Methane gas is emitted during both underground and surface coal mining. Underground coal mines need to monitor methane gas emissions to ensure adequate ventilation is provided to guarantee that methane concentrations remain low under different production and environmental conditions. Prediction of methane concentrations in underground mines can also contribute towards the successful management of methane gas emissions. The main objective of this research is to develop a forecasting methodology for methane gas emissions based on time series analysis. Methane time series data were retrieved from atmospheric monitoring systems (AMS) of three underground coal mines in the USA. The AMS data were stored and pre-processed using an Atmospheric Monitoring Analysis and Database Management system. Furthermore, different statistical dependence measures such as cross-correlation, autocorrelation, cross-covariance, and variograms were implemented to investigate the potential autocorrelations of methane gas as well as its association with auxiliary variables (barometric pressure and coal production). The autoregressive integrated moving average (ARIMA) time series model which is based on auto-correlations of the methane gas is investigated. It is established that ARIMA used in the one-step-ahead forecasting mode provides accurate estimates that match the direction (increase/decrease) of the methane gas emission data.

地下和露天煤矿开采过程中都会排放甲烷气体。地下煤矿需要监测甲烷气体排放，以确保提供足够的通风，以保证在不同的生产和环境条件下甲烷浓度保持在较低水平。地下矿井中甲烷浓度的预测也有助于成功管理甲烷气体排放。本研究的主要目的是开发一种基于时间序列分析的甲烷气体排放预测方法。甲烷时间序列数据取自美国三个地下煤矿的大气监测系统 (AMS)。使用大气监测分析和数据库管理系统存储和预处理 AMS 数据。此外，实施了不同的统计相关性度量，例如互相关、自相关、互协方差和变异函数，以研究甲烷气体的潜在自相关及其与辅助变量（大气压力和煤炭产量）的关联。研究了基于甲烷气体自相关的自回归综合移动平均（ARIMA）时间序列模型。已经确定，在一步提前预测模式中使用的 ARIMA 提供了与甲烷气体排放数据的方向（增加/减少）相匹配的准确估计。研究了基于甲烷气体自相关的自回归综合移动平均（ARIMA）时间序列模型。已经确定，在一步提前预测模式中使用的 ARIMA 提供了与甲烷气体排放数据的方向（增加/减少）相匹配的准确估计。研究了基于甲烷气体自相关的自回归综合移动平均（ARIMA）时间序列模型。已经确定，在一步提前预测模式中使用的 ARIMA 提供了与甲烷气体排放数据的方向（增加/减少）相匹配的准确估计。