Parameter identification of non linear systems for the optimization of biogas production via biological way

Abstract

This thesis explores the Anaerobic Digestion (AD) process model, focusing on optimizing productivity. The International Water Association’s Anaerobic Digestion Model No.1 (ADM1) is a key model in this field. The main challenge is identifying uncertain parameters. To tackle this, a methodology using genetic algorithms (GA) is introduced to fine-tune the parameters of a simplified model, AM2HN. The GA minimizes the proposed objective function on two scales: linear and logarithmic. This methodology, validated through computer simulation, shows significant improvement over traditional techniques.The reduced model AM2 is used to formulate and test a robust control technique (Sliding Mode Control), demonstrating its efficacy and potential. This study contributes significantly to AD process modeling and control.