摘要
猪肉在储藏、加工和运输过程中因为腐败会挥发出氨气、硫化氢等,据此选择8个金属氧化物半导体气敏传感器构成检测阵列,运用改进的BP神经网络算法建立猪肉新鲜度智能检测的数学模型,从而构建了猪肉新鲜度检测电子鼻系统。通过检测实验构建样本数据集,并对识别模型进行训练、测试,结果表明该模型对猪肉新鲜度的预测结果与用理化分析方法所得实际结果具有很好的吻合度,预测准确率大于90%。
Ammonia,hydrogen sulfide and other volatile substances are produced because of spoilage while pork in storage,processing and transportation stage.According to this,the study is aimed to construct an electronic nose system to detect the freshness of pork,which is composed of a detect array with eight metal oxide semiconductors and a mathematical model of intelligent detection for pork freshness using an ameliorated BP network algorithm.The sample data are collected from detection experiments and the recognition model has been trained and tested.Result shows that the predictions of pork freshness between the model and the physicochemical analysis method have a good correlation,and the prediction accuracy derived from the model is greater than 90%.
出处
《计算机应用与软件》
CSCD
2011年第9期82-84,143,共4页
Computer Applications and Software
基金
国家科技支撑项目子课题(2006BAD30B03-02)
关键词
模式识别
BP
网络
猪肉新鲜度
电子鼻
总挥发性盐基氮
Pattern recognition BP network Pork freshness Electronic nose TVBN（total volatile base nitrogen）
