A Soft Sensor for Simultaneous Prediction of Struvite Purity and Recovery Rate

Area of Science:

• Environmental Science
• Chemical Engineering
• Materials Science

Background:

• Phosphorus is a vital, nonrenewable resource essential for life and industry.
• Struvite (magnesium ammonium phosphate) is a key phosphorus recovery product from wastewater, crucial for sustainable fertilizer production.
• Accurate assessment of struvite purity and recovery rates is critical for its market viability but traditional methods are laborious and hinder automation.

Purpose of the Study:

• To develop a novel, non-destructive method for simultaneously determining struvite purity and phosphorus recovery rate.
• To establish a foundation for real-time, automated monitoring in large-scale struvite recovery processes.
• To validate the efficacy of using physical properties for quality assessment.

Main Methods:

• Analysis of optical microscope images of struvite crystals.
• Integration of saturation index information.
• Application of principal component analysis (PCA) using MATLAB scripts.
• Comparison of model predictions with traditional chemical and instrumental analysis.

Main Results:

• The developed model accurately predicts struvite purity and phosphorus recovery rates.
• The model's predictions closely align with results from conventional analytical techniques.
• Physical properties (crystal images, saturation index) are sufficient for reliable quality prediction.

Conclusions:

• A reliable method for real-time struvite quality monitoring has been established using optical images and saturation data.
• This advancement supports the automation of continuous analysis in phosphorus recovery.
• The findings provide a theoretical basis for improving the efficiency and economic feasibility of struvite fertilizer production.