论文部分内容阅读
从辅料运行特性的角度对烧结工艺做了总体分析,并对烧结矿的质量指标及其影响因素做了研究,在此基础上运用了一种带动量项和变学习率的BP神经网络算法建立了烧结矿质量预测模型。模型预报结果表明,用拓扑结构为15-20-4的BP神经网络和0.000 191的网络误差进行训练后,模型的命中率在83.3%以上,充分展示了基于辅料运行特性的烧结矿质量预测模型的准确性和有效性。
From the point of view of the operating characteristics of the auxiliary materials, the sintering process is analyzed in general, and the quality index of sintering ore and its influencing factors are studied. Based on this, a BP neural network algorithm with momentum and learning rate is established The sinter quality prediction model. The results of model prediction show that the hit rate of the model is above 83.3% after BP neural network with topological structure of 15-20-4 and network error of 0.000191, which fully shows the prediction model of sinter quality based on the operation characteristics of auxiliary materials The accuracy and effectiveness.