Multi-Objective optimisation of oil demisters geometry for Ultra-Low charge ammonia chillers

© 2024 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/bync-nd/4.0/).

Driven by global climate objectives, the pursuit for eco-friendly refrigeration solutions has catalysed impressive advancements in the sector. Ammonia, a historically utilized refrigerant, is now widely adopted in light industrial applications, largely due to the growth of Ultra-Low Charge package chillers. Opportunities, however, persist for refining system components to further diminish refrigerant charge, fostering safer and more efficient solutions. This paper focuses on optimizing the oil separator geometry of an experimentally tested Ultra-Low Charge ammonia chiller. The principal aim and novelty is to lower the specific refrigerant charge while enhancing the separation efficiency and the pressure drop performance. Assessing the separation efficiency involves an analytical model grounded in the “orbit-equilibrium” concept, while the pressure drop is evaluated with consideration of liquid particles presence. Utilizing the validated model, a Multi-Objective Optimization process driven by the eaMuPlusLambda algorithm produces geometries able to reduce the cut-size diameter by 1.4 % up to 15.4 %, pressure drop by 66.7 % up to 89.1 %, and internal volume by 51.3 % up to 68.9 %, compared to the baseline setup. This accomplishment stems from meaningful parameter adjustments, encompassing a 12.1 – 30.0 % decrease in the main diameter, 35.8 % reduction in the main height, 25.4 – 41.2 % contraction in the outlet diameter, 10.9 – 47.1 % augmentation in the inlet diameter, and the vortex finder elimination. Consequently, this translates to a 5% to 14% decrease in the chiller’s specific charge. This study intends an initial step in optimizing oil separators for Ultra-Low Charge ammonia chillers, presenting prospects for specific charge reduction and inciting further exploration through experimental testing or Computational Fluid Dynamics.

CDTI (Centre for the Development of Industrial Technology)

Intarcon S.L