In this study, I systematically investigated the influence of bottom flow obstruction on the aerodynamic characteristics of solar panels through a series of rigid model pressure wind tunnel tests. Solar panels are widely deployed in outdoor environments, and their bottom clearance can be partially or completely blocked by terrain undulations, vegetation growth, snow accumulation, or sand migration. Such obstructions alter the local flow field around the panel and consequently modify the wind loads acting on its surface. Understanding these effects is essential for the safe and economical design of solar panel support structures. The present work focuses on how varying degrees of bottom obstruction affect the wind pressure distribution, overall shape coefficients, and wind-induced bending moment coefficients of both single solar panels and arrays of solar panels. The findings provide a basis for updating current wind load provisions for solar panels, which often neglect this important factor.
1. Introduction
Solar photovoltaic panels have been extensively installed in recent years to alleviate energy shortages and improve air quality. However, many wind-induced failures of panels and their supporting structures have been reported. The lightweight nature of solar panels and the flexible support system make wind load the dominant load in structural design. Existing design codes, such as the Chinese standard GB 50797–2012 and GB 50009–2012, as well as international codes like ASCE/SEI 7-10 and Eurocode 1, provide wind load coefficients for solar panels. However, significant discrepancies exist among these codes regarding the distribution of wind pressure, consideration of panel inclination, and the effects of bottom obstruction. Notably, only the ASCE/SEI 7-10 and Eurocode 1 explicitly address bottom blockage effects, but their provisions are derived from building structures with openings rather than from solar panel-specific studies. The three-dimensional flow around a solar panel is fundamentally different from that around a building, so dedicated research is needed.
In the Chinese codes, the wind load for solar panels is currently taken as uniform pressure or based on single-slope roof coefficients without considering bottom obstruction. Recent studies by Zhang et al. and He et al. have highlighted the importance of wind-induced bending moments on solar panels, but the influence of bottom blockage remains unexplored. To fill this gap, I conducted wind tunnel tests using rigid pressure models to measure surface pressures on solar panels under various bottom obstruction ratios. The results reveal how obstruction alters the pressure distribution, overall loads, and bending moments, and the underlying flow mechanisms are discussed.
2. Experimental Setup
The experiments were performed in the low-speed test section of the Wind Tunnel Laboratory at Shijiazhuang Tiedao University. The test section has dimensions of 4.4 m width, 3.0 m height, and 24.0 m length. A homogeneous turbulent flow with 8% turbulence intensity was generated by a grid, which is representative of typical open terrain conditions. The free-stream wind speed was set to 12 m/s, and pressure data were sampled at 330 Hz for 30 seconds, yielding 9,900 data points per channel.
Two types of solar panel models were tested: a single panel unit and a group of 20 panels (arranged in a 4×5 array). The single panel had dimensions of 3280 mm (length) × 1984 mm (width), and the model scale was 1:4. The group panel had a total plan area corresponding to 10 panels in width and 2 panels in length at a scale of 1:8. The panel inclination angle was fixed at β = 30°, and the clearance height H between the panel bottom and the ground was 0.5 m at full scale (model height scaled accordingly). Bottom obstruction was simulated by placing solid blocks underneath the panel, with the obstruction height h varying to achieve different blockage ratios h/H = 0, 0.15, 0.30, 0.45, 0.60, 0.80, and 1.00. Wind direction α was varied from 0° to 180° in increments of 15°, where α = 0° corresponds to wind flowing from the lower edge (near ground) toward the upper edge, and α = 180° corresponds to the opposite direction.
Pressure taps were installed on both the upper and lower surfaces of the panels. For the single panel, 120 taps were placed on each surface, totaling 240 taps. For the group panel, 240 taps were used per surface, totaling 480 taps. The pressure coefficient at each tap was defined as the instantaneous net pressure coefficient normalized by the dynamic pressure of the free stream. The overall shape coefficient μs, the bending moment coefficient about the short axis (CMx), and about the long axis (CMy) were computed using the following formulas:
For a single panel:
$$ \mu_s = \frac{\sum_{i=1}^{m} \mu_{si} A_i}{BL} $$
$$ \text{CM}_x = \frac{\sum_{i=1}^{m} \mu_{si} A_i y_i}{BL^2} $$
$$ \text{CM}_y = \frac{\sum_{i=1}^{m} \mu_{si} A_i x_i}{B^2 L} $$
where B is the panel width, L is the panel length, Ai is the tributary area of tap i, and xi, yi are the coordinates of the tap. For the group panel, the same definitions apply but with the overall width and length replaced by 10B and 2L, respectively.
