摘要
为了采用神经网络方法预测干燥后稻米食味值,依据稻米的食味值与其主要成分(水分、蛋白质、直链淀粉和脂肪酸)和干燥温度有关这一研究结论,用近红外光谱谷物成分分析仪测定了稻米的主要成分值,用专家模糊评判方法确定稻米的食味值,建立了稻米食味值与其主要成分之间的网络结构模型,误差分析结果表明:该方法可以较好预测稻米的食味值,并分析了干燥条件对稻米理化指标的影响规律,影响程度为干燥温度:-0.7;水分:0.68、脂肪酸:-0.56、直链淀粉:0.48和蛋白质含量:-0.33。
The purpose of the research is to evaluate and analyze the rice taste value of the post-drying paddy rice in this paper. The rice taste value is subject to the main constituent content including the moisture, protein, amylase and fat acid and drying temperature. On the basis of the conclusion, the rice taste value of post-drying paddy rice was evaluated. First, the main constituent content(A) of the rice was measured by near-infrared grain analyzer. Then, the rice taste value(B) was determined by sensory evaluation panel. The mathematical model was developed involving in the A and B based on the Neural Network software (NeuroShell2,V3.0) which has one input layer (five input nodes) and three hidden layers and one output layer (one output node). The analysis of results leads to a conclusion that neural network model can be used to evaluate the rice taste value of the post-drying paddy rice. The degree of the impact on the post-drying rice taste value was obtained,which includes the moisture((0.68)), protein(-0.33), amylose(0.48) and fat acid(-0.56) and drying temperature(-0.7).
出处
《农业工程学报》
EI
CAS
CSCD
北大核心
2004年第2期193-195,共3页
Transactions of the Chinese Society of Agricultural Engineering
基金
中国博士后基金(2002031149)
黑龙江省教育厅科技项目(10531002)资助
关键词
稻米
食味值
近红外光谱
干燥
神经网络
rice
taste value
near-infrared spectra
drying
neural network