基于神经网络的花生红衣原花色素提取研究  

Study on Extraction of Proanthocyanidins from Peanut Red Skin Based on Neural Network

作　　者：黄燕[1] 张文凤 HUANG Yan;ZHANG Wen-feng(Jiujiang Vocational University,Jiujiang 332000,China;Jiangxi University of Engineering,Xinyu 338000,China)

机构地区：[1]九江职业大学,江西九江332000 [2]江西工程学院,江西新余338000

出　　处：《中国调味品》2022年第5期193-196,共4页China Condiment

基　　金：2020年度江西省教育厅科学技术研究项目(GJJ203909)。

摘　　要：文章以花生红衣原花色素提取工艺过程为研究对象,利用正交试验的方式获取神经网络训练模型的输入参数,包含提取剂、提取温度、提取剂浓度、提取时间以及料液比值,同时将花生红衣色素综合评价值作为输出参数,采用神经网络预测花生红衣色素提取过程的最佳工艺条件和综合评价值。神经网络预测结果表明,花生红衣色素提取时最佳提取温度为55℃,采用浓度为60%的乙醇作为提取剂,在每35 mL溶液中加入1 g提取物,经过3 h提取可获取最高综合评价值。Taking the extraction process of proanthocyanidins from peanut red skin as the research object,the orthogonal experiment is used to obtain the input parameters of neural network training model,including extractant,extraction temperature,extractant concentration,extraction time and solid-liquid ratio.The comprehensive evaluation value of peanut red skin pigment is taken as the output parameter,and the neural network is used to predict the optimal extraction process conditions and comprehensive evaluation value in the extraction process of proanthocyanidins from peanut red skin.The neural network prediction results show that the optimal extraction temperature is 55℃,and the highest comprehensive evaluation value could be obtained by adding 1 g extract into every 35 mL solution with 60%ethanol as the extractant and extracting for 3 h.

关 键 词：花生红衣 原花色素 神经网络 正交试验 

分 类 号：TS202.3[轻工技术与工程—食品科学]