Component optimization in an alpha-type Stirling engine

Abstract

The increasing integration of intermittent renewable energy sources into the global energy grid necessitates the development of efficient and reliable long-duration energy storage systems. The Electron247, a thermal energy storage device developed by EnergyIntel Services, utilizes an alpha-type Stirling engine for heat-to-power conversion. However, the engine’s baseline performance is a significant bottleneck, operating at a simulated [REDACTED] with a thermal efficiency of [REDACTED] , which limits the overall round-trip efficiency of the system. This thesis presents a systematic approach to improve the power output and thermal efficiency of the Electron247 system through the parametric optimization of its Stirling engine components. A specialized third-order, quasi-steady thermodynamic model was developed to accurately simulate the engine’s performance, incorporating critical loss mechanisms such as imperfect heat transfer, pressure drops, regenerator ineffectiveness, and mechanical friction. The model’s predictive accuracy was validated using Helium as the working fluid, showing strong correlation (average error below 7%) with experimental data from two operational units in Masdar City, Abu Dhabi. Leveraging the validated model, a multi-objective optimization was performed using the Nondominated Sorting Genetic Algorithm II (NSGA-II). The optimization aimed to simultaneously maximize thermal efficiency and power output by varying ten key geometric parameters of the engine’s heater, cooler, and regenerator, subject to manufacturing and system-level constraints. The results produced a Pareto-optimal front of designs offering significant performance gains over the baseline configuration. Analysis of the optimal designs revealed that the regenerator’s geometry (specifically, its total volume and wire mesh characteristics) was the most critical factor in determining engine performance. Notably, the ”Maximum Efficiency” design achieved a thermal efficiency of [REDACTED] , while the balanced ”Closest to Ideal” design improved both power output to [REDACTED] and efficiency to [REDACTED] . Depending on the design choice, from a balanced-performance model to a maximum-efficiency configuration, these enhancements result in an additional 800 to 1,100 tons of CO2 emissions being spared per unit, demonstrating the critical impact of component-level optimization on the viability and environmental benefits of thermal energy storage technologies.