Building Simulation最新文献

The building cross-section shape significantly affects the flow characteristics around buildings, especially the recirculation region behind the high-rise building. Eight generic building shapes including square, triangle, octagon, T-shaped, cross-shaped, #-shaped, H-shaped and L-shaped are examined to elucidate their effects on the flow patterns, recirculation length L and areas A using computational fluid dynamics (CFD) simulations with Reynolds-averaged Navier-Stokes (RANS) approach. The sizes and positions of the vortexes behind the buildings are found to be substantially affected by the building shapes and subsequently changing the recirculation flows. The recirculation length L is in the range of 1.6b–2.6b with an average of 2b. The maximum L is found for L-shaped building (2.6b) while the shortest behind octagon building (1.6b). The vertical recirculation area A_v is in the range of 1.5b²–3.2b² and horizontal area A_h in 0.9b²–2.2b². The L, A_v and A_h generally increase with increasing approaching frontal area when the wind direction changes but subject to the dent structures of the #-shaped and cross-shaped buildings. The area-averaged wind velocity ratio (AVR), which is proposed to assess the ventilation performance, is in the range of 0.05 and 0.14, which is around a three-fold difference among the different building shapes. The drag coefficient parameterized by A_h varies significantly, suggesting that previous models without accounting for building shape effect could result in large uncertainty in the drag predictions. These findings provide important reference for improving pedestrian wind environment and shed some light on refining the urban canopy parameterization by considering the building shape effect.

Deep Reinforcement Learning (DRL)-based control shows enhanced performance in the management of integrated energy systems when compared with Rule-Based Controllers (RBCs), but it still lacks scalability and generalisation due to the necessity of using tailored models for the training process. Transfer Learning (TL) is a potential solution to address this limitation. However, existing TL applications in building control have been mostly tested among buildings with similar features, not addressing the need to scale up advanced control in real-world scenarios with diverse energy systems. This paper assesses the performance of an online heterogeneous TL strategy, comparing it with RBC and offline and online DRL controllers in a simulation setup using EnergyPlus and Python. The study tests the transfer in both transductive and inductive settings of a DRL policy designed to manage a chiller coupled with a Thermal Energy Storage (TES). The control policy is pre-trained on a source building and transferred to various target buildings characterised by an integrated energy system including photovoltaic and battery energy storage systems, different building envelope features, occupancy schedule and boundary conditions (e.g., weather and price signal). The TL approach incorporates model slicing, imitation learning and fine-tuning to handle diverse state spaces and reward functions between source and target buildings. Results show that the proposed methodology leads to a reduction of 10% in electricity cost and between 10% and 40% in the mean value of the daily average temperature violation rate compared to RBC and online DRL controllers. Moreover, online TL maximises self-sufficiency and self-consumption by 9% and 11% with respect to RBC. Conversely, online TL achieves worse performance compared to offline DRL in either transductive or inductive settings. However, offline Deep Reinforcement Learning (DRL) agents should be trained at least for 15 episodes to reach the same level of performance as the online TL. Therefore, the proposed online TL methodology is effective, completely model-free and it can be directly implemented in real buildings with satisfying performance.

In recent years, rapid urban development has led to capsule hotels, sleep pods, and other tiny sleeping spaces that adapt to people’s fast-paced lives, achieving maximum functionality with a very small footprint. However, due to the small space, human metabolic pollutant (such as CO₂) is more likely to accumulate, and the air is not easily circulated. In this paper, a full-size experimental platform is set up with three types of ventilation modes to explore the exclusion efficiency of metabolic pollutants and the overall distribution of age of air under these ventilation modes. The conclusions showed that the mean values of metabolic pollutant exclusion rates for the different ventilation modalities varied very little across the spatial dimensions of the confined space but varied considerably in the area around the head. The double-side attached ventilation method was the most effective in removing human metabolic pollutants, especially in the head region (CN ≥ 0.92), while the single-wall attached ventilation method had the best air exchange efficiency (η ≥ 0.85). This suggests an inconsistent distribution of CO₂ and age of air, which is contrary to general common sense. The conclusions of this paper can guide the design of ventilation for tiny sleeping spaces.

