食品和生物资源转化过程预测技术入门,Journal of Food Process Engineering

当前位置： X-MOL 学术 › J. Food Process Eng. › 论文详情

Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)

食品和生物资源转化过程预测技术入门
Journal of Food Process Engineering ( IF 2.7 ) Pub Date : 2023-03-17 , DOI: 10.1111/jfpe.14325
Jason Sicard ₁ , Sophie Barbe ₂ , Rachel Boutrou ₃ , Laurent Bouvier ₄ , Guillaume Delaplace ₄ , Gwenaëlle Lashermes ₅ , Laëtitia Théron ₁ , Olivier Vitrac ₆ , Alberto Tonda ₇

Affiliation

为了满足当前社会对更可持续的转化过程和生物资源的需求，必须优化这些过程并开发新的过程。可以预测各种系统（原材料、食品或工艺属性）的演变，以优化生物质的使用，以获得更好的质量、安全性、经济效益和可持续性。预测建模可以指导必要的变化并影响工业、政府政策和消费者决策。然而，实现良好的预测能力需要对模型和模型验证进行反思，这可能很困难。本综述旨在通过介绍目前用于食品和相关生物产品预测科学的技术，帮助科学家开始进行预测。第一的，指南帮助读者启动预测过程以及最终提示和有关所涉及风险的警告，特别关注关键的验证步骤。然后介绍了三大类技术：经验、机械和人工智能（或“数据驱动”）。对于每个类别，当前预测技术的优点和局限性都根据它们当前的应用领域进行了解释，并通过文献研究和详细示例进行了说明。因此，本文为工程研究人员提供了有关预测建模的信息，这是食品和非食品生物资源过程优化方面的最新相关进展。机械和人工智能（或“数据驱动”）。对于每个类别，当前预测技术的优点和局限性都根据它们当前的应用领域进行了解释，并通过文献研究和详细示例进行了说明。因此，本文为工程研究人员提供了有关预测建模的信息，这是食品和非食品生物资源过程优化方面的最新相关进展。机械和人工智能（或“数据驱动”）。对于每个类别，当前预测技术的优点和局限性都根据它们当前的应用领域进行了解释，并通过文献研究和详细示例进行了说明。因此，本文为工程研究人员提供了有关预测建模的信息，这是食品和非食品生物资源过程优化方面的最新相关进展。

"点击查看英文标题和摘要"

A primer on predictive techniques for food and bioresources transformation processes

To meet current societal demand for more sustainable transformation processes and bioresources, these processes must be optimized and new ones developed. The evolution of various systems (raw material, food, or process attributes) can be predicted to optimize the uses of biomass for better quality, safety, economic benefit, and sustainability. Predictive modeling can guide the necessary changes and influence industrials, governmental policies and consumers decision-making. However, achieving good predictive capability requires reflection on the models and model validation, which can be difficult. This review aims to help scientists begin to predict by presenting the techniques currently used in predictive science for food and related bioproducts. First, a guideline helps readers initiate a prediction process along with final tips and a warning about the risks involved, with a particular focus on the crucial validation step. Three broad categories of techniques are then presented: empirical, mechanistic, and artificial intelligence (or “data-driven”). For each category, the advantages and limitations of current techniques for prediction are explained in light of their current domains of applications, illustrated with literature studies and a detailed example. Thus this article provides engineering researchers information about predictive modeling which is a recent relevant development in optimization of both food and nonfood bioresources processes.

更新日期：2023-03-17

点击分享查看原文

点击收藏

阅读更多本刊新发论文本刊介绍/投稿指南