Three-dimensional (3D) unsteady Reynolds-averaged Navier–Stokes (URANS) flow simulations are conducted to investigate the highly unsteady flow field at part load operation of a centrifugal pump. By the availability of unsteady flow field measurement data in the impeller wake region, a thorough validation of the simulation method is performed. Grid independence of the results is ensured. Unsteady characteristics in terms of head and shaft power as well as transient blade loads are evaluated to assess the unsteady pump performance. Significant mis-loading of the blading is revealed when one blade passes the volute tongue and associated with the strong unsteady and 3D flow field in the impeller-volute tongue region. Negative radial velocity in the tongue region is the origin of a vortex at the blade pressure side and a subsequent pressure drop that leads to even temporally negative blade loading. The results provide a detailed insight in the complex part load flow field that might be utilized for an improved pump design. As a valuable secondary outcome, a comparison of results obtained by two widely used computational fluid dynamics (CFD) codes for pump flow simulation is provided, i.e., the commercial code ansyscfx and the branch foam-extend of the open source software openfoam. It is found that the results of both methods in terms of unsteady characteristics as well as local ensemble-averaged velocity field are consistent.

The combined radiant floor and fan coil cooling (RFCAFC) system is widely used due to its high comfort and large energy saving potential. In this study, as an example, the combined RFCAFC system was studied in a high humidity environment in Jinan, Shandong Province, China. The novelty of the combined RFCAFC system lies in its ability to automatically adjust the water supply temperature of the radiant floor and fan coil in real time according to outdoor meteorological conditions, achieving thermal comfort while eliminating the likelihood of condensation on the radiant floor surface. Days with similar outdoor meteorological conditions were grouped, and the comfort level and hourly cooling performance coefficient (EER_h) of different operating strategies for different outdoor meteorological conditions were monitored along with other evaluation indicators. The RFCAFC had good energy efficiency and comfort in a high humidity room environment. This study showed that the indoor vertical air temperature difference ranged from 1.6 to 1.8 °C, which met the ASHRAE55-2017 standard. The radiant floor surface temperature uniformity coefficient (S) fluctuated between 0.7 and 1.0, and the time it took the radiant floor surface temperature to reach 63.2% of the total variability range (τ₆₃) for different operation strategies based on different outdoor meteorological conditions ranged between 4.4 and 4.7 h, which was within the normal range. The proportion of the total cooling capacity contributed by fan coil cooling under low temperature and high humidity (LH), high temperature and low humidity (HL), and medium temperature and medium humidity (MM) were 68.0%, 73.8%, and 71.7%, respectively. Based on this study, the following recommendations for the combined cooling system can be made: (1) When the outdoor humidity is high, the radiant floor system should be turned on early to provide cooling capacity. When the outdoor temperature is high, the fan coil system should be turned on early to reduce the indoor temperature. (2) To reduce energy consumption and achieve efficient operation of the system, the radiant floor system should be continuously operated to maximize its contribution to the cooling capacity, while the fan coil can be operated intermittently. Natural cooling can be integrated to provide additional cooling capacity to the room in the hours preceding occupation (i.e., 7:00–9:00). (3) The operation strategy of the combined cooling system must be able to respond in real time to changes in outdoor meteorological conditions to prevent discomfort in times of extreme heat or humidity.

