Webinare PVsyst 8.1, CLI & PVsystBasic 1.0

Added: 2026-05-06

[00:00:38]Hello everybody and welcome to our webinar today on of PVC. My name is Bruno Vitma. Uh with me is Orian Kanes and later on there will be Mikuel Lozi also joining us to present you the novelties on PVCist 8.1 that was released a month ago and also on the other products that we have been developing and which are new PVist basic and the novelties of PVist CLI.

[00:01:06]So uh the outline of the webinar the will be organized in four parts. We will start by presenting a little bit the novelties of the PVS version 8.1. Uh namely the weather data import wizard and how it uh improves the importing of weather data and the subhourly simulation which is really the the main major improvement of version 8.1.

[00:01:30]Then we'll take a short time to briefly present a new product been uh we've been developing. It's called PVC basic and it's a simplified simulation of solar pumping systems. Right now you will see later uh how it works. And uh in the third part we'll have a presentation of the novelties and PVC CLI the command line interface of PVC where we have introduced recently the possibility to create new sites uh automatically with a command line. Finally, at the end, we'll have a Q&A session. So, where you can we'll try to answer your questions. The questions you have, you can insert into the Q&A tab that you will find on Zoom.

[00:02:14]Please write all questions in English.

[00:02:17]questions which are not in English, we will simply refuse and ask you to to reformulate them in English because our team which is behind the scenes and accompanying us are doing a great job uh need you to answer to all all all the questions. Um there is also a separate chat where you can where participants can chat between them. So this you you may use but we will just moderate a little bit on that there will not take questions which you write on on on that chat. So please if you have questions use a uh tab that is for seen for that. Finally, I would like to remind you that this webinar is being recorded and uh in a week or so we will publish it also on YouTube and the questions uh that we will not be able to answer today if there are too many we will nevertheless uh record them write them down give answers to them and when the video of the webinar will be published we'll publish also answers to all the questions that we didn't manage to answer today. So uh in case you are in this situation that you ask a question that we won't answer, you have to be patient and wait for the video.

[00:03:29]And with this uh I start the part on PVC 8.1. So the latest PVC version that has been released and where the major improvement was a subhourly simulation that we will show you later on. But before we do that, I will pass over to Orian who will show you how the new weather data import wizard works. Please, Orian.

[00:03:51]>> Thank you, Widow. Hello everyone. Um, so let's get started. So our major update is subly simulations. And of course, if you need to run a subly simulation, you do need subi uh weather data. So currently in PVsyst 8.1 there are uh two main ways of getting subly data to run your simulations.

[00:04:15]The first option is the new updated methonom DLL meteor 9.0 that allows you to generate synthetic data at the minute scale. And then the second options that you have available is our custom file import that has been completely reworked and uh now allows you to um import any kind of data you wish from the minute scale to the hour scale.

[00:04:46]So I will um share the psyest uh screen and I will show you briefly how it's done. Maybe I will remove this uh presentation. Thank you. Thank you, Reno. Um, so let me just highlight my Perfect. So we'll be looking at the database tab of PVS and we'll start by uh using the new metronome import. So the new metalonom import is uh available when you look at geographical sites here and you'll first need to create a new site. This is um the same interface as we used to have.

[00:05:32]So for any new site you can choose a location on the map accept the selected point. And here uh the main uh update is the weather data import. You now have the metronome version 9.0 O and here you can select to either import audi data or a minute data. So if I click on minute and import the data here in the background you are already generating your met file and the results of the met files will be also saved in the site as uh monthly weather values. So here you have the monthly values updated and when I save my site uh so that's the first stage for saving the site it's saved and after the site creation I get also prompted about saving my uh new weather data and yes here I do want to save it I'll keep the default name here and that's already done I have u my file available and I can look at it in the weather data tables and graphs here. Uh let me open it.

[00:06:44]Yep, it's the right file. And if I look on graph here, you now have the subi uh option to display your data. And if I have a look at my radiance, I can see that uh I have a nice little file for my simulation.

[00:07:01]So that is uh the first approach.

[00:07:04]Now the meteor DLL is also available um in the synthetic data generation tab here. So it's going to be the pro the similar process behind the scene. You can select a site the data here can come from uh any any source you wish and you'll generate the same um data from the monthly values that specified. Here is the choice between hour and minute option. So if you select minute here you'll get one minute data file.

[00:07:39]Now the second option is the import custom file. Um this uh feature has been available in previous version. Uh but the interface has been completely reworked and it looks very different but the process and concepts behind it is quite similar. So you need uh a CSV file as your web source that you first select.

[00:08:05]And then in order to import your file, you need uh what we call a meth file.

[00:08:10]That's a formatting file that will just tell PVC what data is to be found where.

[00:08:15]Uh you can either create a new one or use an existing one. So here in this demonstration, I'll just uh use an existing meth file. from this this first window. You can already uh display your CSV file here and I can already check if I have selected the right uh separator to read my file. If I choose the wrong one, you can see how it displays clearly I need semicolons. So that was a correct choice.

[00:08:46]In the next step, I will then have to choose the site, the geographical site associated uh to the weather I want to import. It's a very important step for us to be able to compute the sun position correctly. Uh again, either create a new site or use an existing site uh in the database. Uh I will look for the site corresponding to my data.

[00:09:11]Oops, sorry. I was typing at the wrong spot.

[00:09:18]up. Here you go. Uh I get all the information displayed and I can move on to the next step.

[00:09:26]And here we arrive at the main window of really uh the beth file itself. So again I can display uh my source file next to it and I will check uh all the parameters that I have defined here. So the most crucial parameter is probably the time step here. Uh that has to match what is in your file. So here I can check in my minute column. I do have five minute data. So my time step is defined correctly. I have to define how many header line I want to skip.