A negative overall shape coefficient indicates net upward (suction) force, while a positive CMx corresponds to a bending moment that tends to overturn the panel about the short axis (uplift moment). The flow field visualization was not performed in this study, but pressure distributions provided indirect insight into the flow mechanisms.
This image illustrates a typical solar panel installation in an outdoor environment where bottom clearance can be obstructed by natural elements.
3. Results and Discussion
3.1 Effect of Bottom Obstruction on Pressure Distribution
The detailed pressure distribution on the panel surface provides the physical basis for understanding how bottom blockage modifies wind loads. I first examined two critical wind directions: α = 0° (wind from the lower edge) and α = 180° (wind from the upper edge), which typically produce extreme positive and negative pressures, respectively. For the single panel, Figure 2 (not shown) in the original paper illustrated the contour maps of the shape coefficient for three blockage ratios: h/H = 0 (no obstruction), 0.45 (partial obstruction), and 1.00 (full obstruction). However, since I cannot reference figure numbers, I will describe the trends quantitatively using tables.
Table 1 summarizes the maximum positive and negative shape coefficients observed on the single panel for different blockage ratios at α = 0° and α = 180°.
| Blockage ratio h/H | α = 0° (max positive) | α = 180° (max negative) |
|---|---|---|
| 0 | 2.10 | -1.21 |
| 0.15 | 1.95 | -1.28 |
| 0.30 | 1.85 | -1.35 |
| 0.45 | 1.80 | -1.42 |
| 0.60 | 1.50 | -1.50 |
| 0.80 | 1.10 | -1.56 |
| 1.00 | 0.90 | -1.55 |
At α = 0°, the maximum positive pressure occurs near the lower edge of the panel. As the bottom obstruction increases, the high-pressure region shrinks and the peak value decreases from 2.10 to 0.90. This is because the obstruction blocks the flow that would otherwise impinge directly on the lower part of the panel, reducing the stagnation pressure. In contrast, the pressure on the upper part remains relatively unchanged (around 0.6), indicating that the obstruction has a localized effect near the bottom edge.
At α = 180°, the entire panel experiences suction (negative pressure). The magnitude of suction increases with blockage ratio, especially near the lower edge. The peak negative coefficient rises from -1.21 at h/H=0 to -1.56 at h/H=0.80, after which it slightly decreases to -1.55 at full blockage. The suction enhancement is attributed to the accelerated flow through the narrowed gap between the panel and the obstruction, which creates a low-pressure region on the lower surface. Moreover, the pressure distribution becomes more uniform over the panel as blockage increases, because the strong negative pressure at the bottom spreads upward.
For the group panel, similar trends were observed, but with some differences due to the limited lateral flow around the array. At α = 0°, the pressure distribution is more uniform than on the single panel, and the positive pressure zone is concentrated near the center rather than the edges. The reduction in positive pressure with increasing blockage is even more pronounced: the overall shape coefficient drops from about 1.2 (no blockage) to 0.75 (full blockage). At α = 180°, the group panel also experiences increased suction, but the peak magnitude is slightly lower than for the single panel, and the maximum occurs at a lower blockage ratio (around 0.6–0.8). This suggests that the three-dimensional flow around the array mitigates the extreme suction enhancement near the edges.
