ENHANCING SIMULATION EFFICIENCY THROUGH MODAL TRUNCATION: A STUDY ON CONVECTION REACTION MODEL REDUCTION

Abstract

The paper explores the application of the Modal Truncation algorithm in Model Order Reduction, focusing on its effectiveness in reducing high-dimensional mathematical model. The algorithm identifies dominant modes governing dynamic responses, discards the high-order model, and reconstructs a new model with reduced dimensions. Much recent scientific literature has demonstrated the algorithm's versatility in systems stability analysis, dynamic analysis of systems, and other diverse applications. This study applies Modal truncation to the Convection Reaction model, an 84th-order system representing a chemical reaction. Results show that 5th and 6th-order reduced-order models effectively replicate the original system's behavior, with the 6th-order model exhibiting superior accuracy. Visualizations of transient and frequency domain responses provide insights into the reduced-order models' applicability. Based on the H∞ norm, error analysis emphasizes the 6th-order model's accuracy, which is crucial for selecting an appropriate reduced-order model based on desired accuracy in various applications. This study underscores the Modal Truncation algorithm's significance in achieving computational efficiency without compromising simulation fidelity. The continued refinement and application of this method play a crucial role in addressing challenges associated with high-dimensional systems.