离差最大化结合BP神经网络评价烟叶化学品质

张勇刚, 任志广, 徐志强, 刘建国, 张晓兵, 刘化冰, 夏琛, 程昌合

Chemical Quality Evaluation of Flue-Cured Tobacco Based on Maximization of Deviation and BP Neural Network

Zhang Yonggang, Ren Zhiguang, Xu Zhiqiang, Liu Jianguo, Zhang Xiaobing, Liu Huabing, Xia Chen, Cheng Changhe

表2 不同赋权方法的比较

Table 2 Comparison of different weight methods

指标 Index 总糖 Total sugar 还原糖 Reducing sugar 烟碱 Nicotine 糖碱比 Reducing sugar/ nicotine 氯 Chlorine 钾 Potassium 总氮 Total nitrogen 氮碱比 Total nitrogen/ nicotine 熵值Entropy value 0.9227 0.9935 0.9810 0.9692 0.9215 0.9957 0.9985 0.9904 传统熵权法Traditional entropy weight method 0.3398 0.0284 0.0834 0.1356 0.3451 0.0190 0.0066 0.0420 改进熵权法Improved entropy weight method 0.2650 0.0621 0.0979 0.1319 0.2684 0.0560 0.0478 0.0709 AHP 0.0508 0.1016 0.2564 0.2351 0.0824 0.0946 0.0795 0.0995 综合赋权Combination weighting 0.1579 0.0819 0.1772 0.1835 0.1754 0.0753 0.0637 0.0852