摘要
在酒品生产中,质量稳定性的快速鉴别对确保酒品质量、改进生产工艺有着重要的作用。运用气敏传感器阵列技术和神经网络模式分类方法研究了酒品质量稳定性的快速鉴别方法。鉴别试验结果表明,合适的传感器阵列可以测试酒品的特征信息,基于遗传算法的RBF神经网络可以建立正确的酒品质量鉴别模型,进而实现了同种酒品生产过程中质量稳定性的快速鉴别。
In drinks production, quick identification of quality stability is very important for insuring drinks quality and improving production technique. Gas sensor array technology and neural network pattern classification technique are used to study quick identification of drinks quality stability. The results of identification experiments showed that characteristic information of drinks could be tested by appropriate sensor array. The proper identification model of drinks quality could be set up by RBF neural network based on genetic algorithms, and the quick identification of quality stability in one drink production could also be achieved.
出处
《食品科学》
EI
CAS
CSCD
北大核心
2005年第11期210-212,共3页
Food Science
基金
河南省杰出青年基金资助项目(0612000400)
关键词
酒
质量
稳定性
传感器阵列
神经网络
drink
quality
stability
sensor array
neural network
