ADMM Network Optimization for Production Well Pressure
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Large-scale surface network optimization in hydrocarbon production systems is computationally challenging due to nonlinearity and complexity, leading to increased computational costs and time-consuming processes, necessitating more efficient methods for optimizing pressure and flux within integrated production networks.
Innovation Solution
The Alternating Direction Method of Multipliers (ADMM) framework is applied to decompose large-scale network optimization problems into smaller sub-network problems, allowing for parallelized solutions and enhanced computational efficiency through augmented Lagrangian methods and proxy models, thereby accelerating optimization processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional sequential programming methods are used to solve large-scale surface network optimization, then the optimization can be performed with established algorithms, but the computational cost increases exponentially with network size and complexity
Solution Approach 1:
The patent divides the large-scale production network into multiple subnetworks or clusters, allowing the optimization problem to be decomposed into smaller, more manageable subproblems. This segmentation reduces the exponential computational burden by solving local optimizations in parallel rather than tackling the entire network as a single monolithic problem.
Solution Approach 2:
The patent introduces an intermediary layer between the detailed network model and the optimization algorithm, using reduced-order models or surrogate models to represent complex subnetworks. This intermediary approach maintains optimization accuracy while significantly reducing computational cost by avoiding direct simulation of every network component.
2Reliability
If traditional sequential programming methods are used to solve large-scale surface network optimization, then the optimization can be performed with established algorithms, but the computational time increases making the process time-consuming
Solution Approach 1:
By segmenting the network into independent or loosely-coupled subnetworks, the patent enables parallel computation of optimization solutions. Multiple processors or computing nodes can simultaneously solve subproblems, dramatically reducing wall-clock computational time while maintaining overall optimization accuracy through coordinated solution integration.
Solution Approach 2:
The patent performs preliminary actions by pre-processing the network to identify natural clusters or subnetworks that can be optimized independently. This pre-segmentation and pre-computation of boundary conditions allows the main optimization algorithm to run faster by avoiding repeated computations of unchanged network portions.
3Measurement precision
If the network model includes detailed constraints and nonlinearity to accurately represent the production system, then the model accuracy is maintained, but the complexity of solving the optimization problem increases
Solution Approach 1:
The patent applies local quality by maintaining high model fidelity and detailed constraints only in critical subnetworks where accuracy is most important, while using simplified representations in less critical areas. This selective detail approach preserves overall model accuracy while reducing the global optimization problem complexity.
Solution Approach 2:
The patent transforms the complex nonlinear optimization problem by changing parameters through variable substitution, linearization of constraints, or reformulation into convex forms. These parameter transformations maintain the essential physics and constraints while making the optimization problem computationally more tractable and easier to solve.
Data Source
AI summary
A method of modeling pressure and flux within an integrated network of multiple wells, including: measuring reservoir pressures at each well; measuring a separator pressure at the separator; receiving or generating a model of the integrated network, the model including a node representing the separator, at least one node representing each well, and pressure constraints at the separator and each well; dividing the model into a plurality of subnetworks; performing an alternating direction method of multipliers (ADMM) optimization by iteratively solving a plurality of optimization equations each corresponding to a different subnetwork; determining a pressure at each node and a flux between each of the nodes in the model based on the optimization; and producing fluids from the network of multiple wells based on the determined pressures and fluxes in the model.