Method and system for predicting yield of oil field in high water cut period and storage medium

The invention discloses a high-water-cut-period oil field yield prediction method and system and a storage medium, and relates to the technical field of oil field development, and the method comprises the steps: obtaining oil reservoir data of a target oil field; the oil reservoir data comprises static variable data and dynamic variable data; on the basis of the static variable data and the dynamic variable data after data cleaning, using a trained variable weight optimization module to calculate the weight of each variable and perform variable selection to obtain an optimized static variable weight matrix and an optimized dynamic variable weight matrix; endowing the optimized static variable weight matrix with a time sequence by using the trained static covariable encoder to obtain a time sequence static variable weight matrix; based on the time sequence static variable weight matrix and the optimized dynamic variable weight matrix, obtaining a variable fusion matrix by using the trained long-short-term memo