煤矿系统安全预测模型与组合预测

Model of system safety forecasting and combination forecasting

为了解煤矿安全预测本质,正确有效地预测煤矿系统的安全状况,基于目前煤矿安全研究现状和手段,从煤矿系统安全预测的内在规律性、有效时间长度及有效性等方面进行了多角度研究,总结出煤矿系统安全预测的本质特征,并研究了安全预测的建模过程.在此基础上,提出了提高预测质量的方法,阐述了组合预测法,最后结合实例,建立了煤矿系统安全预测系统.研究结果表明,采用灰色预测法与自回归法相结合的组合预测模型、自回归法与BP神经网络相结合的非线性组合预测模型均有较高的精度,且非线性组合要好于线性组合.非线性组合预测模型克服了单一预测模型的缺点。

In order to find the essence of coal mine safety forecasting, predict the security situation of mine system effectively and correctly, based on the current status of coal mine safety research and research methods, the internal laws of mine security forecasting system, effective length of time and effectiveness were studied. The essential char- acteristics of coal mine safety forecasting system were summed up and security forecasting modeling process was studied. On this basis, methods to improve the quality of forecasting and the combination forecasting methods were presented. Combining the example, coal mine security prediction system was established. The results show that model of combining the gray forecasting method and the self-regression method and non-linear model of combining self-regression method and BP neural networks have a higher precision and the non-linear is better than the linear combination. Non-linear combination forecasting model overcomes the shortcomings of the single prediction model, and solve the forecast difficulties caused by random changes of the number of safety indicators of system status.