面向多传感器数据融合的白酒发酵评估方法

Liquor fermentation evaluation method for multi-sensor data fusion

针对由传统白酒固态发酵评估方法数据维度单一、可靠性低导致决策结果存在片面性和不确定性问题,提出了一种基于多传感器数据融合技术的白酒固态发酵状态评估方法。利用统计过程控制SPC方法剔除多传感器数据中偏离较大的异常数据,并以自适应加权融合算法将同类传感器数据进行局部融合,将局部融合数据利用模糊集理论建立各环境指标的隶属度函数,对得到的隶属度函数进行基本概率分配并按组合规则进行全局融合。根据决策准则,对融合结果进行判断,得到白酒固态发酵状态评估,将其与传统D-S证据理论、Sun Quan方法和Murphy方法的决策结果进行比较分析。结果表明:该方法有效综合了窖池中各环境参数的不同作用效果,提高了对白酒固态发酵状态评估的准确性,可为白酒生物发酵过程的控制与优化提供有效的技术指导。

Aiming at the problems of one-sidedness and uncertainty of decision-making results caused by single data dimension and low reliability of traditional liquor solid-state fermentation evaluation methods,this paper proposes a liquor solid-state fermentation state evaluation method based on multi-sensor data fusion technology.The statistical process control SPC method is used to eliminate the abnormal data with large deviation from the multi-dimensional sensor data,and then the adaptive weighted fusion algorithm is used to fuse the similar sensor data locally.The local fusion data is used to establish the membership function of each environmental index by using the fuzzy set theory.The basic probability distribution of the obtained membership function is carried out and the global fusion is carried out according to the combination rules.According to the decision criteria,the state evaluation of liquor solid-state fermentation was obtained by judging the fusion results,and the decision results were compared with the traditional D-S evidence theory and other methods.The results indicated that this method more effectively integrated the different effects of various environmental parameters in the cellar,improved the accuracy of the evaluation of the solid-state fermentation state of liquor,and provided strong technical guidance for the control and optimization of the biological fermentation process of liquor.