瓦斯异常涌出混沌时间序列预测模型的建立

Establishment of prediction model of abnormal gas emission based on chaotic time series

运用混沌理论对平煤十矿的实际瓦斯涌出数据进行了分析处理,采用基于关联积分的C-C方法确定了重构空间的时间延迟和嵌入维数,并对时间序列数据进行相空间重构,利用最小数据量法确定了时间序列的最大Lyapunov指数;运用混沌理论加权一阶局域预测方法,建立了混沌时间序列瓦斯异常涌出预测模型;并利用平煤十矿己15-24080掘进工作面31d的瓦斯实际浓度数据对该模型进行了预测效果检验。结果表明:时间序列的最大Lyapunov指数大于零,证明了时间序列数据具有混沌特征;模型中瓦斯异常涌出的预测发生时间和实际发生时间比较吻合,预测精度达93%。预测模型的可靠性为制定煤矿瓦斯防治措施和采取安全防护措施提供了理论依据。

Based on chaos theory, the original gas concentration data of No. 10 Mine of Pingdingshan Tian＇an coal industry Co. , Ltd, was processed, the phase space of time series was reconstructed after estimating the embedding dimension and delay time through C- C method, and the method of small data was used to calculate the largest Lyapunov exponent. With chaotic adding-weight one-rank local- region method, a prediction model of abnormal gas emission was established. Actual gas concentration data of 24080 driving work face in 31 days was applied to inspect the forecast effect. The results showed that the largest Lyapunov exponent was greater than zero,which il- lustrated the chaotic characteristics in time series data and the prediction accuracy reached 93%. The reliable prediction model provided theoretical basis in making coal mine gas prevention and control measures and undertaking safety protection.