André Mermoud

Easy to imagine for a simple system. Most of the PVsyst projects use several inverters (sometimes dozens or hundreds of string inverters), connected to several kinds of sub-array (sometimes different sub-arrays on the MPPT inputs of a given inverter), etc. We are preparing such a tool but this is extremely complex in the general case. This tool will not be ready before several months.

Do you really think that PV plant managers will adjust the tilt of their PV array 8 times a year during 20 years ? By the way, the differences (gains) are probably very low. You can easily evaluate them by performing the simulation for different tilts, and gather the monthly results by choosing different monthts in these simulations.

1 - Is there any difference between simulation and reports in evaluation mode VS licensed software ? The capabilities of the software are exactly the same in "Evaluation mode" and in the the licensed mode. 2 - But at the reports, production of two systems was the same. In the simulation, the production result is mainly related to the STC specifications of your modules. If these are the same the result will be the same. In real, mono string is producing 7% more energy. So what could be the reason ? I don't know what allows you to affirm this. If the production is 7% better with Mono crystalline modules, it is because the STC values of these modules are 7% better. Do we need to use paid weather data ? It depends on what you want to do. An evaluation from Meteonorm V7.1 is quite sufficient for the first evaluation and the optimization of your system. Now paid data (especially for a set of 10 + years) may indeed be useful if you want to estimate the evolution along the future years (P50-P90 estimation). But nothing ensures that these data will better represent the present and future weather. Paid data for a specific year are useful also if you need real data for the present time, for a direct comparison of your simulated data with respect to measured ones.

You can do different simulations. You can use the "Batch mode" for doing several simulations at a time with different situations, and gather the results on an EXCEL document. There is a "ready-to-use" tool performing this optimization of sheds (choose "Advanced simulation > Optimization tool").

We have completely rechecked the import of SolarGIS data for the next version 6.68, to be released very soon.

I can't understand that. Please try again, using "Databases > Geographic Site > New". And make sure that before saving your site, the data are indeed what you have specified. The site will be saved when you click "OK", and at this time you can check the filename.

Probably you are using a very old version of PVsyst (V 5), where I think I remember that there were indeed no sites in Brazil. In the present version, the native Meteonorm 7.1 database shows 14 sites in Brazil. And of course, you can always choose any location using "Databases > Geographic sites > New", and here either choose a location on the google map, or define it by its GPS coordinates.

If you choose the option of applying the limit at the injection point, you will very probably obtain a clean output E_Grid of 27'200 kW.

We intend to improve this situation in the future. But this is really not simple, and it is not in our priorities.

In the present time, PVsyst only allows to define a simple output circuit, with one transformer. The loss is expressed in terms of percentage of the Peff (corresponding to STC) value. Now if you have several transformers, the easiest way id to simply add the ohmic loss (in terms of percentage) of the first and the second transformer. Working explicitly on the resistance of the transmission lines is also possible, but more complex. You should define an "equivalent" resistance (as "seen" from the output of the inverters) of the circuits before the first transformet, between both transformers and up to the injection point, taking the voltage ratios into account.

The use of "AC modules" has indeed been implemented in PVsyst since several years. For the association (PV-module - Micro-inverter), the remarks of the AC module should mention: "AC module (with LM305UE-G1 (240V) AC module of LG Electronics)" and inversely the remarks of the inverter will mention "AC module (with LG 300 A1C-B3 ACModule 240V of LG Electronics)". However only one device has been proposed by manufacturers up to now. This is the LG Electronics LG 300 A1C-B3 AC Module 240V (or --- 208V). When choosing this device, the corresponding micro-inverter LG Electronics LM305UE-G1 (240 or 208V) is automatically selected. In the sub-array, please make sure that you define the same number of inverters than the number of modules.

SolarGIS provides its data in so much different formats. We cannot automatize the import of all of them. Therefore you have to use the general tool "Databases > Import ASCII meteo files", and here press F1 for the procedure.

The clear day model in PVsyst is based on the air mass, and implies the the Linke coefficient, which is set by default to 3.5. It uses the model of Kastens. You can create a Meteo file (*.MET) of clear days for any location in "Databases > Import meteo Files", last choice in the meteo sources.

With the tool "Advanced simulation > button Output file", you can create a CSV file of hourly values. In this tool you can choose any variable among the >50 variables involved in the simulation. Obviously including the system output E_Grid.

The bifacial model for tracking systems is in preparation. It will not be available with the next version 6.68, but probably with the next one.

In the version 6.67, the IAM specified by the user is not correctly read on the custom pan files. During the simulation it always reverts to the Ashsrae model. This is a bug just in V 6.67, which will be corrected in the next version.

You are quite right. PVsyst uses indeed the expression you have calculated, with a RshRef as you propose. I have to change this in the help.

This is in preparation for a next version.

With the tool "Detailed simulation > Output Files", you can choose any variables among the >50 variables involved in the simulation (including of course the system output E_Grid) for hourly values ourput. In this tool, press F1 for the procedure.

Yes, the Site file (*.SIT) and the meteo file (*.MET) are associated to the project when opening the project. Now if you have performed a simulation with a given site, the variant file (*.VCi) will store the values of this site as Input parameters of the already performed simulation.

This is just the definition of the IAM: "Incidence angle Modifier": it is the deviation of the reflexion loss whith the incidence angle, with respect to the normal. The curve of First Solar represents the total irradiance reflexion losses. But in the usual definition of the IAM, the loss under normal incidence is fully included in the STC measurement / definition.

Yes, sorry. There are indeed problems with IAM. In the version 6.66, the IAM was not read in the internal database. This was the main motivation for creating a new version 6.67, where this is corrected. However in the version 6.67, the IAM is now not well read in the *.PAN files of your personal database ! Therefore if you want to perform simulations with your own PAN files with customized IAM, you have to revert to the version 6.66. You can download the version 6.66 from our site www.PVsyst,com, and install it in parallel with your version 6.67. Both versions may run independently.

The simulation is suited for using pyranometer measurements. Reference cells may be used for the evaluation of the stability of the PR, but suffer of IAM and spectral discrepancies. You also have to check the ageing of the reference cell (like the Pyranometers, which should be recalibrated form time to time).

There is no rule. We cannot give an advice about the better data source. And this is evolving along the time. Now I don't see any reason why Meteo stations of higher class would give higher results. Please have a look on the Help of PVsyst "Meteo Database > Import meteo data", which lists and analyzes all the sources available in PVsyst.

Yes, the google map service has been changed by Google. Please see our FAQ When defining a site my "Interactive map" doesn't work.