Numerical Simulation Analysis and Full-Scale Experimental Validation of a Lower Wall-Mounted Solar Chimney with Different Radiation Models

Abstract

As a type of passive architectural structure, wall-mounted solar chimneys enhance the natural ventilation volume of a building’s interior, and maximize reductions in the building’s operational energy consumption. They are indispensable in the building’s energy conservation and emission reduction. Therefore, measuring the wall-mounted solar chimney’s flow characteristics and relevant index parameters is particularly important. This paper uses a combination of full-scale experiments and numerical simulation to conduct a detailed analysis of the wall-mounted solar chimney. Four different radiation models, namely DO (discrete ordinates), S2S (surface-to-surface), MC (Monte Carlo), and Rosseland are used for comparison, and the results of the numerical simulation are compared with the experimental data. The results show that the maximum turbulent viscosity of the fluid predicted by the S2S radiation model is higher than that of the MC and DO models by 16.87% and 8.44%, respectively. The errors of the DO radiation model in the midline and glass cover plate direction concerning the experimental results are only 0.33% and 0.15%, respectively. The mistakes of the MC radiation model in these two directions are 0.51% and 0.47%, respectively. The DO radiation model is more suitable in numerical simulation predictions related to the wall-mounted solar chimney.

1. Introduction

As the global economy proliferates, environmental pollution and energy consumption are becoming increasingly serious problems [1,2]. In particular, air conditioning energy consumption accounts for two-thirds of the energy consumption in buildings. Therefore, the study of how to reduce the energy consumption of air-conditioning systems in cooling and heating, and further reduce the overall energy consumption of buildings has become an important research topic in green buildings [3,4].

Due to its dual advantage of being energy-free and pollution-free, natural ventilation uses the thermal pressure caused by the difference in air density, and the wind pressure caused by the wind to promote airflow [2,3,5]. This brings fresh outdoor air into the room, reduces the temperature and humidity, and improves the quality of the thermal environment inside. Therefore, natural ventilation is essential to a building’s energy-saving strategy. Through scientific and rational building design, natural ventilation can be used to optimize a building’s energy consumption, and improve the comfort of the occupants [3,6]. This enhances the energy efficiency of the building and, at the same time, improves the occupants’ quality of life, achieving care both for the environment and for people [5,7,8].

Traditional natural ventilation methods, such as opening windows and doors, and directing draughts through the hall can help fresh outdoor air to enter the house. However, people’s living habits, and architectural design often influence their effectiveness, making the ventilation efficiency unsatisfactory [9,10]. However, due to technological developments, new natural ventilation techniques, such as atrium ventilation, ventilated roofs, and ventilated chimneys can optimize airflow, and improve the ventilation efficiency to a greater extent [11,12]. Of all the new ventilation technologies, solar chimneys stand out for their clever use of solar energy to enhance and organize natural ventilation, achieving a perfect combination of saving energy and protecting the environment. Therefore, an in-depth study of the flow characteristics of solar chimneys, and their potential application in buildings, could effectively improve the thermal environment in buildings, and reduce buildings’ energy consumption, which is of great significance in promoting the development of green buildings, and improving buildings’ energy efficiency [13].

As shown in Figure 1, wall-mounted solar chimney ventilation is a sustainable technology that uses the interaction of solar radiation and the natural environment to improve buildings’ indoor and outdoor air. This approach effectively improves the indoor air quality, adjusts temperatures, and saves energy, by reducing the reliance on traditional mechanical cooling and air conditioning equipment. Wall-mounted solar chimneys work differently to traditional solar ones [14,15]. A conventional solar chimney relies on convection currents generated by the rising heat of the air in the collector area to boost the chimney. On the other hand, a wall-mounted solar chimney combines the solar chimney with a wall of a building. The sun directly irradiates the wall, which absorbs the solar heat and transfers it to the air in contact with the wall [14,16]. The heated air decreases in density, generating buoyancy and rising, creating natural convection, and driving the air exchange between the interior and exterior, to achieve indoor ventilation. This design not only effectively directs and utilizes the sun’s heat, reducing the reliance on mechanical ventilation equipment, but also improves the indoor air quality, and optimizes the building’s thermal environment, saving energy, and reducing the building’s operation and maintenance costs [16,17].

