Quantitative effects of composting state variables on C/N ratio through GA-aided multivariate analysis

Area of Science:

• Environmental Science
• Biotechnology
• Data Science

Background:

• The C/N ratio is crucial in composting, influenced by various state variables.
• Understanding these nonlinear relationships is vital for optimizing food waste composting.

Purpose of the Study:

• To develop and validate a novel method, genetic algorithm aided stepwise cluster analysis (GASCA), for analyzing food waste composting.
• To identify and rank the key state variables affecting the C/N ratio during composting.

Main Methods:

• Applied GASCA to experimental data from six bench-scale food waste composting reactors.
• Utilized genetic algorithms (GA) for variable and parameter optimization within stepwise cluster analysis (SCA).
• Introduced a proxy table to enhance computational efficiency, reducing calculation time by approximately 70%.

Main Results:

• GASCA produced cluster trees with reduced size and enhanced prediction accuracy compared to conventional SCA.
• Identified NH₄+-N concentration, moisture content, ash content, mean temperature, and mesophilic bacteria biomass as critical factors influencing the C/N ratio.
• Established a descending order of influence: NH₄+-N concentration > Moisture content > Ash Content > Mean Temperature > Mesophilic bacteria biomass.

Conclusions:

• GASCA is a powerful tool for modeling complex relationships in composting and other environmental processes.
• Ammonium nitrogen concentration, temperature, and moisture are critical drivers of C/N ratio variation in food waste composting.
• Coupling direct search algorithms with multivariate analysis offers a promising approach for environmental process modeling.