for a long time due to the need for a fast tool for forecasting production rates or choosing an optimal field development scheme. The existing approaches based on the material balance equation have several limitations and are not very applicable for real objects. Full-scale reservoir modeling is relatively slow and requires large computing resources. In this paper, we propose a proxy model based on advanced capacitance-resistance approach. The model predicts multiphase flow rates based on the available historical data of field production and information about well treatments. In addition, it provides preferable transmissibility trends, the presence of sealed or leaking faults, and the degree of dissipation between injector-producer well pairs. The advanced feature of the model is time-dependent weight coefficients, which have not been studied previously. They help in accounting the shut-in and workover periods and can be found during the optimization procedure simultaneously. Another feature is fast calculations due to a vectorized form of the model and application of modern optimization techniques. All these options allow modeling real oil fields with a large number of wells and a complex system of production control.
This paper explores two case studies indicating how artificial intelligence can be used to automate, predict, and optimize the complexities within offshore logistics. It focuses on Offshore Supply Vessel (OSV) fleet management with multiple points of delivery, spread out across geographic locations, each location needing specific cargo items within a given time window. The study indicates how this can free up assets and reduce fuel consumption, ultimately driving cost reductions and enabling supply-chain efficiency by redeploying assets.
The first use case shows the effect of applying these methods within a ‘greenfield’ environment. That is, we assume that our constraint parameters are generally flexible or, at least, can be modified by the operator within some reasonable bounds. For example, we might consider that a vessel can depart from a port any day of the week rather than be bound to a specific day, or that deliveries can happen at night and not only during the day. Thus, the first use case is answering the question "what would the efficiency gain be if we allow enough flexibility in our operations to adapt to the output of the model?"
The second use case shows the effect of applying these methods with a ‘brownfield’ environment wherein our operational constraint parameters are not generally able to be modified to adapt to the model output. For example, we might have an assumption of the type that "port departures are only allowed on Thursdays." Thus, even if the model suggested that departing three days a week would lead to an optimal outcome, we would not allow ourselves to change from the set departure dates.
In recent years great interest has risen towards surrogate reservoir models based on data-driven methodologies with the purpose of speeding up reservoir management decisions. In this work, a Physics Informed Neural Network (PINN) based on a Capacitance Resistance Model (CRM) has been developed and tested on a synthetic and on a real dataset to predict the production of oil reservoirs under waterflooding conditions.
CRMs are simple models based on material balance that estimate the liquid production as a function of injected water and bottom hole pressure. PINNs are Artificial Neural Networks (ANNs) that incorporate prior physical knowledge of the system under study to regularize the network. A PINN based on a CRM is obtained by including the residual of the CRM differential equations in the loss function designed to train the neural network on the historical data. During training, weights and biases of the network and parameters of the physical equations, such as connectivity factors between wells, are updated with the backpropagation algorithm.
To investigate the effectiveness of the novel methodology on waterflooded scenarios, two test cases are presented: a small synthetic one and a real mature reservoir. Results obtained with PINN are compared with respect to CRM and ANN alone. In the synthetic case CRM and PINN give slightly better quality history matches and predictions than ANN. The connectivity factors estimated by CRM and PINN are very similar and correctly represent the underlying geology. In the real case PINN gives better quality history matches and predictions than ANN, and both significantly outperform CRM. Even though the CRM formulation is too simple to predict the complex behavior of a real reservoir, the CRM based regularization contributes to improving the PINN predictions quality compared to the purely data-driven ANN model. The connectivity factors estimated by CRM and PINN are not in agreement. However, the latter method provided results closer to our understanding of the flooding process after many years of operations and data analysis. All considered, PINN outperformed both CRM and ANN in terms of predictivity and interpretability, effectively combining strengths from both methodologies.
The presented approach does not require the construction of a 3D model since it learns directly from production data, while preserving physical consistency. Moreover, it represents a computationally inexpensive alternative to traditional full-physics reservoir simulations which could have vast applications for problems requiring many forward evaluations, like the optimization of water allocation for mature reservoirs.
Reservoir simulation and adaptation (also known as history matching) are typically considered as separate problems. While a set of models are aimed at the solution of the forward simulation problem assuming all initial geological parameters are known, the other set of models adjust geological parameters under the fixed forward simulation model to fit production data. This results in many difficulties for both reservoir engineers and developers of new efficient computation schemes. We present a unified approach to reservoir simulation and adaptation problems. A single neural network model allows a forward pass from initial geological parameters of the 3D reservoir model through dynamic state variables to well’s production rates and backward gradient propagation to any model inputs and variables. The model fitting and geological parameters adaptation both become the optimization problem over specific parts of the same neural network model. Standard gradient-based optimization schemes can be used to find the optimal solution. Using real-world oilfield model and historical production rates we demonstrate that the suggested approach allows reservoir simulation and history matching with a benefit of several orders of magnitude simulation speed-up. Finally, to propagate this research we open-source a Python-based framework DeepField that allows standard processing of reservoir models and reproducing the approach presented in this paper.
