An electromechanical coupling model for performance enhancement in diesel generator sets

Abstract

Diesel generator (DG) sets are crucial power supply equipment in diverse power systems. The power generation process of DG sets involves an intricate conversion of mechanical energy to electrical energy, which requires tight collaboration and interplay between the mechanical and electrical subsystems. This interaction significantly impacts the safety and efficiency of the entire power system. To investigate this interaction behavior and mechanism, this study proposes an electromechanical coupling dynamics model for DG sets that considers the interaction between the torsional vibration of the shaft system and the electrical system. First, an electromechanical coupling model between the diesel engine speed governor system and shaft torsional vibration is established. Next, the synchronous generator and its excitation system are modeled, and the interrelationship between the generator electrical system and the torsional vibration is derived based on Lagrange’s theory. As a result, an overall electromechanical coupling model is developed. This model is validated using experimental data, demonstrating high accuracy in evaluating both the torsional vibration characteristics and electrical dynamics, with a maximum error below 2.50%. Furthermore, this model is employed to study the effect of the dynamic torsional stiffness of the flexible coupling (FC) on electromechanical dynamic responses. The results show that adjusting the stiffness to 105% to 200% of the original design value can effectively mitigate both the electrical dynamic response and torsional vibration response, thereby improving the system performance. Based on these insights, an FC modification is implemented, resulting in a reduction in the steady-state frequency band from ±0.535% to ±0.175%. This successfully addresses the issue of the steady-state frequency band exceeding the ±0.50% threshold in the original design and significantly improves the DG set performance. This study provides valuable guidance to the parametric design and performance analysis of DG sets.