In research studies, experts have sought the best way of exploring the design of wall-mounted solar chimneys to enhance the natural ventilation of buildings. Since 1990, research into solar chimneys has focused on the design of the chimney geometry, and chimney design angles [18,19]. At the same time, the methodology has been minimal, with full-scale and small-scale experiments being carried out. However, the data obtained from the experiments are not as good as they could be, and Khanal points out that the results are not universally valid for the solar chimney in question, as the experimental conditions and the local climate heavily influence them. In recent years, CFD technology has led to a breakthrough in studying solar chimneys. There are advantages not only in terms of the accuracy, but also in terms of the time efficiency, and savings in resource utilization [2,5,20,21].

Firstly, CFD simulations provide an in-depth understanding of the critical factors that affect the performance of a solar chimney [22,23]. During the design phase, simulations can calculate parameters, such as air temperature, pressure and humidity, and wind speed. This is not easy to achieve through physical experiments and theoretical calculations. By optimizing these parameters, the performance of the chimney can be further improved. Secondly, CFD simulations provide a low-cost platform for numerous experiments and modifications. Traditional experimental methods are both time and resource intensive, and costly in terms of trial and error. However, using CFD simulations, designers can perform representative trials and optimizations of different design variables, without conducting actual physical experiments.

Furthermore, with increased computer processing power, complex CFD simulations can now be completed in an acceptable amount of time. This makes CFD simulations a fast and effective tool in various educational, research, and industrial applications. CFD simulations of solar chimneys also help to advance the development of green energy technologies [24]. CFD simulations can help us to understand and optimize this process, thereby increasing solar energy utilization, and reducing the construction and operating costs of the associated equipment. However, while CFD simulations offer many advantages, we must also know their limitations. The simulation results’ accuracy depends on the model’s accuracy and reliability. In addition, CFD simulations reduce complex physical processes to mathematical models, and some detailed information may be lost. Therefore, there is a need to rely on the experimental validation and correction of the model [25,26].

As the primary tool for simulating and characterizing radiative heat transport phenomena in computing, radiation models have an important influence on the results and accuracy of simulations. An appropriate radiation model can accurately predict the behavior of a system when thermal radiation is considered. Whether this behavior constitutes the temperature distribution, fluid flow, or chemical reaction rate, all are affected by thermal radiation. With a suitable model, predictions can be guaranteed to be accurate, model bias can be reduced, and the process can be more accurately understood and controlled, to optimize the design based on the simulation results. In addition, the choice of the suitable radiation model is closely related to the efficiency of the calculation.

On the one hand, an overly complex model will increase the difficulty and computational time of the calculation, thus consuming more resources. On the other hand, an overly simplified model may neglect or misjudge some essential physical effects, thus reducing the reliability of the simulation results. Therefore, the correct choice of radiation model can balance accuracy and computational efficiency. Aligholami compared the DO, P1, and Rosseland radiation models, based on the horizontal variation of the static temperature along the collector, and the radiation temperature distribution along the vertical absorption of the chimney. The results showed that the DO radiation model was more accurate, with an error rate of 1% [27,28]. Hu et al. evaluated the radiative heat transfer of the DO, P1, and discrete transfer radiation model (DTRM) in Ansys Fluent, and showed that the DO and DTRM outperformed the other models, and had a more comprehensive range of applications [29].

In summary, there is limited information on the selection and accuracy of radiation models in the CFD simulation of solar chimneys. In this study, we use a wall-mounted solar chimney as the object of study, and apply a combination of four different radiation models and full-scale experiments to investigate the variation in velocity and temperature distribution in the internal flow field of a wall-mounted solar chimney. The most suitable radiation model for predicting the performance of wall-mounted solar chimneys is determined.

