Application of Machine Learning to Accelerate Gas Condensate Reservoir Simulation
نویسندگان
چکیده
According to the roadmap toward clean energy, natural gas has been pronounced as perfect transition fuel. Unlike usual dry reservoirs, condensates yield liquid which remains trapped in reservoir pores due high capillarity, leading loss of an economically valuable product. To compensate, produced on surface is stripped from its heavy components and reinjected back thus causing revaporization condensate. optimize this recycling process compositional simulation utilized, which, however, takes very long complete complexity governing differential equations implicated. The calculations determining prevailing k-values at every grid block each time step account for a great part total CPU time. In work machine learning (ML) employed accelerate thermodynamic by providing tiny fraction required conventional methods. Regression tools such artificial neural networks (ANNs) are trained against that have obtained beforehand running sample simulations small domains. Subsequently, regression embedded simulators acting proxy models. prediction error achieved shown be negligible needs real-world condensate simulation. gain least one order magnitude, rendering proposed approach yet another successful implementation ML energy field.
منابع مشابه
Machine Learning Applied to 3-D Reservoir Simulation
The optimization of subsurface flow processes is important for many applications including oil field operations and the geological storage of carbon dioxide. These optimizations are very demanding computationally due to the large number of flow simulations that must be performed and the typically large dimension of the simulation models. In this work, reduced-order modeling (ROM) techniques are...
متن کاملhazard evaluation of gas condensate stabilization and dehydration unit of parsian gas refinery using hazop procedures
شناسایی مخاطرات در واحد 400 پالایشگاه گاز پارسیان. در این پروزه با بکارگیری از تکنیک hazop به شناسا یی مخاطرات ، انحرافات ممکن و در صورت لزوم ارایه راهکارهای مناسب جهت افزایش ایمنی فرا یند پرداخته میگردد. شرایط عملیاتی مخاطره آمیز نظیر فشار و دمای بالا و وجود ترکیبات مختلف سمی و قابل انفجار در واحدهای پالایش گاز، ضرورت توجه به موارد ایمنی در این چنین واحدهایی را مشخص می سازد. مطالعه hazop یک ر...
Experimental and Simulation Constructing of Phase Diagram and Quality Lines for a Lean Gas Condensate Reservoir Fluid
Constructing phase diagram and its subsequent quality lines for lean gas condensate reservoirs is always crucial and challenging topic in reservoir fluid studies. Due to small amounts of liquid drop-outs in lean gas condensates at reservoir conditions, full experimental tracking of phase diagram and its relevant quality lines is almost impossible for lean gas condensates.On the other hand, ...
متن کاملModeling and Optimization of the Rich Gas-Condensate Reservoir
A composition stream from reservoir to the market has been obtained as a result of upstream, midstream and downstream retrograde gas-condensate models integration. Applications are composition, Peng-Robinson EOS has been chosen as a fluid package. Different production scenarios were observed to define development strategy for rich gas condensate production. VGD and C/V gas drive mechanisms have...
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: Clean technologies
سال: 2022
ISSN: ['2571-8797']
DOI: https://doi.org/10.3390/cleantechnol4010011