Sunya | Energy AI Observatory · Open the interactive dossier

PetroChina

2 AI initiative profiles for PetroChina across 2 departments, with evidence grades and source links. Strongest evidence: A · Operator filing — Kunlun large model with value-chain applications.

Profiles (2)

• Kunlun large model with value-chain applications

  A · Operator filing · Operating · Refining, LNG & process operations

  crude purchase evaluation cycle: 5 minutes vs hours or days (Company-reported); deployed scenarios: 152 (Company-reported)

  Source (2026-05-29)

• Machine-learning production forecasting across Dagang Oilfield blocks

  B · Primary publication · Pilot · Reservoir engineering & field development

  CNPC article reports that forecast accuracy “improved by 15%” (预测准确率提升15%). The comparison baseline, calculation, validation set, forecast horizon, and whether this is a relative improvement or percentage-point change are unspecified; it is not an absolute 15% accuracy score.

  Source (2026-02-04)

Company and vendor statements are attributed, not independently validated. Evidence grades: A operator filing, B operator release, C vendor-reported, D secondary or unverified.