2. Radiation Model Theory

The discrete coordinate DO radiation model calculates the radiation heat transfer used to model complex three-dimensional heat fields. The DO radiation model, which converts the radiation transport equation into the transport equation for radiation in the spatial coordinates (x,y,z), discretizes the direction of change in the radiation [13,30,31]. A finite number of discrete solid angles are solved, each of which is solved concerning the direction of the corresponding vector in the global Cartesian coordinates. The numerical solution is obtained through the calculation of the transport equations for the radiation in the angular direction over the entire 4$\pi$ discrete spaces [32,33]. The radiation transport equations for the DO radiation model for the absorption, emission, and scattering media in the directions of the positions $\overrightarrow{r}$ and $\overrightarrow{s}$ are as follows:

$$\nabla \cdot (I(\overrightarrow{r},\overrightarrow{s})\overrightarrow{s})+(a+{\sigma }_s)I(\overrightarrow{r},\overrightarrow{s})=a{n}^2\frac{\sigma T^4}{\pi }+\frac{{\sigma }_s}{4\pi }{\displaystyle \underset{0}{\overset{4\pi }{\int }}I(\overrightarrow{r},\overrightarrow{s^{\prime }})}\Phi (\overrightarrow{s}\cdot \overrightarrow{s^{\prime }})d{\Omega }^{\prime }$$

(1)

The S2S radiation model is based on the basic principles and formulations of thermal radiative transfer, and simplifies the computational model by approximating the source of thermal radiation as a point heat source, and the target surface as a planar surface. The S2S model assumes that the path of light propagation is almost straight; therefore, the linear propagation model can treat the radiative transfer as a point source and planar radiative transfer problem [32,34]. In the S2S model, the radiative flux applied to the point source is distributed uniformly over the target surface, and integrated along the vertical direction on each surface, to obtain the radiative flux of the target. In the S2S model, the target surface is divided into a finite number of surface elements, according to their normal vectors, and parameters such as the viewing angle coefficient, projection coefficient, reflection coefficient, and transmission coefficient are then calculated for each point heat source to each surface element. The radiative flux from the target surface is then calculated using the Stefan–Boltzmann theorem [35,36]:

$$J_k=E_k+{\rho }_k{\displaystyle {\sum }_{j=1}^N{F}_{kj}}{J}_i$$

(2)

The Monte Carlo radiation model (MC) is a computational model based on a random number approach, primarily used to simulate complex radiation transport processes. In the MC radiation model, the radiant energy is treated as a stream of photons, and the transport, absorption, scattering, and emission processes of photons in a medium are simulated utilizing random sampling. MC radiation models can handle complex radiation transport paths, including multiple scattering and reflection, and can be used to analyze the behavior of radiation transport in various materials and media [37,38,39]. The MC model’s theoretical basis is to solve for the probability distribution function in a scene using random sampling. The behavior of photons in a medium can be modeled via the MC approach, by randomly sampling the physical parameters at each point along the radiation transport path.

The basic theory of the Rosseland radiation model is based on the average nature of the absorption coefficients of a medium. The absorption coefficient of a medium is usually related to the relaxation time, frequency response, and other factors within the medium, so a medium’s radiation absorption and scattering properties can be approximated by averaging the absorption coefficients in the medium at high temperatures or high radiation conditions. In the Rosseland radiation model, the frequencies of radiation and absorption are approximated as continuous, and are generally assumed to be blackbody radiation. The absorption properties exhibited by a medium under this radiation can be expressed as the absorption coefficient of the medium to the radiation intensity [18,40]. By counting and calculating the absorption coefficients and radiation intensities in a medium, the Rosseland radiation model can help us to understand the role and interactions of various radiation mechanisms in a medium, and to simulate and analyze the physical characteristics of a medium.

