摘要
针对转炉终点神经网络预测模型数据预处理过程中多变量、大样本的特点,介绍并应用了一种识别及检验异常数据的方法;筛选了神经网络模型的数据样本集数据;并初步验证了筛选结果。经过筛选,神经网络预测模型训练集数据训练误差绝对值的平均值及最大值分别下降了26.7%和41%;测试集数据测试误差的平均值及最大值分别下降10%和45%。结果表明,该方法对于转炉预测模型的数据筛选行之有效,对转炉预测模型的进一步完善有一定的实用价值。
Considering the characteristics of multi-variables and large sample in BOF endpoint neural network forecasting model, a method＇of outliers identification and iudgment is introduced. Using this method, the data set of neural network model is selected, and the result of selection is primarily validated. After data pretreatment with this method, the mean and maximum residual absolute value of training set and testing set are decreased by 26.7% ,41% and 10% ,45% respectively. The result shows that the method is effective to improve the neural network model for BOF endpoint forecasting.
出处
《中国冶金》
CAS
2006年第9期27-31,共5页
China Metallurgy
关键词
转炉预测模型
数据预处理
异常值识别
异常值检验
BOF forecasting model
data pretreatment
outlier identification t outlier judgment
