The demand for lithium-ion batteries has been steadily increasing throughout the years. In the field of lithium-ion battery manufacturing, the cascading desiccant wheel deep dehumidification (DWDD) system has been commonly employed for deep dehumidification process with supply air dew point temperature below −28 ℃. The DWDD system can be highly energy-intensive, and systematic studies on the coordinated regulation are still limited. Therefore, on-site measurements were firstly conducted to investigate the energy performance and energy saving potential of the DWDD system. It is found that energy efficiency during dehumidification decreases significantly as the dehumidification depth increases. The deep dehumidification stage handles the least proportion of the dehumidification load while having the highest energy consumption compared to conventional and low humidity dehumidification stages. A combination of high regeneration temperature in the first stage and low regeneration temperature in the second stage of the desiccant wheel improves the energy efficiency of the DWDD system. The performance predictive model is then developed using operational data to reveal the nonlinear influence of operating parameters on the deep dehumidification efficiency of the DWDD system. Model predictive control (MPC) based optimization strategies are further proposed and applied in a practical application to achieve energy savings while ensuring stable control of the supply air dew point. By properly setting the regeneration temperature range for the two stages of the desiccant wheel, approximately 8.1% of the cooling consumption and 15.6% of the total heating consumption of the DWDD system can be reduced. This study can provide a theoretical support and a feasible solution for the intelligent energy saving control of the deep dehumidification system.

Flexible resources (FRs) in power systems serve as a critical enabler for integration of high share renewable energy. Conventional planning practices mainly focus on supply-side FRs while neglecting demand-side flexibility potential, which challenges the coordination of multi-source flexibility contributions. This paper presents a comprehensive flexibility evaluation framework integrating both supply- and demand-side resources. First, operational models characterize supply-side FRs (thermal, hydro and pumped storage) and demand-side FRs-including centralized air-conditioning systems with cold storage (CAC-CS), plug-in electric vehicles (PEVs), distributed battery storage (DBS), and time-shiftable appliances (TSA)-with parameters extracted from multi-source datasets. Second, a tri-dimensional flexibility metric system encompassing energy shifting, power balancing, and ramp rate is proposed, anchored by baseline net load. Third, a case study of a provincial power system reveals three key trends as the renewable penetration ratio (REPR) rises to 50%: 1) increasing system flexibility requirements dominated by power surplus and upward net load ramping; 2) shifting roles of coal plants toward deficit injection (surplus metrics dropping by >80%) and enhanced pumped storage utilization (4-10 times rise in surplus and ramping metrics); 3) PEV and CAC-CS account for approximately 50% energy shifting flexibility in each season and show similar variation trends in metrics to pumped storage. The framework enables coordinated source-load flexibility planning, demonstrating demand-side FRs’ capacity to offset conventional flexibility limitations and support renewable integration targets.

For a future carbon-neutral society, it is a great challenge to coordinate between the demand and supply sides of a power grid with high penetration of renewable energy sources. In this paper, a general power distribution system of buildings, namely, PEDF (photovoltaics, energy storage, direct current, flexibility), is proposed to provide an effective solution from the demand side. A PEDF system integrates distributed photovoltaics, energy storages (including traditional and virtual energy storage), and a direct current distribution system into a building to provide flexible services for the external power grid. System topology and control strategies at the grid, building, and device levels are introduced and analyzed. We select representative work about key technologies of the PEDF system in recent years, analyze research focuses, and summarize their major challenges & future opportunities. Then, we introduce three real application cases of the PEDF system. On-site measurement results demonstrate its feasibility and advantages. With the rapid growth of renewable power production and electric vehicles, the PEDF system is a potential and promising approach for large-scale integration of renewable energy in a carbon-neutral future.

The liquid desiccant air-conditioning system is considered as an energy-efficient alternative to the vapor compression system. The dynamic response characteristics of the system under variable cooling load play an important role in the air temperature and humidity control performance of the system. However, the dynamic response characteristics have not been fully revealed in previous studies. Thus, a dynamic model for a heat pump driven liquid desiccant air-conditioning (HPLDAC) system is established to investigate the dynamic response characteristics of the system in this study. Subsequently, experiments were conducted to validate the accuracy of the dynamic model. The simulation results show a good agreement with the experimental data. The simulation results reveal that evaporating water from the solution is a time-consuming process, compared to adding water to the solution. It spends a long time for the HPLDAC system to decrease the high relative humidity of supply air to a low value, which limits the air temperature and humidity control performance of the system. The upper band for the water replenishing value opening (Δφ_up) is a crucial parameter to improve the limitation. When Δφ_up decreases from 1.0% to 0.25%, the time consumed to reduce the supply air relative humidity to the new lower set value can be saved by 30.6%.