Conference Agenda

In response to the growing significance of indoor humidity control in energy-efficient buildings, liquid-desiccant systems, capable of independently controlling air temperature and humidity, have emerged as an alternative to traditional vapor compression cooling systems. Subsequently, the integration of a heat pump into the liquid-desiccant system, facilitating control over both desiccant-solution and air temperatures, gives rise to a heat-pump-driven liquid-desiccant (HPLD) air-conditioning system. However, there is a lack of previous studies on model derivation to predict its operating energy consumption in response to real-time variations in outdoor conditions and building thermal loads. Notably, the compressor energy consumption, a primary energy consumer in the heat pump, exhibits a pronounced sensitivity to variations in environmental conditions—encompassing outdoor and indoor conditions, building thermal loads, and desiccant-solution conditions. Relying solely on performance specifications of the compressor under fixed rating conditions poses limitations in accurately predicting the operating energy consumption of the HPLD air-conditioning system in practical building applications.

This study aims to develop an empirical model to predict the compressor energy consumption of the HPLD air-conditioning system required to handle the real-time variations in building thermal loads during summer, utilizing on-site measurement data obtained from an actual building application. The system is installed in a public building located in Seoul, South Korea, and operates by setting target room conditions as 25°C and 0.0105 kg/kg under various outdoor summer conditions. The distribution of training to testing data sets follows an 8:2 ratios. The measurement data during the initial system loading and instances where the target room conditions are not met are deemed unqualified for building thermal loads treatment and thus excluded from model derivation. The model incorporates outdoor temperature and humidity, as well as sensible and latent heat capacity processed by the system, as input factors, with compressor input power designated as the output factor. Herein, the sensible and latent heat capacity processed by the system encompass both building thermal loads and outdoor ventilation loads. This study develops both Polynomial Regression (PR) model to provide an equation-based model and Multi-Layer Perceptron (MLP) model to address unforeseen data complexity and enhance prediction accuracy. The PR model, derived through a forward stepwise regression approach, achieves R-square, root mean squared error, and mean absolute percentage error values of 0.976, 0.068, and 1.84%, respectively. Simultaneously, the MLP model, derived through a hyperparameter optimization for establishing model structure and parameter setting, attains values of 0.98, 0.057, and 1.02% for the corresponding metrics. Upon implementing the PR and MLP models, the predicted percentage errors for seasonal compressor energy consumption during summer are determined as 0.357% and 0.271%, respectively. These results highlight the exceptional accuracy and applicability of the models in real-world scenarios, contributing valuable insights to the dynamic prediction of energy consumption in the HPLD air-conditioning system and promoting its building practices.

This study aims to analyze the theoretical dehumidification effectiveness at different liquid-to-gas ratios and NTU values to demonstrate the feasibility of the mist-atomization dehumidifier. The detailed effectiveness-NTU model was used to estimate the dehumidification effectiveness of both the proposed mist-atomization dehumidifier and the reference packed-bed dehumidifier. The main results of this study is that the effectiveness of dehumidification improves, as NTU and liquid-to-gas ratio increase, and the maximum theoretical dehumidification effectiveness is 1.0 when the liquid-to-gas ratio is greater than 0.2. In the case of 0.2 liquid-to-gas ratio, the maximum effectiveness was 0.42. Based on the theoretical dehumidification effectiveness according to the NTU, the liquid-to-gas ratio of 0.6 and 0.8 was required in the proposed and reference dehumidifier, respectively to achieve the target dehumidification effectiveness. This suggests that the proposed dehumidifier can achieve comparable dehumidification performance to the reference dehumidifier using a smaller liquid-to-gas ratio.

Sustainable dehumidification using liquid desiccants or vapor-selective membranes coupled with sensible cooling can significantly reduce the energy demand in buildings and the associated carbon emissions compared to conventional vapor compression systems. These systems also eliminate synthetic refrigerants with high global warming potential. However, the energy input for dehumidification technology must be accurately estimated. In this paper, we compare the performance of three dehumidification technologies using a first-principles-based system simulation. The modeled systems include a) a combined electrodialysis-thermal regeneration of liquid desiccant dehumidification (ED-LD) system, b) a vacuum membrane dehumidification (VMD) system, and c) a standard vapor compression system. A dew point evaporative cooler sensibly cools the air for ED-LD and VMD systems to supply air in the desired conditions. The simulation includes an overall mass transfer model of the ED-LD module coupled with a heat and mass transfer model of a liquid desiccant dehumidification system. For a design ambient temperature of 35°C and relative humidity of 40%, the baseline ED-LD and VMD systems increase the overall electrical energy requirement by 8.2% and ~130%, respectively, compared to a vapor compression system. Using the ED-LD model, we estimated the membrane sizes and found the limiting inlet LD concentration for dehumidification to be 31.4%. We present optimized ED membrane parameters, including osmotic permeability, diffusion permeability of salt, and conductivity of ion-exchange membranes for electrodialysis to achieve better energy savings than the conventional system. Compared to the vapor compression system, the optimized system configuration can reduce the cooling and dehumidification power requirement by up to 45.7%.

Dehumidification and drying processes play a significant role in the energy consumption of air conditioning and the production of dry materials. Traditionally, water removal involves cooling the airstream below its dew point, requiring substantial energy for phase change and reheating. Vacuum membrane dehumidification, which utilizes vapor-selective membranes and a vacuum pump to separate water vapor without condensing it, holds great potential for energy savings. However, the energy-intensive process of vacuum pumping the permeated gases back to the ambient air hinders the system from realizing its full efficiency. To address these challenges, recent studies have proposed a dual-module humidity pump, incorporating an additional membrane module on the rejection side to keep the pressure ratio low. Despite this innovation, the sub-ambient pressure in the rejection module allows air to permeate into the system, necessitating strategic air rejection to prevent pressure buildup and diffusion barriers. In this study, the performance of a dual-module humidity pump is numerically investigated, incorporating hollow fiber membranes and synchronized operation of a water vapor compressor and vacuum pump. The geometries of hollow fiber membranes are coupled with a partial pressure-driven ε-NTU method to derive accurate water vapor transport under non-ambient conditions in both dehumidification and rejection modules. The rotational speeds of the water vapor compressor and vacuum pump are controlled to adjust the pressure ratio and find the optimal efficiency of the system. The effects of asymmetric module size and geometries are examined to maximize the vapor balance ratio. The results indicate that the synchronized motion of the vacuum pump and compressor can greatly improve system efficiency without recirculating the air for sweep operation. Also, the optimized form factor of the hollow fiber membranes is derived, providing design guidelines for future research and manufacturing processes.