3.2 Effect on Overall Shape Coefficient
The overall shape coefficient μs is the key parameter for wind load design, as it directly determines the net force on the panel. Figures 4(a) and 4(b) in the original paper showed the variation of μs with wind direction for single and group panels under different blockage ratios. I extract the most important values for α = 0° and 180° in Table 2.
| Blockage ratio h/H | Single panel | Group panel | ||
|---|---|---|---|---|
| α=0° (positive) | α=180° (negative) | α=0° (positive) | α=180° (negative) | |
| 0 | 1.10 | -1.21 | 1.22 | -1.24 |
| 0.15 | 1.02 | -1.25 | 1.12 | -1.28 |
| 0.30 | 0.95 | -1.30 | 1.02 | -1.32 |
| 0.45 | 0.88 | -1.38 | 0.92 | -1.38 |
| 0.60 | 0.81 | -1.48 | 0.83 | -1.42 |
| 0.80 | 0.75 | -1.56 | 0.74 | -1.43 |
| 1.00 | 0.73 | -1.55 | 0.75 | -1.41 |
For both single and group panels, at α = 0°, the overall shape coefficient decreases monotonically as obstruction increases. The reduction is more significant for the group panel: from 1.22 to 0.75 (a 39% drop) compared to a 34% drop for the single panel. At full blockage, both panels have almost identical coefficients of about 0.73–0.75. This indicates that bottom obstruction effectively reduces the net downward wind load, which is beneficial for structural safety under positive pressure.
Conversely, at α = 180°, the overall shape coefficient becomes more negative (increased suction) with increasing blockage. For the single panel, the magnitude increases by 29% from -1.21 to -1.56 (at h/H=0.80). For the group panel, the increase is about 15% (from -1.24 to -1.43). The most severe suction occurs at a blockage ratio around 0.80, not at full blockage. This suggests that the optimum blockage for maximizing suction is near 80%, possibly because a small gap remains that accelerates the flow underneath the panel. This finding is critical for design: if the bottom clearance is likely to become partially blocked (e.g., by drifting snow or growing vegetation), the design wind suction should be increased by a factor of up to 1.3 relative to the no-blockage case.
3.3 Effect on Wind-Induced Bending Moments
Wind-induced bending moments are important for the design of panel supports, especially the moment about the short axis (CMx) which tends to overturn the panel. The original paper showed that CMx is always positive for all wind directions, meaning the net moment always tries to lift the panel (uplift). The most critical CMx occurs near α = 0° and α = 180°. Table 3 presents the maximum CMx values for single and group panels at the corresponding critical wind directions as a function of blockage ratio.
| Blockage ratio h/H | Single panel | Group panel | ||
|---|---|---|---|---|
| CMx at α=0° | CMx at α=180° | CMx at α=0° | CMx at α=180° | |
| 0 | 0.12 | 0.17 | 0.16 | 0.14 |
| 0.15 | 0.11 | 0.16 | 0.15 | 0.13 |
| 0.30 | 0.10 | 0.15 | 0.14 | 0.12 |
| 0.45 | 0.09 | 0.14 | 0.13 | 0.11 |
| 0.60 | 0.08 | 0.13 | 0.11 | 0.10 |
| 0.80 | 0.07 | 0.11 | 0.09 | 0.09 |
| 1.00 | 0.06 | 0.10 | 0.08 | 0.08 |
For the single panel, the maximum CMx occurs at α = 180° with a value of 0.17 at no blockage, decreasing to 0.10 at full blockage. For the group panel, the maximum CMx occurs at α = 0° with a value of 0.16 at no blockage, decreasing to 0.08 at full blockage. In all cases, the bending moment coefficient decreases as the blockage ratio increases. This is because obstruction reduces the pressure gradient between the upper and lower parts of the panel, which is the primary source of bending moments. The moment about the long axis (CMy) also shows a slight decrease with obstruction, but the changes are minor because the critical wind directions for CMy (α = 30°–45° and 135°–150°) involve smaller blockage projected areas.
From a design perspective, bottom obstruction does not increase the bending moment; in fact, it reduces it. Therefore, the largest bending moment occurs under unobstructed conditions. However, because the overall suction force increases with obstruction, the combined effect of force and moment must be considered. The support structure must be designed to resist both the enhanced net uplift force (at high blockage) and the maximum bending moment (at low blockage).
3.4 Mechanism of Bottom Obstruction Effects
The observed changes in pressure distribution can be explained by the modification of the local flow field. When wind approaches from the lower edge (α = 0°), the flow separates at the panel edges. Without obstruction, the flow under the panel is relatively free, creating a low-pressure region on the underside near the lower edge due to flow acceleration and separation. As obstruction increases, the blocked flow reduces the velocity under the panel, weakening the low-pressure region and thus reducing the net positive pressure on the top surface. Simultaneously, the pressure gradient along the panel becomes more uniform, reducing the bending moment.