We present a robust control system and methodology for physics-informed artificial intelligence (PAI) used to optimize and improve oil recovery, demonstrated in the Yates Field operated by the Kinder Morgan CO2 company. The system consists of a robust control system (referred to as Dynamic Feedback Loop, or DFL) equipped with novel hydrocarbon sensors that measure oil concentration and other parameters continuously and simultaneously on a set of producing wells. The goal of this system is to optimize operational parameters (e.g. choke valve settings, injection rates) to reach specific target metrics of production (e.g. maximizing produced oil while minimizing produced gas).
The key element of our approach is the use of a multi-layer artificial neural network (deep neural network, or DNN, to be specific) that extracts physics-based parameters from the real-time measurements and predicts relevant parameters of the DFL control system. DNNs are prone to overfitting in training, making them ineffective in unfamiliar or challenging situations outside the training dataset. To overcome this problem, we have developed a physics-informed robust neural network technique, where the reservoir physics and sensor data are used to train DNN representations of the key physical parameters. Typically, only simplified physical models are developed using available geostatic or historical production data. Also, due to the dynamic nature of these systems, the accuracy of the models often changes over time. To improve predictive capability of the model, we combine the DNN with the system-theoretic robust control concepts based on physics with a model uncertainty formulation. The concept was first validated using a combination of simulations, isolated sensor data and analyses based on sets of historic production data.
A study using historic production data on Kinder Morgan's Yates Field Unit (YFU) 4045 Pilot (3 producing wells) indicates application of the DFL system results in an increase in cumulative production of up to 35% per year, compared to what is obtained through a traditional (fixed-point) control system. Currently, the DFL is being field-tested on a different set of wells in the Yates Field, instrumented with the novel hydrocarbon sensors that generate continuous and simultaneous production data.
Replacing all analogue sensors in the oil field is very costly and normally only a fraction of them is done. Currently, there is no cost-effective method to efficiently, reliably and accurately capture analogue meter readings in a digital format. Operators are then left with only two options: either replace them with digital (high capex) or continue with manual gathering (high opex). This paper shows how computer vision and artificial intelligence was used for the first time to capture analogue field gauges data with dramatic reduction of cost and increase reliability.
This unique solution was implemented in the Cheleken Oil field, Caspian Sea, Turkmenistan. In the offshore platforms, only low-cost cameras were necessary, and gauges were identified using QR codes. During the field trial, operators were only required to take pictures of the gauges at a given interval of time and upload the photos to the application. After an innovative process of calibration, the acquired images were processed using artificial intelligence and deep learning computer vision.
Routine manually gathered data was compared with data collected using this solution with the following observations made: Date/time: Operators usually round time. The solution described records time on the captured pictures automatically.Value: Manually gathered data is subject to reading, typing and transcription errors. This solution has no error (provided a good calibration is done).Data Modification: Data gathered automatically with this solution has no human intervention. Therefore, is not subject to alteration, copying or duplication.Data collection with pictures was completed in 1/10th of the time that manual processes take.The business benefits from quicker operator rounds with improved accuracy in meter reading data, and time stamps. The administrative burden for operators of filling in extensive spreadsheets which are prone to error was reduced, this allowed them to collect more meter readings or be reassigned by management to more important scopes of work that bring greater value to the business. Once more it was proved that "a picture is worth a thousand words ".
This solution offers an excellent opportunity for digitizing the marginal section of the field and provides a unique way to turn all analogue data into digital with a very low cost of implementation, on an infinitely scalable platform that is vendor agnostic and simple to manage.
We introduce physics-informed neural networks – neural networks that are trained to solve supervised learning tasks while respecting any given laws of physics described by general nonlinear partial differential equations. In this work, we present our developments in the context of solving two main classes of problems: data-driven solution and data-driven discovery of partial differential equations. Depending on the nature and arrangement of the available data, we devise two distinct types of algorithms, namely continuous time and discrete time models. The first type of models forms a new family of data-efficient spatio-temporal function approximators, while the latter type allows the use of arbitrarily accurate implicit Runge–Kutta time stepping schemes with unlimited number of stages. The effectiveness of the proposed framework is demonstrated through a collection of classical problems in fluids, quantum mechanics, reaction–diffusion systems, and the propagation of nonlinear shallow-water waves.
Numerical reservoir simulation offers the best representation for reservoir fluid flow. However, uncertainty in assigning and distributing static parameters between wells adds a dimension of weakness to its implementation and cultivates modeling doubts. Capacitance-resistance modeling is one of the recent evolving technologies that provides opportunities to study fluid dynamic signatures that can help unveil uncertainties for some aspects of static reservoir parameters such as fractures and faults.
