• DocumentCode
    3298760
  • Title

    Analysis of the effect of Daqu´s physical and chemical parameters on liquor production yield using rough sets and fuzzy cluster

  • Author

    Sun, Xingbo ; Tang, Xiuhua ; Tao, Xuerong ; Hao, Ming

  • Author_Institution
    Dept. of Electron. Eng., Sichuan Univ. of Sci. & Eng., Zigong, China
  • fYear
    2010
  • fDate
    25-27 June 2010
  • Firstpage
    350
  • Lastpage
    353
  • Abstract
    Daqu is important for Chinese liquor production, and it determine the yield as well as quality of Chinese liquor. Daqu can be evaluated through the physical and chemical indicators such as the esterifying power(EP), liquefying power(LP), saccharifying power(SP), fermenting power(FP), acidity, starch and moisture. These characters have different effect in liquor production. Duqu´s significant physical and chemical attributes affected on liquor production yield are studied using rough set theory. The information table has been built with these seven physical and chemical indicators of Daqu as conditional attributes, with the Chinese liquor production yield as decision attribute. Total of 180 samples from nine wine cellar are collected and preprocessed using fuzzy cluster, and the cluster centers are obtained and used as condition and decision attributes in information table after being quantized. The table has been reduced, then it is shown that the acidity, liquefying power and fermenting power are the main indicators impacting on the liquor production yield. Based on the conception of weight, we find that the weights of three indicators are 23%, 30.8% and 46.2%, which indicates that, the fermenting power has most effect on the liquor yield.
  • Keywords
    Chemical analysis; Chemical products; Chemical technology; Clustering algorithms; Data engineering; Fuzzy logic; Fuzzy sets; Power engineering and energy; Production; Rough sets;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Educational and Network Technology (ICENT), 2010 International Conference on
  • Conference_Location
    Qinhuangdao, China
  • Print_ISBN
    978-1-4244-7660-2
  • Electronic_ISBN
    978-1-4244-7662-6
  • Type

    conf

  • DOI
    10.1109/ICENT.2010.5532155
  • Filename
    5532155