论文部分内容阅读
在传统BP神经网络预测模型的基础上,依据灰色理论中的灰色关联度,提出了输出变量各个影响因素的灰色关联度权值,首次建立基于灰色理论的神经网络预测模型,并依据国内某钢厂300组实际生产数据进行仿真试验。试验结果表明:误差绝对值小于5%的炉数有39炉,占总炉数的65.00%;误差绝对值小于10%的炉数共有58炉,占到96.67%。与传统BP神经网络相比,基于灰色理论的神经网络模型的预测精度提高近12.5%,说明基于灰色理论的铁水预处理终点磷含量神经网络预测模型能更精确地反映现场实际水平。
Based on the traditional BP neural network prediction model and gray relational degree in gray theory, the gray relational weight of each influencing factor of output variables is proposed. The neural network prediction model based on gray theory is established for the first time. According to the prediction of domestic steel Plant 300 actual production data simulation test. The test results show that there are 39 furnaces accounting for 65% of the total number of furnaces with error less than 5% and 58 furnaces accounting for 96.67% of the total furnaces with error less than 10%. Compared with the traditional BP neural network, the prediction accuracy of the neural network model based on the gray theory is increased by nearly 12.5%, indicating that the prediction model based on the gray theory can predict the actual level of phosphorus in the field more accurately.