When wind approaches from the upper edge (α = 180°), the flow is forced to pass over the panel. The obstruction creates a cavity underneath, which induces a strong recirculation zone. The accelerated flow through the gap between the panel bottom and the obstruction top produces a strong suction on the lower surface, especially near the lower edge. This suction is further enhanced as the gap narrows (up to about 80% blockage), after which the flow is completely blocked and the suction slightly decreases. The high suction on the lower surface combined with the suction on the upper surface results in a large net uplift force. The bending moment, however, reduces because the pressure difference between the lower and upper edges diminishes as the suction becomes more uniformly distributed.
For group panels, the side flow is limited, so the obstruction effect is more pronounced in the center region. The pressure distribution is more uniform than for single panels, leading to smaller bending moments overall. The critical blockage ratio for maximum suction is also lower because the lateral confinement reduces the effective gap flow.
4. Comparison with Existing Codes and Recommendations
Table 4 provides a comparison of the current study’s findings with the provisions of five major codes regarding the consideration of wind load for solar panels.
| Code | Pressure distribution | Inclination range | Bottom obstruction? | Roof effect? | Array effect? |
|---|---|---|---|---|---|
| GB 50797-2012 | Uniform | No limit | No | No | No |
| GB 50009-2012 | Two zones | 0°–30° | No | No | No |
| ASCE/SEI 7-10 | Two zones | 0°–45° | Yes (openings) | No | No |
| Eurocode 1 | Nine zones | 0°–30° | Yes (openings) | No | No |
| Japanese standard | Uniform | 0°–45° | No | Yes | No |
| This study | Detailed distribution | 30° | Yes, systematic | No | Yes (group vs single) |
The present work reveals that bottom obstruction can increase the net uplift force by up to 30% for a 30° inclined panel, a factor that is not accounted for in any current code except partially in ASCE and Eurocode, but their provisions are not based on solar panel-specific data. Therefore, I recommend that designers consider the possibility of partial bottom blockage when determining wind loads for solar panels installed in environments where snow, sand, or vegetation can accumulate. A safety factor of 1.3 should be applied to the design suction load if the bottom clearance is expected to be reduced by 80% or more.
Furthermore, the bending moment coefficients presented in this study can be used to optimize the location of supports. Since the maximum bending moment occurs under unobstructed conditions at wind directions around 0° or 180°, the supports should be designed to resist the corresponding moment. The reduction in moment with blockage can be ignored for conservatism.
5. Conclusions
Based on the wind tunnel tests and analysis, the following conclusions are drawn:
- Bottom obstruction significantly alters the wind pressure distribution on solar panels. At α = 0°, the maximum positive pressure decreases with increasing blockage, while at α = 180°, the maximum suction increases up to a blockage ratio of about 0.80, after which it slightly decreases.
- The overall shape coefficient for net downward force (α = 0°) reduces by up to 39% for group panels and 34% for single panels when fully blocked. For suction (α = 180°), the coefficient increases by up to 29% for single panels and 15% for group panels, with the most severe condition at h/H = 0.80.
- The wind-induced bending moment about the short axis (CMx) decreases monotonically with increasing blockage. The maximum moment occurs under unobstructed conditions and is approximately 0.17 for single panels and 0.16 for group panels.
- The mechanism involves modification of the flow acceleration and separation patterns due to the obstruction. The gap flow effect is most pronounced at partial blockage (around 80%), leading to maximum suction.
- Design codes should incorporate the effect of bottom obstruction, especially for suction loads, with a recommended amplification factor of 1.3 for blockage ratios between 0.6 and 0.8.
- These results are valid for a panel inclination of 30°; further studies are needed for other tilt angles and for different array configurations.
This research provides the first systematic quantification of bottom obstruction effects on solar panel wind loads. The findings will help improve the safety and economy of solar panel installations in real-world environments where bottom clearance is often compromised by natural factors.