Automated Well Placement Planning in Reservoir Simulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for planning new wells in hydrocarbon reservoirs are manual, time-consuming, and prone to errors, especially in heterogeneous reservoirs, leading to incomplete evaluation of scenarios and suboptimal decision-making due to computational and human resource intensity.
Innovation Solution
An automated method using a reservoir simulator to determine candidate well locations, trajectories, and predicted production rates, ranking them based on combined production rates, and selecting optimal locations for drilling, which reduces time and error by identifying 'sweet-spots' with high permeability and low water saturation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual methods are used to plan new wells, then human expertise and flexibility are maintained, but planning time increases significantly and evaluation completeness decreases
Solution Approach 1:
The reservoir simulator performs self-service by automatically executing the well placement optimization process without requiring manual intervention for each evaluation. The system autonomously evaluates numerous candidate well locations, calculates production rates, and ranks options based on predefined criteria, thereby reducing planning time while maintaining decision-making quality through consistent, repeatable simulations.
Solution Approach 2:
The patent replaces the manual mechanical process of well placement planning with an automated computational system. The reservoir simulator uses numerical models and algorithms to substitute human manual evaluation, enabling rapid assessment of multiple scenarios that would be time-consuming to evaluate manually while ensuring comprehensive coverage of potential well locations.
2Loss of information
If comprehensive scenario evaluation is performed manually, then evaluation completeness improves, but computational resource consumption and time increase prohibitively
Solution Approach 1:
The patent segments the reservoir into a three-dimensional grid of computational cells, allowing the simulator to systematically evaluate each cell and its surrounding neighborhood for optimal well placement. This segmentation enables comprehensive scenario evaluation by breaking down the complex reservoir into manageable units that can be processed efficiently through automated simulations, reducing overall computational resource consumption while maintaining evaluation completeness.
Solution Approach 2:
The reservoir simulator dynamically adjusts simulation parameters such as grid resolution, time step size, and physical model complexity based on the specific well location and reservoir conditions being evaluated. This adaptive parameter adjustment allows comprehensive scenario evaluation across multiple conditions while optimizing computational resource consumption by applying higher resolution only where necessary rather than uniformly across the entire reservoir.
3Productivity
If automated well placement is implemented, then planning time is reduced significantly, but system complexity increases
Solution Approach 1:
The reservoir simulator is designed as a multi-functional platform that handles well placement optimization, production rate prediction, and scenario evaluation within a single integrated system. This universality reduces the need for multiple separate tools and processes, thereby managing system complexity while achieving high productivity through automated well placement and comprehensive scenario evaluation.
Data Source
AI summary
Methods and systems are disclosed. The methods may include, using a reservoir simulator to obtain a reservoir simulation model including a grid of computational voxels, a free-water level boundary, and locations of pre-existing production wells and defining a set of candidate well locations within an area enclosed by the free-water level boundary. The methods may further include forming a set of candidate production well locations, where each well candidate production well location is greater than a first threshold distance from all pre-existing production wells and further than a second threshold distance from the free-water level boundary. The methods may also include determining a trajectory for lateral branches of each candidate production well and selecting a set of planned production well locations based on a ranking of combined predicted production rate from the lateral branches of each candidate production well and drilling a production well guided by the trajectories.