Automatic identification and alarm method for stabbing drilling tool

The invention discloses an automatic identification and alarm method for a stabbing drilling tool. The method comprises the following steps: S1, delimiting key characteristic parameters of the stabbing drilling tool; s2, acquiring to-be-predicted target well adjacent well drilling data of whether drilling tool piercing leakage will occur or not; s3, obtaining preprocessed drilling sequence data; s4, segmenting the preprocessed drilling sequence data into a training set and a verification set; s5, creating and training an LSTM neural network model by using the obtained historical data; s6, optimizing model hyper-parameters by using a Bayesian network, and predicting future drilling sequence data by using the trained model; and S7, predicting whether the to-be-predicted target well is about to be punctured by the drilling tool or not. Fluctuation features with remarkable rules in original drilling tool piercing-through sequence data are removed through preprocessing, and training and prediction of an LSTM neura