Hard roof mine rock burst disaster risk prediction and analysis method

The invention discloses a hard roof mine rock burst disaster risk prediction and analysis method, which comprises the following steps of: 1, uniformly installing optical fiber sensors near a seismic source in a hard roof mine, distributing the optical fiber sensors according to an interval of 3 meters, and collecting an electric signal of vibration through the optical fiber sensors; the drilling cutting data, the micro-seismic data and the stress data are collected, chi-square analysis is carried out on all the selected three data, the relevance between the data and the rock burst is judged, the collected data comprise energy, rail smoothness, the amount of pulverized coal and the movement speed of objects in a mine, and the data are stored in a database; deep learning is carried out on the four data, a rock burst danger preliminary model is established, deep learning is carried out on the subjective danger model and the rock burst danger preliminary model, and therefore a rock burst danger model S is establi