摘要
Predicting NO_(x)in the sintering process of iron ore powder in advance was helpful to adjust the denitrification process in time.Taking NO_(x)in the sintering process of iron ore powder as the object,the boxplot,empirical mode decomposition algorithm,Pearson correlation coefficient,maximum information coefficient and other methods were used to preprocess the sintering data and naive Bayes classification algorithm was used to identify the sintering conditions.The regression prediction model with high accuracy and good stability was selected as the sub-model for different sintering conditions,and the sub-models were combined into an integrated prediction model.Based on actual operational data,the approach proved the superiority and effectiveness of the developed model in predicting NO_(x),yielding an accuracy of 96.17%and an absolute error of 5.56,and thereby providing valuable foresight for on-site sintering operations.
基金
financially supported by the Natural Science Basic foundation of China(Program No.52174325)
the Key Research and Development Program of Shaanxi(Grant No.2020GY-166 and Program No.2020GY-247)
the Shaanxi Provincial Innovation Capacity Support Plan(Grant No.2023-CX-TD-53).
