Experimental Investigation and Component-Level Modelling of a Self-Regulating Small-Scale Vapour-Compression Atmospheric Water Generator
Atmospheric water harvesting using small-scale vapour-compression systems offers a decentralised water source, but performance varies strongly with ambient psychrometric conditions. This study experimentally evaluates a compact propane (R290) atmospheric water generator with series-coupled evaporator and condenser airflow and develops a component-level model of its partially wet fin-and-tube evaporator. Three extended tests were used to characterise operation. Model predictions were compared with cumulative collected water using a dry–wet coil formulation and refrigerant-side capacity matching. The simulated cumulative water production differed from the measured value by 6%. In the two 24 h tests, the unit produced 7.0 L outdoors and 4.5 L indoors using 7.2 kWh in each case, corresponding to specific energy consumptions of 1.03 and 1.60 kWh L−1, respectively. Water production was higher under the more humid inlet conditions. The model predicted higher water production at the lower nominal fan setting, reaching approximately 0.50–0.54 L h−1. Predicted compressor power also increased. The analysed outdoor profile showed its highest production during the nocturnal high-humidity period; this is a profile-specific result rather than an annual operating window. The model provides a basis for comparing operating trends under different inlet conditions
Authors
- Emmanuel Tapia-Brito (ORCID: https://orcid.org/0000-0002-1869-8617)
- Mehmet Tahir ERDİNÇ (ORCID: https://orcid.org/0000-0003-2201-2937)
- Çağrı Kutlu (ORCID: https://orcid.org/0000-0002-8462-0420)
Institutions
- University of Nottingham (GB)
- Necmettin Erbakan University (TR)
- Tarsus University (TR)
Publication Details
- Journal
- Artificial Intelligence for Sustainable Cities
- Published
- 2026-10-29
- DOI
- https://doi.org/10.65582/aifsc.2026.015
- Primary Topic
- Solar-Powered Water Purification Methods
- Type
- article
- Field-Weighted Citation Impact
- 0.00