基于MATLAB工具箱的开采煤层自燃危险性预测

正确预测开采煤层自燃发火的趋势与危险性,对煤矿安全生产具有重要的指导意义。煤层自燃发火的趋势和危险程度与其影响因素之间存在着复杂的非线性关系,而人工神经网络具有极强的非线性逼近能力,能真实刻画出输入变量与输出变量之间的非线性关系。为准确预测开采煤层自燃发火的危险性,笔者针对反向BP神经网络收敛差的缺点,分别采用基于MATLAB神经网络工具箱中的VLBP和LMBP算法的改进BP神经网络模型对开采煤层自燃的危险性进行了预测。根据开采煤层自燃的特点,选取煤本身自燃倾向性、煤层地质及赋存条件、通风技术条件3个关键影响因素作为开采煤层自燃危险性的评判指标,建立了开采煤层自燃危险性预测的神经网络模型。实际应用效果表明,采用基于MATLAB神经网络工具箱的BP网络模型,能克服一般BP网络收敛较慢的缺点,能加快收敛速度;运用LMBP算法比VLBP算法快,但需较大计算机内存;该模型收敛速度快,准确性高,是一种十分有效的开采煤层自燃危险性预测方法。

Prediction of Spontaneous Combustion Risk in Mining Coal Layer by MATLAB Neural Network Toolbox

Accurate prediction of the tendency of spontaneous combustion risk in mining coal layers is of great importance in production safety of coal mine. There exists a kind of complicated nonlinear relation between the risk of coal layer spontaneous combustion and its influencing factors. The neural network could truly show the nonlinear relation. In order to accurately predict the risk in mining coal layer, a kind of modified BP neural network based on VLBP and LMBP algorithm in MATLAB neural network toolbox is put forward to speed up the network convergence speed. According to the characteristics of spontaneous combustion risk in mining coal layer, three key influencing factors are selected as the judging indexes. Then the model for predicting the risk is built. Practical application demonstrates that the modified BP prediction model based on MATLAB neural network toolbox could overcome the disadvantages of constringency and has fast convergence speed and good prediction accuracy. The analysis and computation show that the computing speed by LMBP algorithm is faster than by VLBP algorithm but needs more memory. And the results show that the model is very efficient in predicting the spontaneous combustion risk.