Ventilation Safety Monitoring System Using A Trend Curve Of Moving Average Of Carbon Monoxide Concentration

A power-series trend curve of elapsed time was estimated by using time series data of measured concentrations of carbon monoxide in the Taiheiyo coal mine, Hokkaido, Japan. The difference between the moving average concentration data and the estimated data from the trend curve showed linear correlation when plotted on a normal probability paper and a correlation coefficient value of 0.99. It is concluded that the estimate of carbon monoxide concentration using moving average values of concentrations compares very well with the results obtained from the trend curve within the three-sigma (standard deviation) range of the distribution. Thus, it is believed the results can be used to monitor and detect abnormal ventilation conditions in a ventilation network such as the occurrence of spontaneous combustion at an early stage.