ADM1 application for tuning and performance analysis of a multi-model observer-based estimator.

Area of Science:

• Biotechnology and Bioengineering
• Environmental Engineering
• Process Control

Background:

• Anaerobic digestion is a complex biological process crucial for waste treatment and biogas production.
• Accurate process monitoring and control are essential for optimizing efficiency and preventing failures.
• Existing models require robust estimation techniques for real-time parameter tracking.

Purpose of the Study:

• To tune and evaluate the performance of a multi-model observer based estimator (mmOBE).
• To utilize the Anaerobic Digestion Model no.1 (ADM1) for simulating various process states.
• To assess the mmOBE's capability in estimating key kinetic parameters for advanced process management.

Main Methods:

• Simulations using the Anaerobic Digestion Model no.1 (ADM1) to represent methanogenic, organic overload, and acidogenic states.
• Development of a multi-model observer based estimator (mmOBE) grounded in the variable structure model (VSM).
• Optimization of mmOBE tunable parameters based on ADM1 simulation outputs, using a one-day data acquisition interval.

Main Results:

• The mmOBE demonstrated excellent convergence rates with ADM1 simulation outputs.
• The mmOBE successfully estimated critical kinetic parameters, including maximal transformation rates for CODs, VFAs, and methane.
• A data acquisition interval of one day proved sufficient for acceptable accuracy due to slow process dynamics.

Conclusions:

• The mmOBE is a viable tool for accurately estimating kinetic parameters in anaerobic digestion.
• Estimated parameters enhance the development of knowledge-based systems for failure diagnosis and trend analysis.
• This approach supports advanced process control strategies for improved anaerobic digestion performance.