多因素耦合条件下硫化矿自燃神经网络动态预测模型研究

Research on ANN Dynamic Prediction Model for Spontaneous Combustion of Sulfide Ores with Multi-factors Coupling

硫化矿石自燃是多种因素、多场耦合综合作用的结果,是一典型的非线性问题。笔者应用人工神经网络技术,以Matlab软件为平台,通过现场调查和理论分析,建立了矿石含硫量、通风强度、环境温度3因素与硫化矿石自燃之间的预测模型;通过数据样本学习与部分现场监测数据相结合进行模拟,研究表明预测数据与实测结果基本吻合,误差控制在10%以内,取得了较好的效果。该研究为预防硫化矿石自燃提供一个新的思路和方法,具有一定的理论意义和应用价值。

Spontaneous combustion of sulfide ores is caused by the coupling of multi-factors and multifields and it is a typical nonlinear problem. Based on the field survey and theoretic analysis, An ANN （Artificial Neural Network） forecasting model for sulfide ore spontaneous combustion, which takes the three factors of the content of sulfur, ventilation intensity, environmental temperature as the input variables of this model, has been built with the help of Matlab （ Matrix Laboratory） software. After the simulation through samples study and by combining field data, it shows that the predicting result is basically in accordance with the observation data, and the average error can be controlled within ten percent with satisfactory results. The research fruit provides a new approach and a route for preventing the spontaneous combustion of sulfide ore, which is of great significance both theoretically and practically.