3. Methodology

3.1. Physical Grid Model

The present study focuses on a monolithic building; its dimensions were precisely measured as 3 m × 3 m × 2.8 m. In addition, a particular device, a wall-mounted solar chimney, with the exact dimensions of the building façade, was installed on the south side of the monolithic building, on its external wall. As shown in Figure 2, the unstructured polyhedral meshing method was used when delineating the grid. Moreover, to improve the mesh quality and computational accuracy, the minimum cell length of the body mesh was 2.9 mm, and two buffer layers and one peeling layer were used. The boundary layers were treated with a smooth transition offset type, with a conversion ratio of 0.272 and a growth rate of 1.15, while the aspect ratio of the non-boundary layer mesh was controlled to a maximum of 1:2.1. It is worth noting that the orthogonal grid quality has reached 0.88, which is a reasonably high level, and the grid quality level is fully adequate for the computational requirements.

3.2. Grid-Independent Verification

In the initial stages of using grid computing, the primary purpose of grid-independent verification is to verify whether the calculation results change with the grid size change, and verify the computational model’s stability and reliability. By comparing the results of calculations with different grid sizes, we can determine whether the model has converged, and to what extent. Grid-independent validation is a crucial step in computational fluid dynamics (CFDs) research, helping to ensure the correctness of the model, and to improve its accuracy and reliability. If the model does not pass grid-independence verification, it may have problems, and needs further optimization. When modeling more complex physical processes, such as fluid flow, heat transfer, etc., a high-quality mesh is essential, to reveal the underlying laws more accurately.

Therefore, this study is dedicated to validating the mesh before starting the calculation. The validation of the mesh mainly consists of two aspects: firstly, the quality of the mesh itself is examined, including the checking of its topology and the docking of edges; secondly, the irrelevance of the mesh is verified. By comparing the calculation results under different meshes, and observing their variation patterns, we can judge whether the results are independent of the mesh parameters. If the variation in the grid parameters has a negligible impact on the results, then such a grid can be considered irrelevant. The irrelevant grid has an essential role in improving the calculation results’ accuracy. In this study, the average velocity of the exit stream of a walled solar chimney was chosen as the resultant parameter for validation. We used different grid sizes and shapes, and calculated the velocity of their exit airflow. Through comparing the results under different grids, it is easy to see the effect of the grid size on the dispersion error of the calculation results. The results are also displayed in the form of a graph, to make the results more intuitive and clear at a glance, as a way of verifying the irrelevance of the grid.

From the above grid-independent study, in Figure 3, it can be seen that the error between the fourth and fifth set of grids is 0.96%; to ensure that the most accurate simulation results are obtained with limited computing resources, the fifth set of grids n5 = 4,236,142 is chosen as the most suitable grid set for calculation.

3.3. Boundary Conditions and Turbulence Model

In this study, the wall-mounted solar chimney ventilation performance and flow state simulation analysis is performed using Ansys fluent 2021R1 computational fluid dynamics software. The pressure–velocity coupling is performed using the SIMPLEC algorithm, based on the control body center approach to solve for the gradient, with the pressure selected in PRESTO discrete format, to increase the jet accuracy, and the remaining terms in second-order windward format, to increase the computational accuracy.

The inlet is set as a pressure inlet boundary, with a total inlet pressure of 0 pa, and the incoming flow temperature set at 20 °C. The inlet of the associated wall-mounted solar chimney is a square inlet of 100 × 2600 mm². The outlet is designated as a pressure outlet, with the exact dimensions of the inlet. The outer transparent glass of the wall-mounted solar chimney is made from acrylic panels, with a light transmission rate of 95%. The boundary conditions are the convective mixed radiation boundary, the translucent radiation boundary, the convective heat transfer coefficient determined by Equation (3), internal and external emissivity to 0.49, the sky radiation source temperature determined by Equation (4), the outer wall thickness of 100 mm, Neumann boundary conditions, and the convective heat transfer coefficient determined by Equation (5):

$$h_{glass}=5.62+3.9U_{out}$$

(3)

$$T_{sky}=0.0552T_{out}^{1.5}$$

(4)

$$h_{wall}=5.62+3.9U_{in}$$

(5)

where $T_{out}$ is the ambient temperature, $U_{out}$ is the external airflow velocity, and $U_{in}$ is the internal airflow velocity.

