A Study Of The Prediction Of Subway Gas Concentration Based On GM (1,1)-Markov Chain Model

Due to the chemical character of gas concentration and its serious harmfulness,gas concentration and hazard grade prediction research has been the research topic of scholars at home and abroad,but these studies are mainly focused on high-risk industries such as mines,coal,chemical industry and so on,however,there is little research on the rail transit industry;in recent years,along with the expansion of the subway construction scales,the proportion of the subway in the life is getting larger and larger,besides,the scale of Metro is also increasing.Because the subway mainly uses the shield structure for construction,the gas spillover will occur during the construction of special stratum structure area,which will cause serious safety risks to safety production,so it is necessary to analyze the construction arears for geological and other environmental to find out the mechanism and causes of gas production,meanwhile,the gas concentration and emission during the construction process should be monitored and analyzed to predict the trend in the future and take corresponding safety precautions in the construction to avoid safety incidents.Through the study of geological structure of Fubao Road section of Kunming Metro Line 2,it is found that the generation of gas in this section is mainly related to its stratigraphic structure.Because some soil layers contain organic matter and microorganisms,these organic matter and microorganisms can react to produce gas gas.When they pass through these soil layers during excavation,gas will spill out,and the gas is concentrated in silty clay layer and peat soil layer.Taking Fubao Road of Kunming Metro Line 2 as an example,through the analysis of the gas concentration of the gas spills from the key exploration holes of Drilling 2,Drilling 4,Drilling 7,Drilling 8 and Drilling 9 in the Fubao Road Section,the GM(1,1)-Markov chain model is constructed to predict the gas concentration.Firstly,the 5-8 dimension GM(1,1)grey prediction model is established for the key holes with gas spills in the Fubao Road Section.Then,the prediction accuracy of the model is tested by posterior difference ratio,average relative error and small error probability.The dimension with the highest accuracy is selected and combined with Markov chain model,and the grey GM(1,1)-Markov chain model is established to predict the gas concentration of these key holes in the future.Finally,the accuracy of the model is tested by three evaluation indexes again.By constructing grey GM(1,1)-Markov chain model,it is found that when the fivedimensional grey GM(1,1)-Markov chain model is established for borehole 2,borehole 4 and borehole 7 of Fubao Road section,and seven-dimensional and eight-dimensional grey GM(1,1)-Markov chain model is established for borehole 8 and borehole 9 respectively,the prediction result of the model is the best and the prediction effect is the best,Except that the average relative error test results of borehole 2 and 8 are reluctant and that of borehole 9 are qualified,all the accuracy test results are good and reach the optimal level.The comprehensive average relative error of grey GM(1,1)-Markov chain prediction model is 5.2%,and the accuracy is 94.8%.In addition,according to the grey GM(1,1)-Markov chain model,the prediction results of key holes are as follows: As time goes on,the gas concentration of borehole 2,7,8 and 9 is more than 2%,and that of borehole 4 is more than 4%.According to the relevant provisions of the national safety production standard,the gas concentration in the tunnel can not exceed 0.5%.Therefore,the gas concentration in these key boreholes is in a state of serious exceeding the standard.It is necessary to ventilate and dilute the air lining in the tunnel before construction,otherwise,when the gas concentration is between 5% and 16%,explosion may occur in case of fire.At the same time,high concentration of gas will cause asphyxiation and death of construction personnel,and It is easy to cause serious or especially serious accidents such as mass casualties and injuries of subway construction personnel.