Method for predicting shear wave velocity of tight oil reservoir based on DNN

The invention discloses a DNN-based method for predicting the shear wave velocity of a tight oil reservoir, and the method comprises the steps: employing a Biot model containing viscous fluid to represent the elastic wave propagation of the tight oil reservoir, and employing a full-connection neural network to learn the elastic parameters in the Biot model containing the viscous fluid; a neural network model for learning elastic wave propagation characteristics of a tight oil reservoir is constructed, the shear wave speed is solved through a plane wave analysis method, actual logging data is used for verification, and then the method for predicting the shear wave speed through machine learning under the constraint of a wave propagation equation is achieved. The method is high in prediction precision, can make up for the defects that when pure data drive machine learning is used for predicting the shear wave speed, the requirements for data volume and data quality are high, and physical significance is lacked,