食品科学 ›› 2022, Vol. 43 ›› Issue (17): 50-55.doi: 10.7506/spkx1002-6630-20210717-194

• 基础研究 • 上一篇    下一篇

小麦加工链中重金属镉含量的深度网络预测模型

金学波,张佳帅,郭天洋,王小艺,苏婷立,赖燕群,孔建磊,白玉廷   

  1. (1.北京工商大学人工智能学院,北京 100048;2.北京工商大学 中国轻工业工业互联网与大数据重点实验室,北京 100048;3.北京工商大学食品与健康学院,北京 100048;4.北京服装学院,北京 100105;5.中粮粮油工业(荆州)有限公司,湖北 荆州 434300)
  • 出版日期:2022-09-15 发布日期:2022-09-28
  • 基金资助:
    “十三五”国家重点研发计划重点专项(2017YFC1600605);国家自然科学基金青年科学基金项目(61903009;62006008)

Deep Network Based Prediction Model for Heavy Metal Cadmium Content in Wheat Processing Chain

JIN Xuebo, ZHANG Jiashuai, GUO Tianyang, WANG Xiaoyi, SU Tingli, LAI Yanqun, KONG Jianlei, BAI Yuting   

  1. (1. College of Artificial Intelligence, Beijing Technology and Business University, Beijing 100048, China; 2. Key Laboratory of Industrial Internet and Big Data in China Light Industry Address, Beijing Technology and Business University, Beijing 100048, China; 3. College of Food and Health, Beijing Technology and Business University, Beijing 100048, China;4. Beijing Institute of Fashion Technology, Beijing 100105, China; 5. China Grain and Oil Industry (Jingzhou) Co., Ltd., Jingzhou 434300, China)
  • Online:2022-09-15 Published:2022-09-28

摘要: 镉污染范围广、毒性大、易侵入,被认为是最具危害性的重金属之一,人长期摄入过量的镉会引起很多疾病甚至癌症。因此,在小麦加工链中预测镉元素含量的变化趋势,制定相应对策来降低其危害,具有重要的现实意义。针对小麦加工链镉含量数据含有强非线性、强随机性噪声而导致的传统建模拟合度不高等问题,本研究提出一种基于正则化方法的深度预测模型。首先,利用门控循环单元(gated recurrent unit,GRU)建立深度预测模型。其次,使用正则化方法修改模型的损失函数,通过加入噪声惩罚项来淡化训练时模型对于噪声的拟合程度,减小噪声对模型预测性能的影响。最后,使用贝叶斯优化方法进行超参数的选择,保证所建立的模型能够准确地预测小麦加工链各环节中的镉含量。本研究的预测结果表明,如果原麦中镉含量小于0.1 mg/kg,则经过加工的成品小麦粉也基本满足GB 2762—2017《食品安全国家标准 食品中污染物限量》的要求。

关键词: 小麦加工链;镉;预测模型;门控循环单元;贝叶斯优化

Abstract: Cadmium is considered one of the most harmful heavy metals because of its wide range of dangerous contamination, high toxicity, and easy invasion. Long-term intake of excessive cadmium can cause many diseases including cancers. Cadmium content prediction in the wheat processing chain is of great practical importance for developing countermeasures to reduce its hazards. In this paper, we proposed a deep learning prediction model using the regularization method to address the problem that the data of cadmium content in the wheat processing chain contain strong nonlinear and random noises, which leads to poor fitness of the traditional model. Firstly, a gated recurrent unit (GRU) was used to build the deep learning prediction model. Secondly, the loss function of the model was modified using the regularization method to reduce the impact of noise on the prediction performance of the model by adding a noise penalty term to fade out the noise fit of the model during training. Finally, a Bayesian optimization method was used to select the hyperparameters to ensure that the model could accurately predict the cadmium content at each stage of the wheat processing chain. The prediction results show that flour made from wheat grains with a cadmium content less than 0.1 mg/kg can basically meet the requirements of the national standard (GB 2762-2017).

Key words: wheat processing chain; cadmium; predictive models; gated recurrent unit; Bayesian optimization

中图分类号: