Kourosh Fakhari - Biotechnology Group, Chemical Engineering Department, Faculty of Engineering, Tarbiat Modares University, Tehran, Iran
Armin Rahimieh - Biotechnology Group, Chemical Engineering Department, Faculty of Engineering, Tarbiat Modares University, Tehran, Iran
Mohsen Nosrati - Biotechnology Group, Chemical Engineering Department, Faculty of Engineering, Tarbiat Modares University, Tehran, Iran

Autothermal Thermophilic Aerobic Digestion, a widely used wastewater treatment method, uses heat-loving microbes to efficiently break down organic matter in sludge without external heating. While complex, understanding the intricate kinetics of this process is crucial for optimization. This study proposes an Artificial Neural Network model, trained on real data, to predict predicting the kinetic rate constant (KATAD). The model, enhanced through Genetic Algorithm optimization, demonstrates exceptional accuracy (more than ۹۹%). Evaluation reveals temperature as the most influential parameter (CI = ۱.۲۳), followed by the primary to secondary sludge ratio (CI = -۰.۴۷), and concentration with the least impact on KATAD (CI = ۰.۱۹).

Artificial Neural Network, Autothermal Thermophilic Aerobic Digestion, Biodegradation Kinetics, Genetic Algorithm, Rate Constant