摘要
应用BP神经网络,通过苹果质构特性指标(硬度、可恢复形变、黏性、内聚性、咀嚼性)来预测苹果贮藏品质(出汁率、可溶性固形物、总酸、固酸比)的方法,来建立苹果品质的预测模型。将寒富苹果分别置于温度为0℃和20℃的贮藏条件下,分别测定苹果在贮藏期间的品质的变化。以苹果质构特性指标为输入,品质指标为输出确定网络拓扑结构,训练所建立的苹果品质的神经网络模型。仿真结果表明:该神经网络模型用质构特性指标能预测苹果品质,同时通过2组非样本数据来验证该模型,其预测值与实测值的相对误差在5%以下,故能够实现用质构值评价苹果品质的目的。
Apple storage quality properties（including juice yield, soluble solids, total acid, solid acid ratio） were estimated through the textural properties of apple（including hardness, resilience, adhesiveness, cohesiveness, chewiness）. An artificial neural network model of quality properties was built by the BP neural network. Hanfu apples were storage separately at a temperature of 0 ℃ and 20 ℃, to test the quality changes during storage. The textural properties and the apple quality properties measured were adopted as input and output to establish the BP neural network. The simulated results show that this neural network make a good estimation of apple storage quality properties through textural properties value when tested by two groups of non-sample data, the relative error between the estimation of this model and the measured value is below 5%, which to achieve evaluation of apple quality by the data of TPA measures.
出处
《食品科技》
CAS
北大核心
2012年第6期290-293,298,共5页
Food Science and Technology
关键词
寒富苹果
质构特性
贮藏品质
BP神经网络
预测模型
‘Hanfu＇ apple
textural properties
storage quality
BP neural network
estimation model
