Practical identifiability of ASM2d parameters--systematic selection and tuning of parameter subsets

Area of Science:

• Environmental Engineering
• Water Treatment Technologies
• Biochemical Engineering

Background:

• Activated Sludge Model No. 2d (ASM2d) requires parameter calibration for accurate application.
• Over-parameterization of ASM2d leads to non-unique parameter subsets and challenges in calibration.
• Current calibration practices for ASM2d often rely on ad hoc selection and tuning procedures.

Purpose of the Study:

• To introduce a systematic approach for Activated Sludge Model No. 2d (ASM2d) parameter calibration.
• To address the issue of parameter non-uniqueness in ASM2d by employing parameter identifiability analysis.
• To develop a robust method for selecting and tuning critical ASM2d parameters.

Main Methods:

• A systematic approach based on parameter identifiability analysis of parameter subsets was applied.
• The method includes preliminary prior parameter analysis (prior values, uncertainties, pre-selection).
• An iterative parameter subset selection and tuning procedure using three diagnostic measures was employed.

Main Results:

• The diagnostic measures effectively identify the most important model parameters and their interdependencies.
• The approach facilitates the analysis of how fixed parameter values influence parameter estimates.
• A more systematic and interpretable method for ASM2d calibration was demonstrated.

Conclusions:

• The proposed systematic approach enhances the reliability and interpretability of ASM2d parameter calibration.
• Parameter identifiability analysis provides a robust framework for addressing over-parameterization issues in ASM2d.
• This method offers a significant improvement over traditional ad hoc calibration techniques for wastewater treatment models.