We are proud to announce that recent SPE paper led by Prof. Bicheng Yan from DGYM, "Deep-Neural-Network-Based Workflow Increases Efficacy in Solving Complex Problems," has been highlighted in the Technology Focus column “History Matching and Forecasting” by the prestigious JPT – Journal of Petroleum Technology, April, 2025. Click here for the JPT Article.

As one of only three highlighted for innovation, this SPE paper showcases how deep learning methods can tackle critical challenges in reservoir forecasting, optimization, and characterization. Prof. Yan’s study introduces a novel Deep Neural Network-based History-Matching (DNN-HM) workflow, significantly improving the parameterization, efficiency and scalability of reservoir history matching.

As noted in the JPT article: “Collectively, these studies exemplify the transformative potential of machine learning and hybrid methods in reservoir management, balancing innovation with practical applicability.” This recognition aligns with DGYM’s ongoing commitment to pioneering research in geo-energy and computational geoscience. Read the original paper here.

About JPT:

JPT – Journal of Petroleum Technology is the flagship magazine of the Society of Petroleum Engineers (SPE), known for spotlighting leading-edge innovations and thought leadership in the oil and gas industry. The journal connects global professionals with the latest advancements in petroleum engineering.

About the Journal Author – Zhenzhen Wang, SPE:

Zhenzhen Wang is a lead research scientist and simulation engineer at Chevron Technical Center with over 10 years of industry experience. His expertise spans reservoir simulation, optimization, history matching, enhanced oil recovery, and both pressure and rate transient analysis. 

Congratulations to Bicheng on this significant achievement and contribution to the field!