基于SOM与BP神经网络的葡萄酒质量评价

Wine Quality Evaluation Based on Self-organizing Feature Map Networks(SOM Network)and BP Neural Network

葡萄酒有着非常高的营养价值,而决定其营养价值的主要因素是葡萄酒的质量指标。葡萄酒的主要质量指标大体可分为感官指标和理化指标。根据酿酒葡萄和葡萄酒的理化指标,对给定的葡萄酒样品进行质量评价。采用自组织特征映射网络(SOM)对给定的酿酒葡萄样品进行分类,并对酿酒葡萄和葡萄酒的理化指标以及葡萄酒的质量进行聚类,同时采用BP神经网络对结果进行了验证。研究表明,红葡萄与红葡萄酒的理化指标之间的相关性比白葡萄与白葡萄酒的理化指标之间的相关性要大得多;酿酒葡萄和葡萄酒的某些理化指标(白葡萄酒的总糖、红葡萄酒的蛋白质和pH值)、葡萄酒的果香、发酵香物质对葡萄酒质量有显著影响。

Wine is of high nutritional value,the major factor determining its nutritional value is the quality index,the index can be divided into two types,such as sensory index and physicochemical index.The quality evaluation of particular wine sample is based on the physicochemical index of wine grapes and wine.During the evaluation,the Self-organizing Feature Map Network(SOM Network)is used to classify particular wine grape sample and cluster the quality of wine and the physicochemical index of grape and wines.The result is validated by BP Neural Network.The results show that the correlation between physicochemical index of red grape and the red wine is higher than that of the correlation between physicochemical index of white grape and the white wine;certain physicochemical index of wine grape and wine(total carbohydrate of white wine,protein of red wine and pH,etc),fruity aroma of wine and fermented aroma substances have profound effect on the quality of wine.