基于神经网络红枣红外辐射干燥预测模型建立  

Dates Established Neural Network Prediction Model Based on Infrared Radiation Drying

作　　者：李超新[1] 张学军[1,2] 朱自成[1] 

机构地区：[1]新疆农业大学机械交通学院,乌鲁木齐830052 [2]新疆农业工程装备创新设计重点实验室,乌鲁木齐830052

出　　处：《农机化研究》2015年第5期220-223,共4页Journal of Agricultural Mechanization Research

基　　金：新疆自治区"十二五"科技重大专项(201130102-4);新疆农业大学前期自助项目(XJAU201226)

摘　　要：针对红枣红外辐射干燥含水率的变化具有非线性和时变性、很难利用现有的模型构造一个数学模型来描述其变化规律的问题,利用Mat Lab神经网络工具箱和红枣红外辐射干燥特性试验数据建立了神经网络预测模型。通过对实测值和模型预测值进行分析研究,得出利用BP神经网络可以较快速、准确地建立模型来描述含水率的变化规律,且模型的预测值与试验测试值误差较小,能很好地实现在线预测的效果。Changes in dates for infrared radiation drying moisture nonlinear and time -varying , difficult to use existing models used to construct a mathematical model to describe the variation .Articles using MATLAB neural network toolbox and dates infrared radiation drying characteristics of test data to establish a neural network prediction model .Through the measured and predicted values were analyzed studies that illustrate the use of BP neural network can be more rapid ,accu-rate modeling is used to describe the variation of moisture content ,and the predictive value of the model with experimental test error is small , well implementation of online prediction results .

关 键 词：红外辐射 含水率 预测模型 BP神经网络 红枣 

分 类 号：S375[农业科学—农产品加工]