Enhancing resource recovery from acid whey through chitosan-based pretreatment and machine learning optimization.

Area of Science:

• Dairy Science
• Environmental Engineering
• Materials Science

Background:

• Acid whey, a low pH dairy byproduct, poses disposal challenges due to high organic content.
• Resource recovery from acid whey is crucial for sustainable dairy industry practices.

Purpose of the Study:

• To synthesize chitosan gel for acid whey turbidity reduction.
• To employ machine learning for optimizing the pretreatment process.
• To evaluate the cost-effectiveness and efficiency of chitosan gel for acid whey treatment.

Main Methods:

• Chitosan gel synthesis and characterization.
• Turbidity measurement of acid whey before and after treatment.
• Development and application of machine learning models (Random Forest) for prediction.
• Optimization of pretreatment conditions using Simulated Annealing algorithm.

Main Results:

• Random Forest model achieved a prediction accuracy of 0.78.
• Optimized conditions (2.2% chitosan gel, 24 g/L dosage, pH 4.6, 12 h) yielded 91% turbidity reduction.
• Significant improvement from 71% turbidity reduction before optimization.
• Validation experiments confirmed the model's predictive and optimization capabilities.

Conclusions:

• Machine learning effectively optimizes chitosan pretreatment for acid whey.
• Chitosan gel is a cost-effective and efficient pretreatment agent for acid whey.
• This approach enhances resource recovery potential in the dairy industry.