Heterogeneous reservoir parameter prediction method and system based on transfer learning

• Main Authors: GAO GUOHAI, ZENG XINGJIE, WANG XIN, CHEN YUFAN, WANG YANG, JIANG WEI
• Format: Patent
• Language: Chinese; English
• Published: 19.12.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; DETECTING MASSES OR OBJECTS; EARTH DRILLING; EARTH DRILLING, e.g. DEEP DRILLING; ELECTRIC DIGITAL DATA PROCESSING; FIXED CONSTRUCTIONS; GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; MEASURING; MINING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR ASLURRY OF MINERALS FROM WELLS; PHYSICS; TESTING

The invention relates to a heterogeneous reservoir parameter prediction method and system based on transfer learning, and belongs to the field of reservoir assessment. The problems that the application effect of an existing method on a complex reservoir is not ideal and the model interpretability is poor are solved. According to the technical scheme, firstly, a data enhancement algorithm is adopted for a data imbalance problem, unbalanced logging data are effectively expanded, and data balance is achieved; then, establishing a lithology discrimination model and a seepage capacity discrimination model based on logging data by using a random forest algorithm; and finally, based on the correlation among the reservoir parameters, introducing transfer learning, and constructing a reservoir parameter prediction model. According to the heterogeneous reservoir parameter prediction method and system based on transfer learning, the lithology category and the seepage capacity level can be obtained through autonomous tra