Prediction of cooling moisture content after cut tobacco drying process based on a particle swarm...

Area of Science:

• Agricultural Engineering
• Data Science
• Quality Control

Background:

• Cigarette quality is significantly influenced by stable moisture content.
• Cooling moisture content post-cut tobacco drying is a critical factor for this stability.

Purpose of the Study:

• To develop an accurate prediction model for cooling moisture content in cut tobacco.
• To enhance the stability of moisture content in the final cigarette product.

Main Methods:

• A particle swarm optimization-extreme learning machine (PSO-ELM) algorithm was employed.
• Historical production data from Honghe cigarette factory was utilized.
• The PSO-ELM model was compared against Multiple Linear Regression (MLR), Support Vector Machine (SVM), and traditional Extreme Learning Machine (ELM).

Main Results:

• The PSO-ELM algorithm achieved the highest prediction accuracy.
• The PSO-ELM method exhibited the lowest average prediction standard error.
• The proposed method outperformed the comparative algorithms in prediction performance.

Conclusions:

• The PSO-ELM algorithm offers a robust and accurate method for predicting cooling moisture content.
• This approach provides a novel solution for ensuring moisture content stability in cut tobacco processing.