A Photovoltaic Thermal System, or PVT, is a smart technology. It combines two things: solar panels (PV) and solar thermal collectors. Normal solar panels only create electricity. Solar thermal collectors only create hot water. A PVT system does both at the same time. It captures sunlight to make electricity, and it uses the waste heat to warm up water. This makes the system very efficient because it uses more of the sun’s energy.

In this Photovoltaic Thermal System CFD simulation, we perform a validation study. We recreate the work from a research paper by Maadi et al. (2019). Our goal is to simulate a PVT module using the Discrete Ordinates (DO) Radiation Model in ANSYS Fluent. This ANSYS Fluent PVT tutorial helps you understand how to model complex radiation and heat transfer in solar energy systems. For more heat transfer examples, please visit our Heat Transfer CFD tutorials.

• Reference [1]: Maadi, Seyed Reza, et al. “Coupled thermal-optical numerical modeling of PV/T module–Combining CFD approach and two-band radiation DO model.” Energy conversion and management198 (2019): 111781.

Figure 1- Schematic diagram of the PVT system showing the layers.[1]

Simulation Process: PVT System Fluent setup

For this PVT CFD analysis, we started by designing the system geometry. We used ANSYS Design Modeler to create the different layers of the solar module. As shown in Figure 1, the system has many parts, so we needed to be very careful. We divided the geometry into several blocks. This allowed us to create a high-quality structured grid in ANSYS Meshing. A structured grid is very important for radiation problems. We generated a total of 1,738,000 hexagonal cells to ensure the results were accurate.

Next, we set up the physics in the ANSYS Fluent solver. The most critical part of this simulation is the radiation model. We selected the Discrete Ordinates (DO) Radiation Model. We assumed that the walls and the pc-Si (silicon) layer are “non-gray diffuse surfaces.” This means their optical properties change depending on the wavelength of the light. We assumed the sunlight hits the module perpendicularly. We also had to define the properties of the water flowing through the tubes. The paper showed that water viscosity changes with temperature in a complex way. Therefore, we wrote a User Defined Function (UDF) to define the viscosity accurately. Finally, we set the boundary conditions and ran the calculation until it converged.

Figure 2- Structured mesh grid generated for the PVT computational domain.

Post-Processing: Photovoltaic Thermal System Temperature and Efficiency Validation

In this section, we analyze the results of the Photovoltaic Thermal System CFD simulation. We focus on the temperature distribution and the electrical efficiency to validate our work against the reference paper. First, we examine the Temperature Distribution. The thermal contours show a beautiful interaction between the sunlight and the water. The solar radiation passes through the glass and hits the silicon layer. This layer converts some light into electricity and the rest into heat. The heat then travels to the water tubes. The simulation shows that the water enters at 311 K and leaves at 316.67 K. This is a temperature rise of 5.67 K. This proves that the system is effectively capturing thermal energy. The temperature distribution on the surface is not uniform; it is hotter between the tubes and cooler exactly where the water flows. This “unusual pattern” is captured perfectly by the DO radiation model.

Figure 3- Temperature contours over the absorber plate and tubes.

Second, we validate the Electrical Efficiency. This is the most important test. We compared our PV Cell Temperature (Tpv) and efficiency (ηelec) with the paper’s data.

• Reference Paper: The paper reported a PV cell temperature of 319.31 K and an electrical efficiency of 10.42%.
• CFD Simulation: Our simulation predicted a PV cell temperature of 317.74 K (just 1.57 K lower) and an efficiency of 10.1%.

The difference in efficiency is extremely small. The error is only 3.07%. This excellent agreement confirms two things. First, our mesh resolution of 1.7 million cells was sufficient to capture the physics. Second, our setup for the non-gray DO radiation model and the viscosity UDF was correct. The simulation successfully replicated the coupled thermal-optical processes described in the study.

Key Takeaways & FAQ

• Q: What is a Photovoltaic Thermal (PVT) System?
  • A: A PVT system is a hybrid device that generates both electricity (Photovoltaic) and heat (Thermal) from the sun simultaneously. It is more efficient than using separate panels.
• Q: What is the DO Radiation Model in ANSYS Fluent?
  • A: DO stands for Discrete Ordinates. It is a comprehensive radiation model that solves the radiative transfer equation for a finite number of discrete solid angles. It is suitable for all optical thicknesses.
• Q: Why do we use “Non-Gray” radiation models?
  • A: A “Gray” model assumes radiation properties are the same for all wavelengths. A “Non-Gray” model is more accurate because it considers that materials behave differently under visible light versus infrared heat.
• Q: Why is a UDF needed for this simulation?
  • A: Standard software libraries might not have the specific complex formulas for material properties (like temperature-dependent viscosity) used in a research paper. A User Defined Function (UDF) allows us to add these custom formulas.