Shape Prediction Of Deep Blasting Failure Zone Based On Machine Learning Algorithm

In recent years,due to the rapid growth of world economy,various industries have entered the stage of high-speed competition development,which also leads to the rapid growth of energy demand.At the same time,because of the rapid expansion of social demand for energy,the exploitation of resources in the shallow part of the earth leads to the gradual decrease of the resources in the shallow part of the earth,and the mining of ore bodies is transiting to deep mining.However,many factors should be considered in the deep mining of the mine,including but not limited to the high ground stress,blast shock wave and load,which will have a great influence on the project results.In view of this phenomenon,this paper uses a variety of machine learning algorithms to study the shape of the engineering blasting failure area,and in-depth study the shape of the engineering blasting failure area under various factors,and the results of various machine learning models to find the better model,which is useful for the actual engineering blasting project.The paper mainly carries on the following two aspects: 1.Carry on the correlation analysis of the data characteristics,and explore the main influence factors of the shape change of the engineering blasting failure area.2.The shape of the blasting fracture zone is predicted by various models and then compared to obtain the best prediction accuracy.In this paper,the initial stress,load and other characteristics are regarded as input characteristics.The shape prediction of the blasting failure zone is carried out by using the data generated by the real Blasting simulation in the laboratory.The results of correlation analysis of data characteristics show that the stress intensity and load have a great influence on the shape of the failure zone.The results show that Ada Boost integrated algorithm model has high stability and reliability for a small number of sample data operation.Ada Boost integrated algorithm model has a reference value for the design of deep mine blasting scheme.Therefore,it is feasible to study engineering blasting by machine learning.