A Model for Predicting and Grading the Quality of Grain Storage Processes Affected by Microorganisms under Different...

Area of Science:

• Agricultural Science
• Food Science
• Data Science

Background:

• Grain quality is significantly impacted by storage environments, necessitating accurate monitoring for human health.
• Wheat and corn, vital staple grains, require robust quality assessment during storage across diverse regions.

Purpose of the Study:

• To develop and evaluate a predictive model for grain quality changes during storage.
• To establish a grading system for assessing grain storage quality based on predictive analytics.

Main Methods:

• Constructed a FEDformer-based model for predicting grain quality changes.
• Implemented a K-means++ clustering algorithm for grading storage quality.
• Utilized six key factors affecting grain quality as input variables.

Main Results:

• The developed grain storage quality prediction model demonstrated superior accuracy.
• The model achieved the lowest prediction error compared to existing methods.
• The grading evaluation model effectively categorized storage quality based on predicted and current values.

Conclusions:

• The FEDformer and K-means++ integrated model provides a highly accurate method for predicting grain quality changes.
• This approach offers a reliable solution for monitoring and grading grain quality in various storage environments.
• The findings are crucial for ensuring food safety and optimizing grain storage practices.