مانا بترو
SR-03متوسطةالموجة 2الذكاء الاصطناعي

التصنيف الآلي للسحنات

في جملة واحدة

Labels rock type from well logs across hundreds of wells consistently, which a team of interpreters working separately cannot do.

المشكلة

Facies interpretation in a mature field spans decades and dozens of interpreters. The same log response has been called three different things in three different studies, and the geomodel inherits all three.

كيف تعمل

A supervised classifier is trained on core-calibrated intervals and applied field-wide, with per-prediction confidence. Where the model is uncertain it abstains and routes the interval to an interpreter rather than guessing, and abstention rate is reported as a headline number.

البيانات المطلوبة

  • Standard log suite: GR, resistivity, density, neutron, sonic where available
  • Core descriptions with depth-shifted core-to-log ties
  • Existing facies interpretations for benchmark comparison
  • Environmental corrections and normalisation between logging vintages

ما الذي تحصل عليه

  • Facies curve per well with confidence and abstention flags
  • Field-wide facies consistency report against legacy interpretations
  • RESQML export for the geomodelling workflow

ما لا تستطيع فعله

The model reproduces the scheme it was taught, including its errors. It cannot resolve facies that are not separable in the logged response — a cemented and an uncemented sand with the same porosity will not be split by a standard suite, no matter how the model is trained, and we say that up front rather than after delivery.

كل صفحة قدرة تذكر حداً حقيقياً. إن لم يستطع المورّد إخبارك أين يتوقف منتجه، فلم يُقل لك الحق عن أين يبدأ.

قدرات ذات صلة