Data-driven model for hydraulic fracturing design optimization. Part II: Inverse problem
نویسندگان
چکیده
We describe a stacked model for predicting the cumulative fluid production an oil well with multistage-fracture completion based on combination of Ridge Regression and CatBoost algorithms. The is developed extended digital field data base reservoir, fracturing design parameters. database now includes more than 5000 wells from 23 oilfields Western Siberia (Russia), 6687 operations in total. Starting 387 parameters characterizing each well, including construction, reservoir properties, features production, we end up 38 key used as input training process. demonstrates physically explainable dependencies plots target (number stages, proppant mass, average final concentrations rate). set methods those use Euclidean distance clustering techniques offset selection (search similar terms certain metrics), which useful engineer to analyze earlier treatments wells. These approaches are also adapted obtaining margings optimization particular pilot part testing campaign methodology. An inverse problem (selecting optimum maximize production) formulated optimizing high dimensional black box approximation function constrained by boundaries solved four different methods: surrogate-based (pSeven), sequential least squares programming, particle swarm differential evolution. A recommendation system containing all above designed advise stimulation optimized design. • Workflow data-drive forecast model. Inverse several gradient-free (with probability improvement) optimization. Methodology presented algorithm best practices Recommendation these assist engineer.
منابع مشابه
the algorithm for solving the inverse numerical range problem
برد عددی ماتریس مربعی a را با w(a) نشان داده و به این صورت تعریف می کنیم w(a)={x8ax:x ?s1} ، که در آن s1 گوی واحد است. در سال 2009، راسل کاردن مساله برد عددی معکوس را به این صورت مطرح کرده است : برای نقطه z?w(a)، بردار x?s1 را به گونه ای می یابیم که z=x*ax، در این پایان نامه ، الگوریتمی برای حل مساله برد عددی معکوس ارانه می دهیم.
15 صفحه اولData-driven Inverse Optimization with Incomplete Information
In data-driven inverse optimization an observer aims to learn the preferences of an agent who solves a parametric optimization problem depending on an exogenous signal. Thus, the observer seeks the agent’s objective function that best explains a historical sequence of signals and corresponding optimal actions. We formalize this inverse optimization problem as a distributionally robust program m...
متن کاملData-driven inverse optimization with imperfect information
In data-driven inverse optimization an observer aims to learn the preferences of an agent who solves a parametric optimization problem depending on an exogenous signal. Thus, the observer seeks the agent’s objective function that best explains a historical sequence of signals and corresponding optimal actions. We focus here on situations where the observer has imperfect information, that is, wh...
متن کاملAn analytical model for hydraulic fracturing in shallow bedrock formations.
A theoretical method is proposed to estimate post-fracturing fracture size and transmissivity, and as a test of the methodology, data collected from two wells were used for verification. This method can be employed before hydrofracturing in order to obtain estimates of the potential hydraulic benefits of hydraulic fracturing. Five different pumping test analysis methods were used to evaluate th...
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: Journal of Petroleum Science and Engineering
سال: 2022
ISSN: ['0920-4105', '1873-4715']
DOI: https://doi.org/10.1016/j.petrol.2021.109303