Through the use of the pressure-based 3D double precision solver Ansys Fluent for transient solving, a two-sided turbulence model is sufficient for the relevant operations. After repeated comparisons and calculations, the RNG turbulence model has been selected as the turbulence calculation model, which can adequately describe and explain the flow state inside a walled solar chimney. The RNG model is an improved flow model based on the energy dissipation rate equation, and is designed to predict the details of the turbulent flow more accurately. Particularly in complex flow environments, such as close to walls, it improves on the standard model with a retarding function. The RNG model has been chosen for its ability to reproduce the airflow behavior within the chimney, and to handle complex flows and turbulence, and the fine detail it provides, providing us with more accurate simulation results. This helps us understand the solar chimney’s performance better, and provides a scientific basis for optimizing the design, and improving the chimney’s efficiency. Its transport equations are given in the equation:

$$\frac{\partial }{\partial t}(\rho k)+\frac{\partial }{\partial x_i}(\rho \epsilon u_i)=\frac{\partial }{\partial x_j}({\alpha }_{\epsilon }{\mu }_{eff}\frac{\partial \epsilon }{\partial x_j})+G_k+G_b-\rho \epsilon -Y_M+S_k$$

(6)

$$\frac{\partial }{\partial t}(\rho \epsilon )+\frac{\partial }{\partial x_i}(\rho \epsilon u_i)=\frac{\partial }{\partial x_j}({\alpha }_{\epsilon }{\mu }_{eff}\frac{\partial \epsilon }{\partial x_j})+C_1{}_{\epsilon }\frac{\epsilon }{k}(G_k+C_{3\epsilon }{G}_b)-C_{2\epsilon }\rho \frac{{\epsilon }^2}{k}-R_{\epsilon }+S_{\epsilon }$$

(7)

3.4. Full-Scale Experiments

This experiment was tested in a full-size building, with an external wall-mounted solar chimney, in Urumqi. The relevant dimensions of the wall-mounted solar chimney were 3000 × 2800 × 180 mm³, as shown in the diagram. During the experiment, a thermometer was used to measure the ambient temperature. The surface temperatures of the glass and collector panels were measured using eight-channel thermocouples. The temperature of the air in the chimney channel, and the surface temperature of the building facade were measured using temperature and wind speed recorders, to measure the inlet and outlet wind speed and air temperature. At the same time, the distribution of heat flow through the glazing facade was measured using a temperature and heat flow thermal cycle acquisition logger. The distribution of the relevant measurement points is shown in Figure 4, below.

4. Results and Discussion

4.1. Specific Enthalpy at the Outlet and Turbulence Distribution at the Inlet

It is shown that specific enthalpy, an essential energy transfer indicator during heat exchange in wall solar chimneys, is most prominent under the S2S radiation model, where the peak specific enthalpy at the boundary on both sides of the outlet appears at its maximum value. Based on the measured data, the maximum specific enthalpy at the outlet of the wall solar chimney under four different radiation models is found in Figure 5a, and the peak specific enthalpy of S2S is calculated to be 53.2%, 23.4%, and 54.2% higher than that under the three radiation models DO, Rosseland, and MC. This indicates that the wall solar chimney is more effective in absorbing and converting solar energy in the S2S radiation model, a complex radiation model that considers the effects of solar radiation, atmospheric radiation, and cloud reflection. Compared to the other three radiation models, the S2S model has a more substantial radiation effect, and is more effective in capturing and converting solar energy during the energy conversion process. However, the mean value of the specific enthalpy distribution at the exit of the wall solar chimney under the Rosseland radiation model is much higher than that of the remaining three radiation models, and the results of the Rosseland model appear to be lower, compared to the S2S, DO, and MC radiation models, in the study of the exit enthalpy analysis for the wall solar chimney. This may be due to the fact that the Rosseland model does not adequately account for the effects of solar radiation, atmospheric radiation, and cloud reflections in simulating the energy conversion process in the solar chimney. In addition, the heat exchange process in wall solar chimneys is complex, and involves a number of different thermodynamic processes, which may be beyond the handling capacity of the Rosseland model.

