Dynamic Modelling and Simulation of a Single-Effect Absorption Chiller

Abstract

Absorption refrigeration systems driven by low-grade thermal energy offer a sustainable alternative to conventional vapour compression chillers, replacing the mechanical compressor with a thermochemical cycle driven primarily by waste heat (Florides et al., 2003). The lithium bromide– water (LiBr/H₂O) working pair is well-suited for single-effect cooling, operating effectively with heat source temperatures of 75–120 °C and making it directly compatible with marine waste heat applications (Salmi et al., 2017a). Because absorption machines possess large thermal inertia, dynamic simulation is essential for predicting transient responses, evaluating part-load performance, and designing effective control strategies (Kohlenbach and Ziegler, 2008a). This thesis presents the development and successful implementation of a dynamic, lumped-parameter simulation model for a single-effect LiBr/H₂O absorption chiller in MATLAB/Simulink. Following the state-space framework of Wen et al., (2019), the model is governed by coupled first-order ordinary differential equations derived from conservation of mass, species, and energy. Thermodynamic properties are computed using the correlations of Pátek and Klomfar, (2006) and Florides et al., (2003), with specific heat capacity derived as the analytical temperature derivative of the enthalpy correlation to ensure thermodynamic consistency. To overcome the numerical instability of the baseline model, two architectural modifications were introduced: a dynamic condenser pressure scaling mechanism proportionally linked to instantaneous liquid refrigerant mass, and an 8-stage spatially discretised counter-flow solution heat exchanger with finite thermal mass at each node. These modifications yield a robust, fully functional integrated model. Simulation results across three parametric scenarios demonstrate stable transient convergence, with a baseline COP of 0.718 (driving heat input is waste heat) establishing a computationally efficient tool for predicting absorption chiller behaviour under varying operational conditions.