AI-Based Steam Estimation for Excess Air Ratio Optimization in Industrial Boilers

Optimizing the excess air ratio in industrial boiler systems is essential for enhancing combustion efficiency and reducing emissions. This paper proposes an AI-based approach for estimating steam output using boiler operation data collected from real industrial environments. A deep learning model is designed to learn the complex and nonlinear relationships between various operational variables, including external environmental factors, and steam output. By leveraging real-time field data, the proposed model enables accurate steam prediction under diverse operating conditions. Experimental results show that the AI-based model achieves high prediction accuracy and can infer the optimal air flow control setting via inverter to maximize steam output under varying load conditions.