基于证据理论的煤矿瓦斯涌出组合预测

为提高煤矿瓦斯涌出量预测的准确度,引入证据理论组合预测方法。根据瓦斯涌出量及其主要影响因素间的实验数据,采用3个不同的粒子群神经网络模型对涌出量进行初步预测。并由BP、RBF网络对预测误差及预测点的影响因素进行分析建模,以获取每个模型的可信度。再利用证据理论对其进行合成,确定组合模型的权值,最终实现对瓦斯涌出量的组合预测。实例结果表明,该组合预测方法的平均绝对误差、均方误差分别为18.5%、5.8%,均小于神经网络组合法及等权平均法的相应预测误差,适用于煤矿瓦斯涌出量预测。

Combination Forecast of Gas Emission in Coal Mine Based on Evidence Theory

In order to improve the prediction accuracy of coal mine gas emission,an evidence theory combination forecasting method is applied.According to the experimental data between gas emission and its main influence factors,three different particle swarm optimization-neural network models are used for the initial prediction.BP(Back Propagation)and RBF(Radical Basis Function)networks are selected to establish models for the forecasting errors and the factors,so as to get the credibility of each model.Then the evidence theory is employed to fuse them to obtain the weights of the combination model,thus,the gas emission combination forecasting is fulfilled.Examples results show that the average absolute error and the mean square error of the combination forecasting method are 0.185 and 5.8%,respectively,less than the corresponding prediction error of the neural network and the equal weight method.The method is suitable for gas emission prediction of coal mine.