The mean turbulent viscosity at the inlet of the walled solar chimney for the four different radiation models is shown in Figure 5b. The graphs reveal the differences between the radiation models DO, S2S, Rosseland, and MC when it comes to modeling the mean turbulent viscosity of the fluid at the inlet of a wall-mounted solar chimney. These models are calculated using the fluid flow in a turbulent state, caused by solid vortex cluster diffusion, and random pulsations due to cascading hashing. The turbulent viscosity is an important parameter that reflects the velocity gradient and can be used to measure the intensity of mixing and transport in the flow regime. The comparative results show that the average turbulent viscosity of the fluid under the MC radiation model is slightly lower than that under the DO and S2S radiation models. This may be because the MC model more rigorously simulates the radiative heat transfer process, taking into account the random motion of the particles. In particular, the S2S radiation model focuses only on the radiative heat transfer process at the surface of the heat collection shed, and neglects the radiation effect within the air; hence, its predicted mean turbulent fluid viscosity is higher than that of the MC and DO models, at 20.4% and 24.3%, respectively. Compared with the MC and S2S radiation models, the DO radiation model can more comprehensively consider the interaction of the fluid’s three heat transfer modes: radiation, convection, and conduction. Therefore, its predicted turbulent fluid viscosity distribution at the chimney inlet is more reasonable, and closer to the actual situation. The results of this study provide an essential reference for understanding the fluid heat transfer process, and the simulation design optimization, of wall-mounted solar chimneys.

4.2. Comparison of Exit Velocity and Temperature Characteristics

The distribution of the exit airflow velocity of a wall-mounted solar chimney at t = 4000 s is shown in Figure 6. The basic premise of the Rosseland radiation model is that the radiation and absorption frequencies are approximately continuously distributed, a phenomenon commonly considered in physics to be blackbody radiation. However, this premise leads to overly idealized model predictions, meaning that actual wall-mounted solar chimney operational data differ significantly from model predictions. In contrast, the three radiation models DO, S2S, and MC all consider the effects of radiant heat exchange on the internal surfaces of wall-mounted solar chimneys. It is worth noting that the S2S model does not consider the scattered radiation from the surface of the wall-mounted solar chimney. Examining the results for all calculation periods, it can be seen in Figure 7 that the predictions of the DO and MC radiation models for the exit velocity of the wall-mounted solar chimney are in better agreement with the test curves based on the experimental data. At t = 4000 s, the error between the DO and MC models and the experimental data is only 0.004% and 0.01%, which further validates the accuracy of the DO radiation model in predicting the exit velocity of the wall-mounted solar chimney, over the other models.

From Figure 8, it can be seen that the outlet temperature distribution and outlet velocity distribution of the wall-mounted solar chimney show a synchronous trend under the different radiation models. Among the four different radiation models, the simulation calculation results of the MC and DO radiation models are closer to the actual experimental results, which proves the validity and accuracy of these two models. This is crucial to further understanding and improving the design and operational performance of solar chimneys. To analyze and compare the performance of the different models more comprehensively, we chose the data point of t = 4000 s for comparison. The specifics of the associated errors are shown in Figure 9. The comparative analysis provides a clearer understanding of the strengths and weaknesses of the different radiation models. Overall, the graphs and data comparisons show that the MC and DO radiation models can accurately predict the exit temperatures and velocities of wall-mounted solar chimneys. This provides an essential theoretical basis and technical support for the research and application of solar chimneys.

