Automated Well Placement Planning Using Hierarchical Evolutionary Algorithms
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Determining optimal well placement in oil and gas fields is complex due to geological, geomechanical, and surface constraints, with existing automated and semi-automated computation methods being limited by simplistic models and lack of consideration for uncertainty and non-linear optimization.
Innovation Solution
An automated process using a hierarchical workflow with increasing algorithm complexity to identify and refine well placement candidates, combining static reservoir models with dynamic flow simulation and cost functions to maximize recovery or economic benefit, involving multiple independent sets of wells and accounting for drilling and surface constraints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If simple algorithms are used to process large candidate populations, then computational speed is improved, but solution accuracy deteriorates
Solution Approach 1:
The patent segments the computational process into multiple stages with different algorithmic complexity levels. A simple fast algorithm processes the largest candidate population first, then progressively more complex algorithms process smaller subsets of refined candidates. This segmentation allows the system to achieve both speed (in early stages) and accuracy (in later stages) without compromising either aspect.
Solution Approach 2:
The patent implements dynamic algorithm selection based on the candidate population size and refinement stage. The system transitions from simple algorithms to more complex algorithms as the candidate population is reduced through filtering stages. This dynamic adaptation ensures optimal computational efficiency at each stage while maintaining solution accuracy through progressive refinement.
2Measurement precision
If complex algorithms are used to maximize solution accuracy, then measurement precision is improved, but computational time increases
Solution Approach 1:
The patent applies preliminary filtering actions using simple algorithms to reduce the candidate population before applying complex algorithms. By pre-processing and eliminating obviously inferior candidates through fast, simple evaluation criteria, the system reduces the computational burden on complex algorithms, thereby achieving high accuracy without excessive computational time.
Solution Approach 2:
The computational process is divided into sequential segments where simpler algorithms handle initial screening and more complex algorithms handle final optimization. This segmentation ensures that complex algorithms are applied only when necessary to the reduced candidate set, minimizing overall computational time while maintaining solution accuracy.
3Productivity
If a hierarchical workflow with increasing algorithm complexity is used, then solution efficiency is improved, but device complexity increases
Solution Approach 1:
The patent segments the workflow into distinct hierarchical levels, each with specific algorithmic complexity and processing objectives. This segmentation organizes the increasing complexity into manageable stages, improving solution efficiency through progressive refinement while keeping the overall system architecture clear and controllable.
Solution Approach 2:
The hierarchical workflow dynamically adapts algorithmic complexity based on the processing stage and candidate population characteristics. This dynamic structure allows the system to achieve high solution efficiency through appropriate algorithm selection at each level while managing overall system complexity through structured progression.
Data Source
AI summary
A hybrid evolutionary algorithm (“HEA”) technique is described for automatically calculating well and drainage locations in a field. The technique includes planning a set of wells on a static reservoir model using an automated well planner tool that designs realistic wells that satisfy drilling and construction constraints. A subset of these locations is then selected based on dynamic flow simulation using a cost function that maximizes recovery or economic benefit. In particular, a large population of candidate targets, drain holes and trajectories is initially created using fast calculation analysis tools of cost and value, and as the workflow proceeds, the population size is reduced in each successive operation, thereby facilitating use of increasingly sophisticated calculation analysis tools for economic valuation of the reservoir while reducing overall time required to obtain the result. In the final operation, only a small number of full reservoir simulations are required for the most promising FDPs.