Coal Gas Concentration Predication Based on Chaotic Time Series

A novel coal gas concentration predication model is introduces based on the chaotic time series theory in this paper. According to the Takens theorem, the gas concentration phase space is reconstructed, the embedded dimension m and the time delay τ are calculated by C-C algorithm, the Lyapunov exponent λ is solved with wolf method, and the time series neural network prediction model is established. Research results show that the gas concentration time series has a chaotic characteristic when the Lyapunov exponent λ is 0.2392. While the embedded dimension m and the time delay τ are 6, respectively, the original gas concentration changes can be restored with the gas concentration reconstruction in sequence. Therefore the coal gas concentration predication model is feasible to predict gas concentration change in short time.