Prediction Method And System Realization Of Coal Mine Pressure

Coal energy has long been in the leading position of all energy sources in China,and is an important industry to maintain the national economic level and energy production,however,in the process of continuous deep mining of coal mines has led to the frequent occurrence of mine pressure disasters at the coal mine working face,which has brought great impact to the coal mine safety production and caused serious risks to the life safety of coal mine workers.In the process of coal mine safety production,most of the roof mine pressure disasters occur during the incoming pressure,therefore,for the problem of coal mine roof mine pressure disasters,effective prediction of coal mine working face cycle incoming pressure is of great significance for the prevention and guidance of mine pressure disasters.This thesis focuses on the effective prediction of incoming pressure of coal mine working face roof cycle,and the research work is as follows:(1)A CEEMD-SVM-ARIMA based model for predicting the incoming pressure of coal mine roof period is proposed.The model first preprocesses the collected raw data,then decomposes the processed data using CEEMD modal decomposition method,divides the decomposed sequences into high-frequency sequences and low-frequency sequences,applies SVM model to predict high-frequency sequences and ARIMA model to predict low-frequency sequences,and finally superimposes the predicted results to construct a CEEMD-SVM-ARIMA based coal mine The experimental results show that in the No.1 support,compared with the SVM and ARIMA models,the MAE in the prediction of the CEEMD-SVM-ARIMA model based on the CEEMD-SVM-ARIMA model is reduced by 26.26% and 99.04%,respectively,and the RMSE were reduced by23.72% and 44.61%,respectively.In scaffold #9,the MAE in prediction based on CEEMD-SVM-ARIMA model was reduced by 44.51%,80.95%,and RMSE by 39.39%,32.95%,respectively,compared to SVM and ARIMA models.In the 18 th scaffold,the MAE in prediction based on the CEEMD-SVM-ARIMA model was reduced by 34.71%,99.3%,and RMSE by 34.98%,56.14%,respectively,compared to the SVM and ARIMA models.The prediction accuracy of the three brackets is significantly improved,and the predicted results are consistent with the periodicity of mine pressure in coal mines and reach the coal mine site conditions,which can respond to the maximum value of the incoming pressure of the roof.(2)A gray correlation-based CEEMD-SVM-ARIMA coal mine roof cycle incoming pressure prediction model is proposed;the gray correlation method is introduced to analyze the spatial correlation between coal mine working face mine pressure time series,and the data set with high correlation is used as the input data set,and then the data set is reconstructed;the data is used as the input data of the CEEMD-SVM-ARIMA model for prediction;a CEEMD-SVM-ARIMA coal mine roof cycle pressure prediction model based on gray correlation is constructed.And the model was applied in a coal mine,and the experimental results showed that compared with the gray correlation degree VMDSVM prediction model and the gray correlation degree based SVM model,the MAE of the gray correlation degree based CEEMD-SVM-ARIMA prediction model was reduced by 67.22% and 41.78%,and the RMSE was reduced by 64.58% and 33.92%,respectively.(3)The two proposed algorithms for prediction of coal mine roof pressure cycle are applied in practice,and a coal mine roof pressure monitoring and early warning system is designed to realize the intelligent function of coal mine working face roof pressure monitoring and early warning system,which is tested in a coal mine.Figure[26] Table[13] Reference[80]...