Research On Dust Concentration Prediction Of Open-Pit Mine Based On Random Forest-Markov Model

The problem of dust pollution in open-pit coal mines has a very important significance and impact on improving the productivity of open-pit coal mines.It is also an urgent technical issue concerning the continuous operation of open-pit mines and protecting the health of employees.Only by fully understanding the current research status of open-pit dust can the research and work of open-pit dust prediction be carried out correctly and efficiently.This paper collects dust concentration data at the monitoring point of the Harwusu open-pit mine.Based on the dust concentration monitoring data from July to August at the monitoring point,a machine learning algorithm is used to establish a random forest dust concentration prediction model belonging to the open-pit mine and determine its For the model parameters,the analysis of the influence factors of dust concentration and the importance of characteristic variables based on the random forest prediction model were carried out,and the key influence factors of dust in the open pit were obtained.The effect evaluation,analysis and optimization of the random forest dust concentration prediction model are carried out,and the Markov model is proposed to modify the prediction results of the random forest prediction model.Through extensive training on the existing open-pit mine dust data,the data rules are summarized,and part of the data not involved in the training is used to test the sensitivity and accuracy of the dust concentration prediction model.The final results show that the average relative error and root mean square error of the prediction results obtained after Markov correction are smaller,and the accuracy is improved.Finally,a high-precision random forest-Markov dust concentration prediction model is constructed.There are 30 figures,15 tables and 75 references in this thesis.