论文标题
实地发展优化的深入强化学习
Deep Reinforcement Learning for Field Development Optimization
论文作者
论文摘要
现场开发优化（FDO）问题代表了一个具有挑战性的混合企业非线性编程（MINLP）问题，我们寻求获得井的数量，其类型，位置和钻井顺序，从而最大程度地提高了经济指标。进化优化算法已有效地应用于解决FDO问题，但是，这些方法仅提供了确定性（单个）解决方案，这些解决方案通常不适合问题设置的小变化。在这项工作中，目标是将基于卷积神经网络（CNN）深入强化学习（DRL）算法应用于现场开发优化问题，以获取从不同状态或基础地质模型的代表映射到最佳决策的策略。近端策略优化（PPO）算法被考虑使用两种CNN架构，这些结构有不同的层和组成。与混合粒子群优化相比，两个网络均获得了令人满意的结果 - 网状自适应直接搜索（PSO -MADS）算法，该算法已被证明可有效解决FDO问题。
The field development optimization (FDO) problem represents a challenging mixed-integer nonlinear programming (MINLP) problem in which we seek to obtain the number of wells, their type, location, and drilling sequence that maximizes an economic metric. Evolutionary optimization algorithms have been effectively applied to solve the FDO problem, however, these methods provide only a deterministic (single) solution which are generally not robust towards small changes in the problem setup. In this work, the goal is to apply convolutional neural network-based (CNN) deep reinforcement learning (DRL) algorithms to the field development optimization problem in order to obtain a policy that maps from different states or representation of the underlying geological model to optimal decisions. The proximal policy optimization (PPO) algorithm is considered with two CNN architectures of varying number of layers and composition. Both networks obtained policies that provide satisfactory results when compared to a hybrid particle swarm optimization - mesh adaptive direct search (PSO-MADS) algorithm that has been shown to be effective at solving the FDO problem.