In this study, energy loss within a centrifugal pump is investigated by post-processing three-dimensional unsteady flow field through kinetic energy dissipation theory. The three-dimensional unsteady flow field is predicted by solving unsteady Reynolds-averaged Navier–Stokes equations. The kinetic energy dissipation consists of three parts: averaged kinetic energy dissipation, turbulent kinetic energy dissipation, and near-wall revised kinetic energy dissipation. The total value variations of three kinetic energy dissipations in the centrifugal pump with flowrate are investigated and compared. Results show that with the increase in flowrate, the total near-wall revised kinetic energy dissipation gradually increases, the total turbulent kinetic energy dissipation first gradually decreases and then gradually increases, and reaches the minimum value at the design flowrate. The total averaged kinetic energy dissipation is less than the total turbulent and the total near-wall revised kinetic energy dissipations, and the total near-wall revised kinetic energy dissipation is larger than the total turbulent kinetic energy dissipation when the flowrate is larger than 0.75 Q_des. The space variation of the near-wall revised kinetic energy dissipation with flowrate shows that large near-wall revised kinetic energy dissipation mainly occurs at the volute and transfers from the small cross-section casing to large cross-section casing and discharge pipe with the increase in flowrate. The space variations of the turbulent kinetic energy dissipation with time for three flowrates are also discussed. Results indicate that large turbulent kinetic energy dissipation near the volute tongue evidently changes with the rotation of the impeller, particularly in 0.5 Q_des. The large turbulent kinetic energy dissipation gradually expands to the pressure side of the blade when the volute tongue gradually approaches the middle of the impeller blade passage. The large turbulent kinetic energy dissipation transfers from the impeller inlet and outlet to the volute tongue and discharge pipe with the increase in flowrate. The findings of this study can serve as guide to improve the design of centrifugal pumps.

The accurate prediction of residential heat load is crucial for effective heating system design, energy management, and cost optimization. In order to further improve the prediction accuracy of the model, this study introduced principal component analysis (PCA), the minimum sum of squares of the combined prediction errors (minSSE), genetic algorithm (GA), and firefly algorithm (FA) into back propagation (BP) and ELMAN neural networks, and established three kinds of combined prediction models. The proposed methodologies are evaluated using real-world data collected from residential buildings over a period of one year. The obtained results of the PCA-BP-ELMAN, FA-ELMAN, and GA-BP models are compared with the neural network before optimization. The experimental results show that the combined prediction models have higher prediction accuracy. The Mean Absolute Percentage Error (MAPE) evaluation indices of the three combined models are distributed between 5.95% and 7.05%. The FA-ELMAN model is the combination model with the highest prediction accuracy, and its MAPE is 5.95%, which is 2.25% lower than the MAPE of an individual neural network. This research contributes to the field by providing a comprehensive and effective framework for residential heat load prediction, which can be valuable for building energy management and optimization.

The increasing quantity of air pollutants generated by automobiles can cause significant harm in relatively enclosed indoor environments. Studying the distribution of pollutants under different conditions in underground parking garages is of great significance for improving indoor air quality and reducing casualties in the event of a fire. This article presents a geometric model of an underground parking garage based on PHOENICS modeling. The related results of CO concentration distribution and fire temperature distribution under ventilation and fire conditions are obtained. Based on the CO concentration and velocity distribution as well as the temperature distribution during a fire, reasonable suggestions are proposed to improve indoor air quality and reduce casualty rates in fire incidents. The results show that under ventilation conditions, adjusting the position of the induced ventilation fan can maintain CO concentrations below 30 ppm in partitions one to three and below 37 ppm in partitions four to six. The temperature of smoke gases remained below 50 °C during the evacuation time, and only a small area exhibited CO levels exceeding 2000 ppm. The existing ventilation exhaust system provides effective fire protection, as it minimally affects personnel evacuation due to the relatively lower smoke temperature.

This study proposes a control method that integrates deep reinforcement learning with load forecasting, to enhance the energy efficiency of ground source heat pump systems. Eight machine learning models are first developed to predict future cooling loads, and the optimal one is then incorporated into deep reinforcement learning. Through interaction with the environment, the optimal control strategy is identified using a deep Q-network to optimize the supply water temperature from the ground source, allowing for energy savings. The obtained results show that the XGBoost model significantly outperforms other models in terms of prediction accuracy, reaching a coefficient of determination of 0.982, a mean absolute percentage error of 6.621%, and a coefficient of variation for the root mean square error of 10.612%. Moreover, the energy savings achieved through the load forecasting-based deep reinforcement learning control method are greater than those of traditional constant water temperature control methods by 10%. Additionally, without shortening the control interval, the energy savings are improved by 0.38% compared with deep reinforcement learning control methods that do not use predictive information. This approach requires only continuous interaction and learning between the agent and the environment, which makes it an effective alternative in scenarios where sensor and equipment data are not present. It provides a smart and adaptive optimization control solution for heating, ventilation, and air conditioning systems in buildings.