基于混沌时间序列的神经网络对瓦斯涌出量预测

瓦斯灾害是煤矿中最严重的灾害之一,瓦斯治理是我国煤矿安全的主攻方向。瓦斯涌出量的预测对遏制矿山瓦斯灾害、保证矿山安全和矿山技术经济指标都有重要意义。传统的方法预测精度不高,而神经网络在构建网络模型时具有一定的主观性。将混沌时间序列理论引入瓦斯预测中,为构建神经网络模型提供理论依据。通过实例证明在实际应用中是可行的。

Anticipation of the Quantity of Gushing Gas According to Neural Net of the Muddleheaded Time Series

Gas explosion is one of the serious calamities in the coal mine. Governing the gas is crucial in guaranteeing the safety of the coal mine in our country. The anticipation of the quantity of gushing gas means much to the inhibition of gas combustion in the coal mine, maintaining the safety arid the index of technology and economic in the coal mine. The traditional method is not accurate enough and the neural net is subjective once it is used to construct circuital models. With the introducing the theory of muddleheaded time series into the anticipation of gas, the theoretical basic is provided to construct the neural circuital model. This method is improved to be helpful by examples.