摘要
利用近红外光谱技术(NIRS)建立一种高效的测定小麦淀粉含量(干基)的检测方法。收集181个小麦样品,随机分成定标集(n=144)和验证集(n=37)。样品分别以颗粒原样、中药粉碎机粉碎及粉碎至0.5 mm进行建模比较。结果表明,3种不同制样方式建立的模型R^(2)均大于0.8,RMSECV均低于0.7,RMSEP均低于0.6,RPD均在2以上,说明3个模型预测效果良好,均能满足日常检测需求。其中,中药粉碎机粉碎和粉碎至0.5 mm的模型预测效果更好一些。但3个模型的RPD均小于3,在后期的使用中仍需添加代表性样品,以提高模型预测能力和稳定性。
An efficient method for the determination of wheat starch content(dry basis)was used by near infrared spectroscopy technique(NIRS).181 wheat samples were collected and randomly divided into calibration set(n=144)and validation set(n=37).The samples were modeled and compared with the original sample,smashed by traditional Chinese medicine pulverizer and smashed to 0.5 mm.The results showed that the R^(2) of the three models were greater than 0.8,RMSECV were less than 0.7,RMSEP were less than 0.6,RPD were more than 2,which indicated that the three models had good prediction effect and could meet the needs of daily detection.Among them,the prediction effect of traditional Chinese medicine pulverizer grinding and grinding to 0.5 mm model was better.However,the RPD of the three models were less than 3,so it is necessary to add representative samples in the later use to improve the prediction ability and stability of the models.
作者
王晓琼
陈丽
向娜娜
纪昌正
潘子幸
杨露
黎鸿彬
谭会泽
张瑜
WANG Xiao-qiong;CHEN Li;XIANG Na-na;JI Chang-zheng;PAN Zi-xing;YANG Lu;LI Hong-bin;TAN Hui-ze;ZHANG Yu(Wens Foodstuffs Group Co.,Ltd.//Key Laboratory of Animal Nutrition and Feed Science of the Ministry of Agriculture,Yunfu 527400,China)
出处
《粮食与饲料工业》
CAS
2021年第6期58-60,共3页
Cereal & Feed Industry
关键词
近红外
小麦
淀粉
测定
NIR
wheat
starch
determination
