摘要
试验探究了不同温度(50、60、70、80℃)和切片直径(22、29、36、43 mm)对甘薯切片热风干燥特性曲线、有效水分扩散系数及活化能的影响。采用5种薄层干燥数学模型对试验值进行非线性拟合分析,以决定系数(R^(2))、和方差(SSE)和均方根误差(RMSE)为评判标准。结果表明:Weibull模型与其他干燥模型相比能够更好地模拟甘薯切片的热风干燥过程,而BP(back propagation)神经网络模型预测值的决定系数为0.999254,略高于Weibull模型预测值的决定系数0.998756,因此2种模型均能较好地预测甘薯切片的热风干燥特性。甘薯切片的有效水分扩散系数在9.9759×10^(-10)~2.07153×10^(-9) m^(2)/s范围内,其大小随着热风温度的升高和切片直径的减小而增大。通过Arrhenius方程计算得到甘薯切片的干燥活化能为24.469 kJ/mol。
The effects of different temperatures(50,60,70,80℃)and slices diameters(22,29,36,43 mm)on hot air drying characteristic curve,effective moisture diffusivity and activation energy of sweet potato slices were investigated.Five kinds of thin-layer drying mathematical models were used for the nonlinear fitting analysis of the experimental results.The coefficient of determination(R^(2)),the sum of squares due to error(SSE)and root mean square error(RMSE)were used as evaluation criteria.The results showed that the Weibull model could better simulate the hot air drying process of sweet potato slices compared with other drying models.At the same time,the prediction coefficient of back propagation(BP)neural network model for the experimental data was 0.999254,which was higher than that of the Weibull model validation of 0.998756.Both models could well predict the drying characteristics.The effective moisture diffusivity of sweet potato slices was between the range of 9.9759×10^(-10)-2.07153×10^(-9) m^(2)/s,and its value increased with the increase of the hot air temperature and the decrease of slice diameter.According to the Arrhenius equation,the calculated activation energy of drying sweet potato slices was 24.469 kJ/mol.
作者
刘鹤
田友
焦俊华
刘佳敖
吴学红
LIU He;TIAN You;JIAO Jun-hua;LIU Jia-ao;WU Xue-hong(School of Energy and Power Engineering,Zhengzhou University of Light Industry,Zhengzhou 450002,Henan,China)
出处
《粮食与油脂》
北大核心
2022年第8期30-36,共7页
Cereals & Oils
基金
国家自然科学基金项目(51806200)
热流科学与工程教育部重点试验室(西安交通大学)开放基金资助项目(KLTFSE2020KFJJ03)。
关键词
甘薯切片
热风干燥
数学模型
BP神经网络
活化能
sweet potato slice
hot air drying
mathematical model
back propagation(BP)neural network
activation energy
