Time Series Analysis on Flame Length of Micron Coal Dust Explosion with Ignition Delay Time

Abstract To discuss P m variation law with t in large scale coal dust explosion equipment, shock wave pipe is used and a 60 linear test system built. of measured by flame collection system. ARIMA ( p,q,d ) time series analysis model built, differential methods are to eliminate upward trend the maximum length data. By calculating autocorrelation, partial correlation coefficient, value AIC SBC, autoregression moving average orders got. Model parameters estimated least squares method. 20 groups prediction show that: have high precision, optimum ignition delay consistent value, which realizes reasonable on fluctuation characteristics . This can reasonably predict up down fluctuations local data, reflecting superiority model, not achievable general models. The results this study greatly reduce number tests, save material costs provide new idea for studying relationship between length.