Method for establishing CO2 huff and puff effect prediction model and CO2 huff and puff effect evaluation method

The invention provides a method for establishing a CO2 huff and puff effect prediction model and a CO2 huff and puff effect evaluation method. The method for establishing the CO2 huff and puff effect prediction model comprises the steps of firstly obtaining a sample data set of CO2 throughput development, dividing the sample data set into a training set and a test set, then setting a neural network structure, initializing hyper-parameters of a neural network model, then training the model, and by modifying the hyper-parameters, controlling a loss function value of the model within an error range so as to obtain a trained model. The CO2 huff and puff effect evaluation method comprises the steps that an influence factor parameter data set of CO2 huff and puff development of a target well area is obtained, the influence factor parameter data set of the target well area is input into a CO2 huff and puff effect prediction model, and the oil change rate is predicted. The method can be applied to rapid prediction of