论文部分内容阅读
为准确有效地预测煤层底板突水的危险性,在分析大量观测实例数据的基础上,选取底板含水层水压、煤层采高、隔水层厚度、断层落差、煤层倾角和断层距工作面距离等6项指标作为影响煤层底板突水的初始特征指标。针对指标之间具有相关性的问题,利用主成分分析(PCA)法提取6项特征指标的主成分,将其作为概率神经网络(PNN)的输入向量,建立基于PCA的煤层底板突水危险性的PNN预测模型。选取21组煤矿实测数据作为学习样本,用于训练模型。采用回代估计法对模型回检。利用学习好的模型,预测另外4组矿井突水数据样本。结果表明,该方法有效降低了指标数据相关性,实现了降维,使PNN模型工作复杂度减弱。将该模型应用于工程实例中,所得预测结果准确率为100%。
In order to accurately and effectively predict the risk of water inrush from coal seam floor, based on the analysis of a large number of observational data, the author selected aquifer pressure, coal seam height, thickness of aquifuge, fault drop, coal seam dip and fault distance from the working face And other six indicators as the initial characteristics of coal seam water inrush index. Aiming at the problem of the correlation between the indexes, the principal component analysis (PCA) method was used to extract the six components of the characteristic index, which was used as the input vector to the probabilistic neural network (PNN) to establish the risk of water inrush from coal seam floor based on PCA PNN prediction model. Select 21 groups of coal mine survey data as a learning sample, used to train the model. Use back-to-back estimation method to check the model. Using the learned model, another four sets of mine water inrush data samples are predicted. The results show that this method effectively reduces the correlation of index data and realizes dimensionality reduction, which makes the work complexity of PNN model weakened. Applying the model to the engineering example, the accuracy of the forecast result is 100%.