Prediction Method of Blasting Vibration by Optimized GEP Based on Spark

In order to minimize the damage of engineering blasting vibration, it is very important to accurately predict the blasting peak velocity. The GEP algorithm is used to analyze the relationship between blasting vibration and related parameters. Aiming to improve the time performance of GEP in dealing with large-scale engineering blasting data, the gene structure of GEP is adjusted to head, body and tail. Then the GEP algorithm is paralleled on Spark cluster. The optimized parallel GEP can greatly improves the global search efficiency. The experimental results show that the method can obviously reduce the running time and can improve the prediction accuracy in most cases.