Abstract:Abstract A Derivative-Free Trust-Region (DFTR) algorithm is proposed to solve the robust well control optimization problem under geological uncertainty. Derivative-Free (DF) methods are often a practical alternative when gradients are not available or are unreliable due to cost function discontinuities, e.g., caused by enforcement of simulation-based constraints. However, the effectiveness of DF methods for solving realistic cases is heavily dependent on an efficient sampling strategy since cost function calculations often involve time-consuming reservoir simulations. The DFTR algorithm samples the cost function space around an incumbent solution and builds a quadratic polynomial model, valid within a bounded region (the trust-region). A minimization of the quadratic model guides the method in its search for descent. Because of the curvature information provided by the model-based routine, the trust-region approach is able to conduct a more efficient search compared to other sampling methods, e.g., direct-search approaches. DFTR is implemented within FieldOpt, an open-source framework for field development optimization, and is tested in the Olympus benchmark against two other types of methods commonly applied to production optimization: a direct-search (Asynchronous Parallel Pattern Search) and a population-based (Particle Swarm Optimization). Current results show that DFTR has improved performance compared to the model-free approaches. In particular, the method presented improved convergence, being capable to reach solutions with higher NPV requiring comparatively fewer iterations. This feature can be particularly attractive for practitioners who seek ways to improve production strategies while using an ensemble of full-fledged models, where good convergence properties are even more relevant.

A Derivative-Free Trust-Region Algorithm for Reliability-Based Optimization

An Off-Policy Trust Region Policy Optimization Method with Monotonic Improvement Guarantee for Deep Reinforcement Learning

Globally Convergent Composite-Step Trust-Region Framework for Model-Based Real-Time Optimization

A Trust-Region Framework for Real-Time Optimization with Structural Process-Model Mismatch

Improved Algorithms Based on Trust Region Framework for Solving Unconstrained Derivative Free Optimization Problems

A Derivative-Free Trust-Region Method for Biobjective Optimization

An Efficient and Multi-Fidelity Reliability-Based Design Optimization Method Based on a Novel Surrogate Model Local Update Strategy

A Modified Probability Collectives Optimization Algorithm Based on Trust Region Method and a New Temperature Annealing Schedule

A Trust-Region Algorithm for Noisy Equality Constrained Optimization

Reliability-based design optimization using adaptive Kriging-A single-loop strategy and a double-loop one

A general methodology for reliability-based robust design optimization of computation-intensive engineering problems

Derivative-free trust region optimization for robust well control under geological uncertainty

A Multi-Constraint Failure-Pursuing Sampling Method for Reliability-Based Design Optimization Using Adaptive Kriging

Trust-Region Stochastic Optimization with Variance Reduction Technique

Reliability-based Topology Optimization using Stochastic Gradients

A New Modified Nonmonotone Adaptive Trust Region Method for Unconstrained Optimization

S-BORM: Reliability-based optimization of general systems using buffered optimization and reliability method

Efficient Trust Region Filter Modeling Strategies for Computationally Expensive Black-Box Optimization

A TRUST-REGION METHOD FOR NONSMOOTH NONCONVEX OPTIMIZATION

A nonmonotone trust region method for unconstrained optimization

An Improved Non-Monotonic Adaptive Trust Region Algorithm for Unconstrained Optimization