Abstract: Horizontal well bridge plug stage fracturing is one of the main technologies for the efficient development of unconventional oil and gas. Among them, the real-time diagnosis of the effectiveness of the bridge plug ball seat setting seal is the key link, once the tea seat setting fails, it will not be able to follow up normally. At present, it mainly relies on manual observation of the characteristics of wellhead pressure changes, which is difficult to quickly and accurately identify. To this end, this paper integrates expert experience qualitative judgment and quantitative annotation of setting data feature mining, and slices the sliding window data to form 5792 sets of label data. A long short-term memory neural network with two-dimensional input of wellhead pressure-displacement is preferred. An intelligent diagnostic model for the effectiveness of fracturing tee setting was established, and the under sampling balance dataset was used to improve the prediction accuracy of the model. The results show that the setting data shows a significant three-stage characteristic of steep rise, steep drop and gentle rise of wellhead pressure, and if the wellhead pressure lacks a certain stage feature, it is invalid set. The statistical value of wellhead pressure slope has a large distribution range, and it is impossible to form clear rules to achieve accurate diagnosis. Artificial intelligence technology is used to learn the valid/invalid setting data characteristics of different wellhead pressure forms, and the diagnostic results are output per second, with an accuracy of 96.8% for the test set and 84.3% for the verification set. Compared with the long short-term memory neural network with the dual input of wellhead pressure and displacement, the accuracy is increased by 5.1 percentage points compared with the single input of wellhead pressure. The results of this study are expected to provide a model method for the real-time automatic diagnosis of the effectiveness of the bridge plug ball seat.