SAM's degradation methodology results in reduced accuracy for clipping inverters

2 posts / 0 new
Last post
bobsil
SAM's degradation methodology results in reduced accuracy for clipping inverters

SAM currently calculates the effects of degradation by simulating only the first year and then simply applying the calculated degradation value to derive the power output for each subsequent year.

In the case of a clipping inverter, this leads to inaccurate results as the actual effects of module degradation will have a lesser effect on the output power than what is predicted by this methodology.

Please consider providing the option to apply the degradation factor to the DC power calculated for each hour and running a complete simulation iteration for each year.

Paul Gilman

Thank you for the suggestion.

Your observation is correct. In addition to simulating only the first year, SAM applies the degradation loss to the system's AC output, so, for a PV system, it only very roughly estimates the effect of module degradation on the system's performance. The main reason SAM only runs a simulation for a single year is to minimize the number of calculations. This is especially important for people who run parametric simulations or write scripts to run hundreds of simulations.

As it is now, for a simulation based on an hourly weather file, SAM performs a set of 8760 calculations for the first year simulation, and then a set of 20-30 calculations (one for each year in the financial model). To model degradation more accurately, it would have to perform 20 or 30 sets of 8760 calculations. In some cases, people use SAM with subhourly data, which increases the number of calculations for the annual simulation. Also, to model module degradation more accurately with a simulation for each year, you would need a separate weather file for each year. It would not make sense to simulate each year separately using a typical-year weather file.

There are a number of modeling scenarios that require running simulation for each year, so adding this capability is on our list of things to consider for a future version of SAM.

UPDATE (MAY 11, 2017): In the latest version of SAM, there is an option to run a separate simulation for each year, which allows for more accurate modeling of inverter clipping losses over the life of the project, and for systems with batteries, allows for modeling of battery replacement as a function of the number of charge-discharge cycles. To enable this option, on the Lifetime input page, choose PV simulation over analysis period:

screenshot of PV lifetime options

Note that the "PV simulation over analysis period" uses a single weather file. For now, it is not possible to use a separate weather file for each year.

Best regards,
Paul.

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer