Thermal PV degradation

[BenWinchester]

Issue

PV-T collectors, introduced as part of #76 to the dev branch, are hybrid collectors capable of producing both electricity and heat, in the form of hot water, from a single device. They are a novel technology, and provide several benefits over stand-alone solar-thermal and PV systems, namely

• A reduction in the temperature of the PV cells, leading to
  • Higher-efficiency devices,
  • Reduced thermal degradation,
  • Reduced likelihood of damage due to "hot spots" occurring on the collectors;
• A reduction in space as both of these outputs can be generated from a single collector;
• Potentially higher combined efficiencies than either of the stand-alone devices.

Whilst the above benefits can be listed in a qualitative manner, it is best to include as many of them into the quantitative calculations of CLOVER as possible. The area of the collectors can be easily introduced down the line, and the combined efficiencies are already being calculated. However, the current functionality misses out on one of the key major benefits, that the PV cells' lower temperatures results in both a higher efficiency and reduced likelihood of degradation.

Proposal

1. Thermally-driven degradation of the PV cells should be included as part of CLOVER's run_simulation functionality. This would enable a direct comparison of the degradation in performance of the PV cells as part of the integrated PV-T collectors as opposed to the stand-alone PV cells and would enable direct comparisons within CLOVER and its outputs.
2. The renewables.ninja web interface should be checked to see whether thermal-based efficiency calculations are carried out and, if not, these should be included into CLOVER.

The only alternatives considered to a robust calculation are a non-robust calculation, i.e., making an approximation of this degradation.

[BenWinchester]

Thermal PV-T Degradation

NOTE: Once introduced to the PV panels, it will logically make sense to implement this in the PV layers of the PV-T panels as a simultaneous release. In this way, they can be accurately compared. However, this will likely result in potential hotspots forming/high levels of thermal degradation in closed-loop scenarios, especially closed-loop series solar-thermal and PV-T collectors. There is hence potential to optimise across scenario configurations.

PV-T Panels

What are PV-T panels?

PV-T panels are hybrid photovoltaic-thermal panels and, as per their name, they produce both electricity (through photovoltaic cells) and heat (via thermal collection) from the same collector. There is no one single design of a PV-T collector, but there are designs which are more common within the industry. Flat-plate sheet-and-tube collectors were introduced as part of #76 to the dev branch and are awaiting release as part of 5.1 or 5.2.

Why are they good?

By collecting both heat and electricity from the same area their benefits are twofold:

• They reduce the area required for systems, enabling both electricity and heat to be generated from a smaller area and hence allowing better use of limited space such as islands (🏝️ Area Limitations #113 ) or urban rooftops;
• They provide synergistic efficiency and degradation benefits as the temperature of the PV cells can be reduced, both improving their efficiency and reducing the risk of thermal damage.

Series connections

What do we mean by series connections?

Once solar-thermal collectors are introduced into CLOVER #94 , there will be the option to introduce series connections with PV-T collectors #92 . The "series" nature of these connections means that a heat-transfer fluid (HTF) flows first through the PV-T collectors and then through the solar-thermal collectors.

Why do we want series connections?

By heating the HTF through a larger area of panel, we can potentially achieve higher temperatures whilst also reducing the "half-way temperature" of the HTF when it leaves the PV-T collectors. This could result in

• A greater amount of heat being available to the system.
• Whilst keeping the PV cells at an even lower temperature.

Hotspots

Depending on how the panels are connected, hot spots may form.

Closed loop

"Closed loop" connections have the HTF passing through a heat exchanger, transferring its heat to a hot-water tank, and then passing back through the collectors.

Direct heating

"Direct heating" connections have the HTF being water which is passed through the collectors from a cold-water source to a hot-water source, with no heat exchangers present, and the HTF being the required water.

Where is there a potential for hotspots?

In closed-loop scenarios, the nature of the series connections means that higher temperatures may be achieved. This may result in the HTF passing back into the PV-T collectors at a higher temperature than in PV-T-only configurations, potentially causing more damage. There is hence scope to, once this thermal degradation has been included, allow for the comparison between different scenarios, either within CLOVER or as a post-analysis script.

[BenWinchester]

General thoughts

The way that CLOVER currently considers the performance of PV collectors is to

1. ask renewables.ninja what the performance of a 1 kWp mono-crystaline Silicon (m-Si) PV panel would be at the given location with the given tilt angles etc.

The model that renewables.ninja uses is an empircal model based on the performance of m-Si in the real world [1,2]. Whilst this is good for some siturations, is limits CLOVER's ability to consider

• panels made from materials other than m-Si,
• how effects not considered by Pfenniger and Staffell [1] and Staffell and Pfenniger [2] affect the performance of solar panels,
• consider thermal degradation of PV panels (mentioned first in Thermal PV degradation #110 ).

Including other panel arrays, as discussed in #198 and re-raised in #203 from #166 , would require, potentially, semi-transparrent materials, such as those other than m-Si, to be considered. Including the thermal degradation of these panels would hence be a "nice-to-have" achievement of the work to add different panel topologies and materials into CLOVER.

@hamishbeath , it would be good to have your thoughts on this. Potentially, we could bundle a large number of these discussions together under a single issue to consider multiple PV technologies, orientations, topologies and performances within CLOVER?

References

1. Pfenninger, Stefan and Staffell, Iain (2016). Long-term patterns of European PV output using 30 years of validated hourly reanalysis and satellite data. Energy 114, pp. 1251-1265. doi: 10.1016/j.energy.2016.08.060

2. Staffell, Iain and Pfenninger, Stefan (2016). Using Bias-Corrected Reanalysis to Simulate Current and Future Wind Power Output. Energy 114, pp. 1224-1239. doi: 10.1016/j.energy.2016.08.068