论文部分内容阅读
煤自燃是煤氧复合的结果,在不同温度下煤氧复合的耗氧速率及CO、CO2产生率与煤的实验自然发火期之间存在复杂的对应关系,采用S型函数的前向多层人工神经网络来描述这种对应关系,用煤自然发火实验测定的数十个煤样的自然发火期及不同温度下耗氧速率及CO、CO2产生率作为训练样本,用BP算法对网络进行训练,得到了神经元间的联结强度.通过少量煤样程序升温氧化实验得到不同温度下煤样的耗氧速率及CO、CO2产生率,将其代入此人工神经网络程序就可以确定煤的实验自然发火期.该方法实验时间短、用煤量少得多,结果与实际吻合.
Spontaneous combustion of coal is the result of coal-oxygen complex. At different temperatures there is a complex correspondence between the oxygen consumption rate of coal-oxygen complex and the CO and CO2 production rates and the experimental spontaneous combustion period of coal. The forward multi-layer Artificial neural network to describe this correspondence between coal spontaneous combustion experiments measured dozens of samples of natural spontaneous combustion and oxygen consumption rates at different temperatures and CO, CO2 production rate as a training sample, using the BP algorithm to train the network , The strength of the connection between neurons was obtained.A small amount of coal sample temperature oxidation experiments were obtained under different temperatures, oxygen consumption rate of CO 2 and CO, CO 2 generation, which will be substituted into this artificial neural network program can determine the experimental nature of coal The ignition period of the method is short, the use of coal is much less, the result is consistent with the actual.