大红袍花椒对流-辐射并联干燥BP神经网络模拟与品质研究  

BP neural network simulation and quality of convection and radiation parallel drying of Dahongpao Zanthoxylum bungeanum Maxim

作　　者：薛韩玲[1] 王楠 牛婷婷 陆泽华 廖帮海 习红军[1] XUE Hanling;WANG Nan;NIU Tingting;LU Zehua;LIAO Banghai;XI Hongjun(College of Energy and Engineering,Xi’an University of Science and Technology,Xi’an 710000,China;School of Architecture and Civil Engineering,Xi’an University of Science and Technology,Xi’an 710000,China)

机构地区：[1]西安科技大学能源学院,陕西西安710000 [2]西安科技大学建筑与土木工程学院,陕西西安710000

出　　处：《食品与发酵工业》2024年第19期265-273,共9页Food and Fermentation Industries

基　　金：陕西省重点研发项目(2019NY-166)。

摘　　要：为探究热风干燥、热风-红外和热风-微波对流-辐射并联干燥对大红袍花椒干燥特性及品质的影响,实验分析了不同温度、装载量、干燥功率等条件下的大红袍花椒干燥曲线特征,运用BP神经网络进行拟合,并采用感官评价与GC-MS对3种干燥方式干制大红袍花椒挥发油进行分析。结果表明,热风-红外并联干燥的恒速期干燥速率高于热风干燥一个数量级,热风-微波并联干燥时长最短,出现二次升速且降速期不明显;升温、减少装载量或加大微波功率均有利于提高干燥速率和缩短干燥时间。BP神经网络的相关系数R值均能达到0.985以上,均方误差最低可达1.010×10^(-4),平均相对误差值E为4.55%,可很好地描述大红袍花椒的干燥动力学特性,预测大红袍花椒干燥过程含水率准确且迅速。3种干燥方式挥发油分别鉴定出40、39、28种化学成分,热风-微波并联干燥获取大红袍花椒挥发油中烯烃类化合物相对含量最大,但色泽和口感差,热风干燥花椒色泽最佳,挥发油中化合物种类最多,热风-红外并联干燥既缩短了大红袍花椒干燥时间,又能保持其良好品质,为较佳选择。To investigate the effects of hot air drying,hot air-infrared and hot air-microwave convection and radiation parallel drying on the drying characteristics and quality of Dahongpao Zanthoxylum bungeanum Maxim(DZBM),experiments were conducted on the drying kinetics curves under different temperature,loading capacity,and drying power conditions.BP neural network was used for fitting analysis,and sensory evaluation and volatile oil by GC-MS were used to analyze the quality of dried pepper.Results showed that the drying rate during the constant rate period of hot air-infrared parallel drying was one order of magnitude higher than that of hot air drying,and the time of hot air-microwave parallel drying was the shortest,with a second increase in speed and no obvious decrease in speed period.Heating up,reducing loading capacity,or increasing microwave power were all beneficial for increasing drying rate and shortening drying time.The correlation coefficient R value of the BP neural network could reach 0.985,the minimum mean square error(MSE)could reach 1.010×10^(-4),and the average relative error E was 4.55%.Therefore,the BP neural network could well describe the drying kinetics characteristics of DZBM and accurately and quickly predict its moisture content.Three drying methods identified 40,39,and 28 chemical components in the volatile oil.The hot air-microwave parallel drying method obtained the highest relative content of olefins in the volatile oil of DZBM but with slightly poorer color and taste.DZBM dried by means of hot air had the best color and the most types of compounds in the volatile oil.Hot air-infrared parallel drying was a better choice,which not only shortened the drying time of DZBM but also maintained its good quality.

关 键 词：并联 干燥 大红袍花椒 BP神经网络 品质 

分 类 号：TS264.3[轻工技术与工程—发酵工程]