QGA-RBF神经网络在矿井瓦斯涌出量预测中的应用

Application of QGA-RBF for Predicting the Amount of Mine Gas Emission

煤矿的安全事故中有80%以上为瓦斯事故,为了更加准确的预测瓦斯涌出量,使得煤矿安全进一步得到保障,采用足够的具有代表性的瓦斯检测数据作为样本,利用QGA算法优化RBF神经网络的参数,建立了瓦斯涌出量的预测模型,并使用MATLAB进行仿真研究。结果表明,经过优化后的预测模型较单一的RBF网络模型有更好的预测精度,可以为煤矿瓦斯防治提供理论依据。

Safe accidents in coal mine are aroused more than 80 percent by the excess of gas. In order to make gas emission quantity forecasting result more accurate and guarantee the safety of coal mine, detecting sufficient and typical gas data are collected as samples in this paper. The quantum genetic algorithm is adopted to optimize the pa- rameters of Radial Basis Function neural networks, and the forecasting model used for carrying out the gas emission quantity forecasting is established. The simulating result obtained by using Matlab indicates the optimized forecasting model has an more accuracy forecasting result than the forecasting model based on RBF neural networks. A theoretical basis is provided for the prevention and control of gas accidents in coal mine.