- Researchers from Spain’s Technical University of Madrid (IES-UPM) have proposed a novel methodology to determine the structure shading factor (SSF) and mismatch loss (MML) for the rear side of bifacial PV modules using single-axis trackers.
In their experimental analysis, the team has considered both one-in-portrait (1P) setups, where the torque tube runs directly behind the center of the panel, as well as two-in-portrait (2P) setups, where the torque tube sits between the two panels.
“The calculation of SSF and MML is important for two reasons. First, they are input parameters required by PV simulation software for energy yield estimations. It is necessary to have accurate estimates that help reduce the uncertainty associated with PV performance simulations,” corresponding author, Pablo Merodio, told pv magazine. “This uncertainty control is crucial for industrial and financial participants, whose main concern is the limitation of financial risks associated with bifacial PV systems. Second, they are typically requested from tracker manufacturers by EPCs and other stakeholders of the PV industry.”
To experimentally determine SSF and MML, the group has designed two instruments with high spatial resolution rear irradiance maps at the module level. For a 1P setup, they used a wooden board with dimensions 2 m × 1 m, on which 21 solar cells were strategically placed. The instrument for the 2P measurement was similar in size but included 35 cells uniformly distributed in five columns and seven rows. “It creates high spatial resolution rear irradiance maps measured at the module level,” said Merodio.
The proposed methodology was tested at a PV plant in Cheste, Spain, which includes 1P and 2P trackers equipped with 355 Wp bifacial Longi Solar. The ground is gravel-covered, with albedo being approximately 30%. The torque tubes have a square section of 14 cm × 14 cm. The tracker heights were 1.35 m for the 1P and 2.1 m for the 2P, and the array widths were 2 m for the 1P and 4.16 m for the 2P, which gives a ratio height to width of 0.7 for the 1P and 0.5 for the 2P.
Based on the readings from the instruments, the group developed calculations for SSF and MML. To calculate SSF, they suggested comparing how much sunlight the back of the panel receives with and without shading from the structure. To calculate MML, on the other hand, they used a formula that connects uneven sunlight to power loss. “Care should be taken when drawing conclusions from these comparisons, as the SSF and MML parameters strongly depend on the specific characteristics of the PV system they describe,” they highlighted.
The testing demonstrated that, for 1P trackers, SSF is 10.0% and MML is 0.46% for full-cell modules and 0.40% for half-cell modules. For 2P trackers, SSF is 2.7% and MML is 0.35% and 0.21% for full-cell half-cell modules, respectively. “We also provide a linear fit of SSF as a function of the distance (d) from the torque tube to the PV module: SSF= -0.2d+11.6. Equivalent linear relationships are found for MML, considering full-cell, MML=-0.02d+0.62, and half-cell modules, MML=0.02d+0.56,” the academics said.
“The uncertainty in SSF associated with our measurement procedure is estimated to be 4%(2%) for the 1P(2P) tracker, while the uncertainty in MML is expected to be at most 4% for both trackers,” they concluded. “When our results are used as constant annual inputs for energy yield estimations, these uncertainties must be adjusted to 10% in SSF for both trackers and 20% (30%) in MML for the 1P(2P) tracker, but their impact on the uncertainty in the energy yield is significantly reduced to less than 1%. The varying Sun positions throughout the year do not affect single-day measured SSF and MML values.”
Their findings were presented in “Experimental determination of the structure shading factor and mismatch losses for bifacial photovoltaic modules on variable-geometry, single-axis trackers,” published in Solar Energy. According to Dr. Merodio, the team is now working on extending their measurement campaign to cloudy days and studying “second-order effects, such as the shading caused by the pillars and the tracking system mechanisms, as well as edge effects on modules at the end of a tracker.”
Author: Lior Kahana
This article was originally published in pv magazine and is republished with permission.