Abstract

Seasonal thermal energy storage (STES) allows storing heat for long-term and thus promotes the shifting of waste heat resources from summer to winter to decarbonize the district heating (DH) systems. Despite being a promising solution for sustainable energy system, large-scale STES for urban regions is lacking due to the relatively high initial investment and extensive land use. To close the gap, this study assesses the potentials of using two naturally available structures for STES, namely valley and ground pit sites. Based on geographical information system (GIS) methods, the available locations are searched from digital elevation model and selected considering several criteria from land uses and construction difficulties. The costs of dams to impound the reservoir and the yielded storage capacities are then quantified to guide the choice of suitable sites. The assessment is conducted for the northern China where DH systems and significant seasonal differences of energy demand exist. In total, 2,273 valley sites and 75 ground pit sites are finally identified with the energy storage capacity of 15.2 billion GJ, which is much larger than the existing DH demand in northern China. The results also prove that 682 valley sites can be achieved with a dam cost lower than 20 CNY/m³. By conducting sensitivity analysis on the design dam wall height and elevations, the choices of available natural structures are expanded but practical issues about water pressures and constructions are also found. Furthermore, the identified sites are geographically mapped with nearest urban regions to reveal their roles in the DH systems. In general, 560 urban regions are found with potential STES units and most of them have STES storage capacities larger than their own DH demand. The novel planning methodology of this study and publicly available datasets create possibilities for the implementations of large-scale STES in urban DH systems.

A radiant floor cooling system (RFCS) is a high-comfort and low energy consumption system suitable for residential buildings. Radiant floor systems usually work with fresh air, and their operating performance is affected by climatic conditions. Indoor and outdoor environmental disturbances and the system’s control strategy affect the indoor thermal comfort and energy efficiency of the system. Firstly, a multi-story residential building model was established in this study. Transient system simulation program was used to study the operation dynamics of three control strategies of the RFCS based on the calibrated model. Then, the performance of the control strategies in five climate zones in China were compared using multi-criteria decision-making in combination. The results show that control strategy has a negligible effect on condensation risk, but the thermal comfort and economic performance differ for different control strategies. The adaptability of different control strategies varies in different climate zones based on the consideration of multiple factors. The performance of the direct-ground cooling source system is better in Hot summer and warm winter zone. The variable air volume control strategy scores higher in Serve cold and Temperate zones, and the hours exceeding thermal comfort account for less than 3% of the total simulation period. Therefore, it is suggested to choose the RFCS control strategy for residential buildings according to the climate zone characteristics, to increase the energy savings. Our results provide a reliable reference for implementing RFCSs in residential buildings.

Abstract

The issue of unoccupied or abandoned homesteads (courtyards) in China emerges given the increasing aging population, rapid urbanization and massive rural-urban migration. From the aspect of rural vitalization, land-use planning, and policy making, determining the number of unoccupied courtyards is important. Field and questionnaire-based surveys were currently the main approaches, but these traditional methods were often expensive and laborious. A new workflow is explored using deep learning and machine learning algorithms on unmanned aerial vehicle (UAV) images. Initially, features of the built environment were extracted using deep learning to evaluate the courtyard management, including extracting complete or collapsed farmhouses by Alexnet, detecting solar water heaters by YOLOv5s, calculating green looking ratio (GLR) by FCN. Their precisions exceeded 98%. Then, seven machine learning algorithms (Adaboost, binomial logistic regression, neural network, random forest, support vector machine, decision trees, and XGBoost algorithms) were applied to identify the rural courtyards’ utilization status. The Adaboost algorithm showed the best performance with the comprehensive consideration of most metrics (Accuracy: 0.933, Precision: 0.932, Recall: 0.984, F1-score: 0.957). Results showed that identifying the courtyards’ utilization statuses based on the courtyard built environment is feasible. It is transferable and cost-effective for large-scale village surveys, and may contribute to the intensive and sustainable approach to rural land use.

