AI-Based Prediction of Numerical Earthquakes Using (Pseudo) Acoustic Emission

The Discrete Element Method is widely used in applied mechanics, particularly in situations where material continuity breaks down (fracturing, crushing, friction, granular flow) and classical rheological models fail (phase transition between solid and granular). In this study, the Discrete Element Method was employed to simulate stick–slip cycles, i.e., numerical earthquakes. At 2000 selected, regularly spaced time checkpoints, parameters describing the average state of all particles forming the numerical fault were recorded. These parameters were related to the average velocity of the particles and were treated as the numerical equivalent of (pseudo) Acoustic Emission. The collected datasets were used to train the Random Forest and Deep Learning models that successfully predicted the time to failure. SHapley Additive exPlanations (SHAP) was used to quantify the contribution of individual physical parameters of the particles to the prediction results. The main novelty of this study was the prediction of time to failure for entire event sequences. Using only instantaneous particle velocity statistics and without using information about the history of previous events, coefficients of determination in the range R2 = 0.81–0.96 were obtained.