4.3. Internal Flow Field and Temperature Distribution

The velocity distribution inside the wall-mounted solar chimney is shown in Figure 10. Due to the small cross-sectional area of the chimney inlet and outlet, the pressure increases under the action of the nozzle, making the airflow velocity larger. The simulation and experimental results are shown graphically, with the four points selected as data samples equally spaced at 4000 s in the airflow’s central axis inside the wall-mounted solar chimney, and in the direction of the glass cover surface. In Figure 11, the temperature at the probe points simulated by the Rosseland radiation model is compared to the experimental results, with an error of 4.11% in the mid-axis direction, and 3.21% in the glass cover direction. This is because the Rosseland radiation model assumes an approximately continuous distribution of radiation and absorption frequencies, which is not the case in practice and, therefore, the model is not optimal for use in wall-mounted solar chimney simulations. The errors between the DO and MC radiation models and the experimental results are much smaller than those of the Rosseland radiation model, showing a high degree of agreement between the two, with 0.47% in these two directions. Although the S2S radiation model also considers the surface radiative heat transfer inside the wall-mounted solar chimney, its simulation results disagree with the experimental results, with errors of 1.81% and 1.61% in the central axis and glass cover directions, respectively. Therefore, the DO and MC radiation models appear more reliable for simulating the actual process of wall-mounted solar chimneys, while the Rosseland and S2S radiation models are less adaptable.

In-depth data analysis shows that the translucent glass cover plays a crucial role in the operation of the wall-mounted solar chimney heat collection system. The cover effectively absorbs solar radiation, and converts it into heat, then heats the air inside the solar chimney. When this hot air is convectively exchanged with the building facade, the conversion and utilization of solar energy are achieved. Such a process is a vital function of a collector system.

The three different radiation models, MC, DO, and S2S, all consider the radiative heat exchange process on the surface of a wall-mounted solar chimney. However, as the S2S radiation model does not consider the absorption, emission, and scattering of radiation from the chimney surface, the results lead to some errors regarding the fluid temperature, the average exit fluid velocity, and the average temperature at the glass cover in the test results. In contrast, the MC and DO radiation models show better accuracy. The errors between these two models and the experimental results are within acceptable limits for the core parameters of the fluid temperature, the flow velocity, and the average temperature at the fluid outlet and associated components. In particular, the DO radiation model showed excellent accuracy and adaptability in the simulations. Therefore, selecting a suitable radiation model is crucial to improve the calculation of the heat collection and ventilation, and to optimize the performance of wall-mounted solar chimneys, with the DO radiation model standing out from the crowd as the ideal choice for numerical simulations of wall-mounted solar chimneys.

5. Conclusions

The present study reproduced experiments on the heat collection and ventilation flow of a walled solar chimney in a full-size building using four different radiation models. Four different radiation models, DO, S2S, MC, and Rosseland, were simulated numerically in three dimensions for the walled solar chimney, and the numerical simulation results were compared with the experimental data for analysis. The relevant conclusions are shown below.

• The peak specific enthalpy of the S2S radiation model at the boundary on both sides of the outlet is 53.2%, 23.4%, and 54.2% higher than that of the DO, Rosseland, and MC radiation models.
• The S2S radiation model focuses only on the radiative heat transfer process at the surface of the walled solar chimney, ignoring the radiation effect within the air and, therefore, its predicted maximum turbulent fluid viscosity is higher than that of the MC and DO models by 16.87% and 8.44%, respectively.
• The error between the DO radiation model and the experimental results in the mid-axis and glass cover directions is only 0.33% and 0.15%, respectively; the error of the MC radiation model in these two directions is 0.51% and 0.47%, respectively. Although the S2S radiation model also considers the surface radiative heat transfer inside the walled solar chimney, its simulation results disagree with the experimental results, with errors of 1.81% and 1.61% in the central axis and glass cover directions, respectively.

It is worth mentioning that the DO radiation model outperforms the other three radiation models in the simulation of radiative heat transfer in a walled solar chimney combined with solar ray tracing, in terms of the numerical prediction of various index parameters, the temperature, and the velocity, as well as the specific enthalpy and turbulence viscosity of the walled solar chimney. The experimental results and error analysis also verify this. Therefore, in the radiation model application for numerical simulation predictions related to walled solar chimneys, it is more appropriate to select the DO radiation model. However, in the process of model construction, the selection of some crucial parameters is also very critical, such as the radiation coefficient, absolute emissivity, etc., as this directly affects the model’s prediction accuracy. However, the suitable parameter values depend on the actual application environment. Therefore, it is essential to investigate further the influence of relevant parameters on the future prediction accuracy of wall solar chimneys based on the DO model.