Sequential field development plan through robust optimization coupling with CNN and LSTM-based proxy models

Machine-learning technologies have exhibited robust competences in solving many petroleum engineering problems. The accurate predictivity and fast computational speed enable a large volume of time-consuming engineering processes such as history-matching and field development optimization. The Southwest Regional Partnership on Carbon Sequestration (SWP) project desires rigorous history-matching and multi-objective optimization processes, which fits the superiorities of the machine-learning approaches. Although the machine-learning proxy models are trained and validated before imposing to solve practical problems, the error margin would essentially introduce uncertainties to the results. In this paper, a hybrid numerical machine-learning workflow solving various optimization problems is presented. By coupling the expert machine-learning proxies with a global optimizer, the workflow successfully solves the history-matching and CO2 water alternative gas (WAG) design problem with low computational overheads. The history-matching work considers the heterogeneities of multiphase relative characteristics, and the CO2-WAG injection design takes multiple techno-economic objective functions into accounts. This work trained an expert response surface, a support vector machine, and a multi-layer neural network as proxy models to effectively learn the high-dimensional nonlinear data structure. The proposed workflow suggests revisiting the high-fidelity numerical simulator for validation purposes. The experience gained from this work would provide valuable guiding insights to similar CO2 enhanced oil recovery (EOR) projects.

Download Full-text

Offshore-Field Development Plan Changed From Steel Structures to Artificial Islands

Journal of Petroleum Technology ◽

10.2118/1011-0065-jpt ◽

2011 ◽

Vol 63 (10) ◽

pp. 65-68

Author(s):

Dennis Denney

Keyword(s):

Steel Structures ◽

Development Plan ◽

Field Development ◽

Offshore Field

Download Full-text

Integrated Halo Model And Hydrodynamic Simulation For Optimization Oil Production In Crystalline Fractured Reservoirs

10.21203/rs.3.rs-832230/v1 ◽

2021 ◽

Author(s):

Hung Vo Thanh ◽

Kang-Kun Lee

Keyword(s):

Oil Recovery ◽

History Matching ◽

Fractured Reservoirs ◽

Petroleum Industry ◽

Oil Production ◽

Modelling And Simulation ◽

Development Plan ◽

Sweep Efficiency ◽

Field Development ◽

The World

Abstract Basement formation is known as the unique reservoir in the world. The fractured basement reservoir was contributed a large amount of oil and gas for Vietnam petroleum industry. However, the geological modelling and optimization of oil production is still a challenge for fractured basement reservoirs. Thus, this study aims to introduce the efficient workflow construction reservoir models for proposing the field development plan in a fractured crystalline reservoir. First, the Halo method was adapted for building the petrophysical model. Then, Drill stem history matching is conducted for adjusting the simulation results and pressure measurement. Next, the history-matched models are used to conduct the simulation scenarios to predict future reservoir performance. The possible potential design has four producers and three injectors in the fracture reservoir system. The field prediction results indicate that this scenario increases approximately 8 % oil recovery factor compared to the natural depletion production. This finding suggests that a suitable field development plan is necessary to improve sweep efficiency in the fractured oil formation. The critical contribution of this research is the proposed modelling and simulation with less data for the field development plan in fractured crystalline reservoir. This research's modelling and simulation findings provide a new solution for optimizing oil production that can be applied in Vietnam and other reservoirs in the world.

Download Full-text

A practical method for planning large number of horizontal wells with a reservoir model for a field development plan

Petrovietnam Journal ◽

10.47800/pvj.2021.10-02 ◽

2021 ◽

Vol 10 ◽

pp. 17-32

Author(s):

Guido Fava ◽

Việt Anh Đinh

Keyword(s):

Horizontal Well ◽

Horizontal Wells ◽

Simulation Software ◽

Production Performance ◽

Development Plan ◽

Regular Pattern ◽

Production Forecast ◽

Advanced Technique ◽

Large Reservoir ◽

Field Development

The most advanced technique to evaluate different solutions proposed for a field development plan consists of building a numerical model to simulate the production performance of each alternative. Fields covering hundreds of square kilometres frequently require a large number of wells. There are studies and software concerning optimal planning of vertical wells for the development of a field. However, only few studies cover planning of a large number of horizontal wells seeking full population on a regular pattern. One of the criteria for horizontal well planning is selecting the well positions that have the best reservoir properties and certain standoffs from oil/water contact. The wells are then ranked according to their performances. Other criteria include the geometry and spacing of the wells. Placing hundreds of well individually according to these criteria is highly time consuming and can become impossible under time restraints. A method for planning a large number of horizontal wells in a regular pattern in a simulation model significantly reduces the time required for a reservoir production forecast using simulation software. The proposed method is implemented by a computer script and takes into account not only the aforementioned criteria, but also new well requirements concerning existing wells, development area boundaries, and reservoir geological structure features. Some of the conclusions drawn from a study on this method are (1) the new method saves a significant amount of working hours and avoids human errors, especially when many development scenarios need to be considered; (2) a large reservoir with hundreds of wells may have infinite possible solutions, and this approach has the aim of giving the most significant one; and (3) a horizontal well planning module would be a useful tool for commercial simulation software to ease engineers' tasks.

Download Full-text

Utilizing Pseudo Well Connections to Simulate Multi-Stage Hydraulic Fracturing - Example Based On the Study of a Tight Gas Asset

10.2523/iptc-21318-ms ◽

2021 ◽

Author(s):

Aamir Lokhandwala ◽

Vaibhav Joshi ◽

Ankit Dutt

Keyword(s):

Hydraulic Fracturing ◽

Oil And Gas ◽

Flow Simulation ◽

Oil And Gas Industry ◽

Development Plan ◽

Hydraulic Fractures ◽

Tight Gas ◽

Gas Field ◽

Field Development ◽

Fracture Modeling

Abstract Hydraulic fracturing is a widespread well stimulation treatment in the oil and gas industry. It is particularly prevalent in shale gas fields, where virtually all production can be attributed to the practice of fracturing. It is also used in the context of tight oil and gas reservoirs, for example in deep-water scenarios where the cost of drilling and completion is very high; well productivity, which is dictated by hydraulic fractures, is vital. The correct modeling in reservoir simulation can be critical in such settings because hydraulic fracturing can dramatically change the flow dynamics of a reservoir. What presents a challenge in flow simulation due to hydraulic fractures is that they introduce effects that operate on a different length and time scale than the usual dynamics of a reservoir. Capturing these effects and utilizing them to advantage can be critical for any operator in context of a field development plan for any unconventional or tight field. This paper focuses on a study that was undertaken to compare different methods of simulating hydraulic fractures to formulate a field development plan for a tight gas field. To maintaing the confidentiality of data and to showcase only the technical aspect of the workflow, we will refer to the asset as Field A in subsequent sections of this paper. Field A is a low permeability (0.01md-0.1md), tight (8% to 12% porosity) gas-condensate (API ~51deg and CGR~65 stb/mmscf) reservoir at ~3000m depth. Being structurally complex, it has a large number of erosional features and pinch-outs. The study involved comparing analytical fracture modeling, explicit modeling using local grid refinements, tartan gridding, pseudo-well connection approach and full-field unconventional fracture modeling. The result of the study was to use, for the first time for Field A, a system of generating pseudo well connections to simulate hydraulic fractures. The approach was found to be efficient both terms of replicating field data for a 10 year period while drastically reducing simulation runtime for the subsequent 10 year-period too. It helped the subsurface team to test multiple scenarios in a limited time-frame leading to improved project management.

Download Full-text