Target seepage prediction method based on BO-GBRT strategy

The invention discloses a target object seepage prediction method based on a BO-GBRT strategy, and belongs to the technical field of seepage prediction, and the method comprises the steps: S1, determining initial target object seepage influence factor data based on the influence and hysteresis effect of different factors on target object seepage; s2, preprocessing the initial target object seepage flow influence factor data, the historical seepage flow monitoring data and the influence factor data, and constructing a sample data set; s3, on the basis of the sample data set, optimizing the seepage influence factors of the initial target object through a Spearman correlation coefficient method, and constructing an optimized sample data set; s4, establishing a gradient regression tree model, training the optimal sample data set, and introducing a Bayesian optimization algorithm to perform hyper-parameter optimization to obtain an optimal seepage prediction model; and S5, performing seepage flow prediction by tak