As the mining depth increases, the problem of high-temperature thermal damage mainly caused by heat dissipation of surrounding rock is becoming more and more obvious. It is very important to solve the environmental problem of mine heat damage to improve the efficiency of mineral resource exploitation and protect the physical and mental health of workers. One can apply thermal insulation coating on the walls of mine roadways as a means of implementing active heat insulation. In this paper, expanded perlite (EP) and glazed hollow bead (GHB) are used as the main thermal insulation materials, ceramsite and sand as aggregate, plus glass fiber and sodium dodecyl sulfate to develop a new lightweight composite thermal insulation coating through orthogonal experiment method. According to the plate heat flow meter method and mechanical test method, the thermal insulation and mechanical properties of EP-GHB mixed ceramsite coating were studied by making specimens with different parameter ratios, and according to the analysis of the experimental results, the optimal mix ratio of the coating was selected. In addition, Fluent numerical simulation software was used to establish the roadway model, and the thermal insulation effect of the coating in the roadway under different working conditions was studied. The results show that the thermal conductivity of the prepared composite thermal insulation coating material is only 8.5% of that of ordinary cement mortar, and the optimal thickness of adding thermal insulation coating is 0.2 m, which can reduce the outlet air temperature of the roadway with a length of 1000 m by 4.87 K at this thickness. The thermal insulation coating developed in this study has the advantages of simple technology and strong practicability, and has certain popularization and application value in mine heat damage control.

Building surface cool materials are novel materials that can reduce urban heat island intensity and decrease building energy consumption. This study investigated the impact of radiative properties of materials, façade orientation, and morphological parameters on energy consumption in six typical residential neighborhoods in Nanjing, China. The neighborhood energy consumption of 16 application schemes considering the façade orientation factor is compared to determine the best energy-saving scheme. Seasonal and annual energy-saving rates, savings in electricity costs, and the price ceiling for materials per unit area are analyzed. The results show that for low-rise buildings, using cool materials only on the roof can reduce the annual energy consumption by 1%. When cool or super cool materials are also used on the building façade, the annual energy saving rate can be up to 3.4% and 4.3%, respectively. Using cool materials on the south façade of buildings is not recommended due to significant heat loss in winter. Considering savings in electricity costs and the price ceiling for materials per unit area, the price of cool and super cool materials should be less than 3.0 and 3.7 RMB/m², respectively, assuming a lifespan of eight years in Nanjing.

The acceleration of industrialization worsening indoor environments of industrial buildings has drawn more attention in recent years. Natural ventilation can improve indoor air quality (IAQ) and reduce carbon emissions. To evaluate gaseous pollutant levels in industrial buildings for the development of buoyancy-driven natural ventilation, two theoretical models of pollutant flushing (Model I and Model II) are developed based on the existing thermal stratification theory in combination with the mixing characteristics of lower pollutant. The results show that indoor pollutant flushing is mainly dependent on the pollution source intensity and effective ventilation area. The mixing characteristics of lower pollutant has an important effect on pollutant stratification and evolution during ventilation, but it does not change the prediction results at steady state. When the dimensionless pollution source intensity is larger than 1, the pollution source should be cleaned up or other ventilation methods should be used instead to improve IAQ. In addition, the comparisons between Model I and Model II on instantaneous pollutant concentration are significantly influenced by the pollution source intensity, and the actual pollutant concentration is more likely to be between the predicted values of Model I and Model II. To reduce pollutant concentration to a required level, the pollution source intensity should be in a certain range. The theoretical models as well as the necessary conditions for ventilation effectiveness obtained can be used for the ventilation optimization design of industrial buildings.

Abstract

Optimization for the multi-chiller system is an indispensable approach for the operation of highly efficient chiller plants. The optima obtained by model-based optimization algorithms are dependent on precise and solvable objective functions. The classical neural networks cannot provide convex input-output mappings despite capturing impressive nonlinear fitting capabilities, resulting in a reduction in the robustness of model-based optimization. In this paper, we leverage the input convex neural networks (ICNN) to identify the chiller model to construct a convex mapping between control variables and the objective function, which enables the NN-based OCL as a convex optimization problem and apply it to multi-chiller optimization for optimal chiller loading (OCL). Approximation performances are evaluated through a four-model comparison based on an experimental data set, and the statistical results show that, on the premise of retaining prior convexities, the proposed model depicts excellent approximation power for the data set, especially the unseen data. Finally, the ICNN model is applied to a typical OCL problem for a multi-chiller system and combined with three types of optimization strategies. Compared with conventional and meta-heuristic methods, the numerical results suggest that the gradient-based BFGS algorithm provides better energy-saving ratios facing consecutive cooling load inputs and an impressive convergence speed.