论文部分内容阅读
针对煤尘固体表面接触角测定过程繁琐和煤尘润湿等级划分不合理的问题,以煤质化学组成及其结构参数共13个影响因子为输入参数,采用两层双曲正切S形函数为激励函数,构建有关煤尘接触角估算及润湿性分级的3层BP神经网络.结果表明,隐含层节点数为10时,估算结果相对误差为0.19%~13.99%,平均相对误差为5.18%,煤尘润湿接触角估算结果与实测结果相关性系数为R2=0.933,煤尘润湿分级正确率达91.67%.BP神经网络模型的接触角估算结果和润湿性分级结果可用于指导煤矿井选择降尘措施.
In view of the complicated process of measuring the solid surface contact angle of coal dust and the unreasonable classification of coal dust wetting grade, 13 influencing factors of coal chemical composition and its structural parameters are taken as input parameters, and the two-layer hyperbolic tangent S-shaped function Incentive function to construct a 3-layer BP neural network for estimating the contact angle of coal dust and classification of wettability.The results show that when the number of hidden layers is 10, the relative error of the estimation results is 0.19% ~ 13.99% and the average relative error is 5.18 %, The correlation coefficient between the estimated result of wet contact angle of coal dust and the measured result is R2 = 0.933, and the correct rate of coal dust wet classification is 91.67% .The contact angle estimation results and wettability grading results of BP neural network model can be used to guide Coal mine dust selection measures.