Study On Prediction For Gas Concentration In Fully-Mechanized Coal Mine Based On Time Series Analysis

Prediction of gas concentration in fully mechanized coal face is an important way to prevent and control gas accidents,it is of great significance to ensure mine safety in production.In this paper,the R language of statistical software is used as a tool,build a combination model of ARIMA and GARCH by time series analysis method,seeking the rule of gas concentration variation in fully mechanized coal face and short term prediction,the main research contents are as follows:The ARIMA model fitting on the fully mechanized mining working face gas concentration monitoring number,determine the prediction equation of ARIMA model,and then to fit the gas concentration monitoring data,the result shows that the ARIMA model of gas concentration monitoring data fitting degree is high,the ARIMA model is suitable for forecasting.Based on the ARIMA model,the residual series using GARCH model,ARIMA model and GARCH model to simulate the noise prediction by ARIMA model in the value of the final combination of ARIMA model and GARCH model of fully mechanized coal face gas concentration prediction,the empirical results show that the model has higher in gas concentration forecasting precision.Based on GUI in MATLAB,the visualization interface of fully mechanized mining face prediction is established.The visual interface is stable,reliable and visualized.It provides technical support for visualization of gas concentration prediction.In this paper,combined forecasting as the leading thought,the prediction model of gas concentration in working face not only predicts the accuracy,with the help of visual operation design,the effectiveness of visual application is improved.