Capacitance-resistance modeling (CRM) is a tool that relies on signal processing in associating actions (i.e. injection and drawdown) to reactions (i.e. production). This is accomplished through means of multivariate nonlinear regression. There are many documented limitations to this tool that reduce the model reliability, or prevent it from attaining an acceptable match. Capacitance-resistance modeling limitations result from varying model parameters, or missing input data. Examples of these limitations include changes to the number of active producers, significant change in well productivity index (e.g. well workover), high variation of fluid compressibility, and presence of aquifer support. Most reservoirs will encounter many of these limitations which makes it important to find a rectifying solution.
In this paper, CRM limitations are addressed and a new modeling equation is introduced to produce a flexible capacitance-resistance model. Shutting-in producers, or introducing them, changes the number of active production wells. This limitation is solved by modifying the injection rates in a way that honors the mass balance and keeps the modeling weights constant. Changes in a well's productivity index are addressed by considering the well as two separate wells (before/after change), while the presence of aquifer support is engaged by adding a pseudo injection well, to mimic influx support, and adjusting the injection rate to reach the best possible match quality. As for the new modeling equation, it performs very well when compared to current CRM equations, has fewer variables, and has more applications due to the incorporation of a drainage control-volume that acts as a filter to all supporting streams. This new formulation also gives clearer indications to noncontributing injectors; the information from which can be used to detect the presence of faults and fractures.
Hydraulic fracturing is a common applied stimulation method to develop the tight gas resource which is widely distributed all over the world. A lot of field measurement and experimental observation have shown the hydraulic fracture (HF) complexity in the environment of tight sandstones. This complex fracture is strongly influenced by the natural fractures (NF) which are inherent in the tight gas formation. It was documented that, as a result of interaction, the hydraulic fracture will cross, dilate or slip into the natural fracture. Though some researches have been done, the interaction mechanism is still not fully understood. The lack of knowledge in this topic calls for further research in it.
With this study, two neural networks methods are used to establish models to predict the interaction pattern between HF and NF. Artificial neural networks have proven to be excellent predictive tools in various petroleum-engineering applications. The excellent predictive capability of artificial neural networks comes from the fact that neural networks have large degrees of freedom that allows them to capture the non-linearity of the system being studied better than conventional regression techniques. The back-propagation neural network (BPNN) and probabilistic neural network (PNN) which display strong mapping ability are chosen to establish the prediction models. These models can be used to predict whether a HF will cross, dilate or slip into a NF under certain conditions, i.e. under different approach angles, different differential horizontal stress, net pressure and so on. Results have shown accuracy in predicting the interaction pattern.
Conventional HF design is based on the assumption that the rock is homogeneous and the fracture propagates symmetrically in a plane perpendicular to the minimum stress. In naturally fractured reservoirs due to interaction with NF, the fracture may propagate asymmetrically or in multiple strands or segments. The result from this work can be applied to estimate the interaction picture thus update and optimize the fracturing design.
The social and environmental impacts of the petroleum industry have serious consequences and call for sustainable solutions and practices. Embedding sustainability into organisations is vital to address these issues, and requires the integration of sustainability into performance management systems. The aim of this paper is to investigate the barriers to, and enablers of, sustainability integration in the performance management systems of an oil and gas company. A qualitative case study has been used to identify the stages and means of integration in the organisation, based on Gond et al.’s (2012) framework depicting the role of control systems in supporting sustainability integration within strategy. The findings revealed that although cognitive, organisational and technical enablers moved integration of sustainability forward in the organisation, certain cognitive barriers considerably affected the attainment of full integration. Institutional pressures provided the impetus for the development of enablers, giving rise to several implications for governments, academics and other parties. The study shows that sustainability integration in performance management systems could lead to better management and control of sustainability performance in organisations. This study provides a more comprehensive approach towards understanding the integration of sustainability in control systems from a socio-technical perspective.
Accurate forecasts are crucial to successful organizational planning. In 2001, 40 international experts published a set of principles to guide best practices in forecasting. Some of these principles relate to the use of management judgment. Most organizations use judgment at some stage in their forecasting process, but do they do so effectively? Although judgment can lead to significant improvements in forecasting accuracy, it can also be biased and inconsistent. The principles show how forecasters should use judgment and assess its effectiveness. We conducted a survey of 149 forecasters to examine the use of judgment based on these established principles and to investigate whether their forecasting procedures were consistent with the principles. In addition, we conducted four in-depth case studies. Although we found examples of good practice, we also discovered that many organizations would improve forecast accuracy if they followed basic principles such as limiting judgmental adjustments of quantitative forecasts, requiring managers to justify their adjustments in writing, and assessing the results of judgmental interventions.
CO2 and greenhouse gas emissions
- Ritchie
The Use of Multilinear Regression Models in Patterned Waterfloods: Physical Meaning of the Regression Coefficients
- P H Gentil