Research On Anomaly Recognition And Warning System Of Workface Gas Emission

With researching of original monitoring gas emission data, analysising abnormal status of gas emission characteristic curve and wavelet thresholding method for trend analysising,this paper establish a hierarchical model to identify early warning based on wavelet threshold denoising and a coal and gas outburst recognize early warning system by software programming techniques. The main contents are as following:(1)The establishment of gas emission time-seriesThe analysis found that the original gas data interval ranging and exists missing data and anomaly data. with 1-minute intervals, taking minute average gas concentration method to establish the time series of gas emission,and supplemented with a averaging method of adding and cleaning establish gas emission time-series in line with the principle of comparability.(2) Recognition and warning of coal and gas outburstBased on digital characteristics of gas emission time-series analysis obtain the general nature of gas emission time series; analysising and comparing different states of gas emission time-series come to the characteristics of the glaring danger states;with gas emission time-series dynamics trend analysis in wavelet thresholding method establishs a hierarchical recognition and warning model based on wavelet thresholding.(3) System implementation and verificationWith the client / server architecture, the development platform of visual studio, Microsoft SQL Server database, the paper built recognition and warning system for coal and gas outburst based on gas emission time-series and wavelet threshold denoising and verified by the actual gas monitoring data.