Variance-based sensitivity analysis for wastewater treatment plant modelling.

Area of Science:

• Environmental Engineering
• Water Treatment Technologies
• Mathematical Modeling

Background:

• Global sensitivity analysis (GSA) is crucial for understanding complex mathematical models.
• Existing GSA methods for wastewater models often struggle with non-linear systems like membrane bioreactors (MBRs).
• MBRs exhibit non-linear kinetics and interactions, necessitating advanced analysis techniques.

Purpose of the Study:

• To apply variance-based GSA, specifically Extended Fourier Amplitude Sensitivity Testing (Extended-FAST), to an MBR model.
• To quantify non-linear effects and interactions among model factors in MBR systems.
• To identify key model factors driving MBR performance.

Main Methods:

• Application of the Extended-FAST method to an integrated activated sludge model (ASM2d) for an MBR.
• Analysis of 21 model outputs and 79 model factors.
• Variance decomposition to assess factor contributions and interactions.

Main Results:

• Significant interactions among model factors were identified.
• Relationships between variables and factors were found to be non-linear and non-additive.
• Key model factors with the highest variance contributions were pinpointed by analyzing variance decomposition patterns.

Conclusions:

• Variance-based GSA methods, like Extended-FAST, are highly effective for MBR modeling due to inherent non-linearities.
• The study highlights the importance of considering both biological and physical processes in MBR models.
• Accurate quantification of non-linear effects is essential for robust MBR system characterization.