Parallel Automatic History Matching Algorithm Using Reinforcement Learning
نویسندگان
چکیده
Reformulating the history matching problem from a least-square mathematical optimization into Markov Decision Process introduces method in which reinforcement learning can be utilized to solve problem. This provides mechanism where an artificial deep neural network agent interact with reservoir simulator and find multiple different solutions Such formulation allows for solving parallel by launching concurrent environments enabling learn simultaneously all at once, achieving significant speed up.
منابع مشابه
fault location in power distribution networks using matching algorithm
چکیده رساله/پایان نامه : تاکنون روشهای متعددی در ارتباط با مکان یابی خطا در شبکه انتقال ارائه شده است. استفاده مستقیم از این روشها در شبکه توزیع به دلایلی همچون وجود انشعابهای متعدد، غیر یکنواختی فیدرها (خطوط کابلی، خطوط هوایی، سطح مقطع متفاوت انشعاب ها و تنه اصلی فیدر)، نامتعادلی (عدم جابجا شدگی خطوط، بارهای تکفاز و سه فاز)، ثابت نبودن بار و اندازه گیری مقادیر ولتاژ و جریان فقط در ابتدای...
A New Algorithm for Automatic History Matching
W. H. CHEN G. R. GAVALAS J. H. SEINFELD M. L. WASSERMAN History-matching problems, in which reservoir parameters arc to be estimated from well pressure data, are formulated as optimal control problems. The necessary conditions for optimality lead naturally to gradient optimization methods for determining the optimal parameter estimates. The key feature of the approach is that reservoir properti...
متن کاملAutomatic Abstraction in Reinforcement Learning Using Ant System Algorithm
Nowadays developing autonomous systems, which can act in various environments and interactively perform their assigned tasks, are intensively desirable. These systems would be ready to be applied in different fields such as medicine, controller robots and social life. Reinforcement learning is an attractive area of machine learning which addresses these concerns. In large scales, learning perfo...
متن کاملAlgorithm Selection using Reinforcement Learning
Many computational problems can be solved by multiple algorithms, with different algorithms fastest for different problem sizes, input distributions, and hardware characteristics. We consider the problem of algorithm selection: dynamically choose an algorithm to attack an instance of a problem with the goal of minimizing the overall execution time. We formulate the problem as a kind of Markov d...
متن کاملAutomatic Bridge Bidding Using Deep Reinforcement Learning
Bridge is among the zero-sum games for which artificial intelligence has not yet outperformed expert human players. The main difficulty lies in the bidding phase of bridge, which requires cooperative decision making under partial information. Existing artificial intelligence systems for bridge bidding rely on and are thus restricted by human-designed bidding systems or features. In this work, w...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Energies
سال: 2023
ISSN: ['1996-1073']
DOI: https://doi.org/10.3390/en16020860