Degradation – the danger within

The return on investment from a PV plant depends largely on the annual yield output. As such, the effects of degradation cannot be ignored. Over the past years module manufacturers have tried to ramp up production and increase throughput. Though most of these advancements have come through better process and optimization of resources, questions remain as to how this process has affected quality. During the past years, module quality has been largely an issue for investors but more recently yield losses faced by investors due to degradation is an emerging concern.
In recent studies there have been a number of causes identified as the source of degradation. Light Induced Degradation, Potential Induced Degradation, degradation due to lamination and EVA quality issues have all been observed. The question as how to measure and quantify degradation and its effects is therefore becoming a pertinent theme.

Measuring degradation
Common practices include laboratory-based measurements under various conditions, on field performance evaluations by means of string I-V measurements. Though these are reliable tests to measure degradation, it is important to understand that these tests are based on sampling principles and require flash list data to compare and quantify degradation. Another alternative would be long-term performance evaluation based on real time monitoring data. These monitoring data based evaluations do provide a measure of on field degradation losses for the entire plant. As the annual yield is the primary concern of an investor, monitoring data based analysis provides a more substantial indication of the degradation loss and also provides for comparative analysis based on long-term data.
As the ill effects of degradation, reduction in module output power and subsequent yield losses, are quite well understood, Solarpraxis Engineering analyzed in real time what exactly contributes to the reduction in module output power. Is it the loss in current or the loss in voltage that plays a vital role in reduction of module output?

Four plants
The monitoring data was collected and analyzed by Solarpraxis Engineers from four different PV plants in the Brandenburg area. The data was analyzed for a period of two years starting January 2011 to December 2012. Of these four power plants three were thin film parks (plant a) utilizing micro amorphous silicon (?-Si and µC-Si Tandem), (plant b) micro crystalline silicon tandem and (plant c) copper indium selenide (CIS) with the fourth plant (plant d) utilizing poly crystalline silicon. Graphs plotting the current and voltage output data can be seen on page XX.
The monitoring data analysis of all the four power plants show drop in plant yield, but more importantly very little deviation can be observed from the plots of maximum power point (MPP) currents against irradiation whereas a definite pattern, a marked decrease, could be seen in terms of MPP voltages.
It is clear from the data collected, at irradiation levels above 700 W/m², the fall in module output power shows a linear dependency with the observed drop in MPP voltage. Though the slopes vary from one technology to another, the observations are compelling and conclusive to assert the role of voltage drop in degradation.
Though the influence is irrefutable, the question could be asked, "everyone knows degradation means loss of output power, so why it is important to understand that drop in MPP voltage causes degradation?" The answer to this question lies in the inverter’s input I-V plot. Especially the I-V plots of the micro amorphous silicon (plant a) and micro crystalline silicon tandem (plant b) power plants are quite alarming. Modules utilizing two crystalline silicon thin film technologies clearly are degrading at worrying rates. This trend is clearly observable in that the 2012 data points are considerably further to the left than the 2011 ones.

Consequences
At this rate in the next few years of operation the MPP voltages of the PV arrays would run out of the inverter’s MPP window. Why is it a problem only in the two crystalline silicon thin film plants but not in the other two?
In general PV plant designs are based on extremities. The maximum system voltage of the modules and inverters determines the number of modules in a series connection. Thus a planner has to make sure the open circuit voltage of the string at higher irradiation and lower temperature conditions lies below 1000V, the current industry standards for maximum permissible system voltage.
Especially in the two crystalline silicon thin film plants, the technology characteristics such as higher initial open circuit voltages in comparison to stabilized values add more constraints; reduce the number of modules in series connection. Thus, a string designed to be within the maximum permissible system voltage at the initial year of operation runs out of the inverter’s MPP window in the next few years as a result of degradation.

Inverter trouble
Running out of the inverter’s MPP window means that the DC string operates at the fixed voltage, thus the inverter losses increase indirectly as a result of degradation. As such, this unexpected yield loss, in addition to the degradation yield losses, reduces the annual output of the PV plant – a worrying development for plant owners and investors.
What then is to be done? Within the current PV plant operation and maintenance frame work, identification of this behavior through monitoring data is very much possible. Once identified, the current system configurations have to be changed, such as rewiring of the DC string in order to increase the number of modules in series. This would bring back the array’s MPP voltage within the inverter MPP window.
But on the other hand, rewiring the entire park brings in practical challenges such as DC cabling running across a number of module tables and combiner box configurations. This would add further stress on the project's cash flow, as it brings in newer costs that weren’t accounted for during the power plant design and subsequent financial planning stages. Unanticipated costs are never welcomed by investing parties.
Alternatively looking beyond the current standards, increasing the maximum system voltage from 1000V to 1200V or 1500V would provide for additional modules in series right from the planning stages, thus giving the planners a little more room to plan strings which operate within limits all throughout the power plant's entire useful life. Though this seems to be a promising option, it can be useful only for new power plants that are at the planning stage. For existing power plants problems remain.
For plants already in operation, increasing the MPP voltage range of the inverters by lowering the lower MPP threshold, therefore allowing for lower MPP voltages at high temperatures and high irradiation conditions, would eliminate the risk of PV strings running out of inverter’s MPP window.

Next steps
Solarpraxis Engineering is continuing its research and analysis in this increasingly important field. Ongoing collaboration with inverter manufacturers is showing promising signs in working towards solving this problem. Consequences for crystalline silicon thin film manufacturers have still to become clear.