[00:09:57]Then in the date type I'm going to define how the date is read. So here my dates are read on the file. There's a one column per uh month, day, hour, minute. That is all correct. I can check it here. I can then have the finer uh settings. Is there daylight saving in my file? Do I use legal time, universal time, time shift? All of this is available on this window.

[00:10:25]Uh then on the next tab, I have my variable mapping. This is where I choose which variables uh I am going to import in my in my file. So here I'm importing uh irradiance, ambient temperature, wind and humidity.

[00:10:42]Uh and the PVCs variable you have chosen will appear on the top. So you can check that you have selected the right columns.

[00:10:50]So the file has already been converted in the background. But here if I look at the message uh box I do see that there's something wrong. I have a red message.

[00:11:00]Many many things were were generated. So I should really check uh what's going on with my data. And I will open the weather table and graph uh window here.

[00:11:10]That's what I get. Um so here don't panic. I only have one point because my file is one month of data only. So that's okay. But I do have a message telling me here that there's something wrong with my relative humidity. If I want to learn more, we now have a new tab called the generation log that will display every single message uh from fatal errors to infos uh that happened during the convers conversion process.

[00:11:39]And here it tells me there's something wrong with my uh humidity that the range is weird and I should check the unit. So I can go back to my uh window here and check what I've done. I imported the humidity as a ratio but clearly the values are for a percentage. So I will correct that. Change the ratio to percentage. And as soon as you change something PVC uh will recalculate and like reimport the data in the background. Uh now it looks much much better. I have almost nothing in my generation log. I'm just told only that uh I have a few time stamp missing in my file and that PVC has filled some of the missing values. If I want to check that uh I can go even more in depth for example in my graphs check my uh sub values uh the missing data was indicated at the very end of my file. So I'll have a look at what's happening. And here what you see is uh unfortunately two hours of irradiance data missing. So in this specific case I have just truncated my file a bit too early and I'm missing uh two hours of sunshine. If you ever get missing irradiance data in your imported data uh it will be considered as zero.

[00:12:58]So just there will be no simulation at this hour. So let's say I'm happy with what I have obtained. I can close this window. go to next. Here I'll be prompted to save first my meth file and yes I want to save the correct unit.

[00:13:16]I'll write it and finally on this uh final window I can just save the weather data that I have created.

[00:13:27]It already exists. I'll override it and and I'm done.

[00:13:33]So that's really um that's really it for the import and generation of subhourly weather data for now. In the future plans uh we do need to um improve this import to allow you to import multiple years of data at the same time. Right now we still have uh the limit that we had in PVC 8.0. you can only uh import one year of data. Each meth met file sorry is always composed of one year of data at maximum. And uh in the next step we will also uh update the known format import. Right now known format only works with hourly data but in the next uh releases we'll eventually switch to be able to also import subi data uh with the known format function.

[00:14:26]And I think that's the end of my part, Bruno Buo. So I'll give you control again.

[00:14:32]>> Thank you very much Arian for this nice presentation. I will quickly close this and come back to to our slides.

[00:14:40]Uh so once we have um created um a subhourly weather file and uh let me say that you can choose the you can do a simulation now with that and the simulation that you will run in PVC is based on the time steps that PVC will find in the weather file and you have the possibility to create time steps down to one minute but also larger time steps like five minutes we've just seen an example of Oran 10 minutes, 15 minutes. I mean, it's your choice. And you see here on the bottom the simulation results of hourly steps and 1 minute steps makes a big difference in the precision and and the details you get out of your simulation.

[00:15:23]This is an example of generated PV power with clipping. And you see that the clipping is much better uh modeled if you have have the minute detail than if you use hourly averages.

[00:15:37]So to be able to implement this subharly or to to make a reasonable subhourly simulation in PVC cyst we had to adapt some physical models. So it's not just simulating the same way in shortest time steps but we had to make sure that also the physical models are adapted to these shorter time steps and there were mainly three models that we needed to be adapted for uh good good results. The first one is a de composition of direct and diffuse irradiance that you need when you have weather data or import weather data that has only GHI for example or only POA radians. PBCS needs to figure out which part is diffuse which part is direct and models are used for those and they had to be adapted for the subarly case. Then you have the transposition model that goes from the irradiance of the on the horizontal plane to the irradiance on the POA.

[00:16:31]These models also that's a Paris model that's typically used by default in PVC and these models also were developed and validated for hourly time steps and when you do this exactly the same thing with several time steps you get biases so we needed also to adapt those models to avoid this bias in the subhourly simulations and finally there's a thermal model of the PV modules if you make hourly simulations you can admit that uh you can use an average temperature a stationary temperature for for the whole hour. But if you go down to the minute scale, you have to take into account the thermal inertia of your PV modules. And this was also um put in this is the example you see here on the right side of the slide. You see the difference between the steady state model which basically follows instantaneously the radiance and a more sophisticated system uh model that that has some thermal inertia built in.

[00:17:27]There are other things we need to adapt also namely the the importing of time series for simulation parameters. I remind you that if you have for example self consumption you can import load profiles or now you can also import time series for the grid limitation for soiling power shifting and so on. And of course if you import these time series you don't import them together with the weather data. So you can have a mismatch between the time steps of your time series and of your weather data. and the simulation needs to manage that. So all these were developments that were necessary to get a subhourly simulation and I will show you now how the severly simulation looks like uh by just following up the example from Uran and making a simulation with a weather data she generated. So I'm going here to project creating a new grid connected project. It's completely empty. I will choose a file. By default, PVC will present me the last generated site which is the one that Orian just did. So I say okay and PVS identified this 5 minute weather data that was generated. This is one month in July that has uh has been generated for for this demonstration. We take only one month to make simulation time short enough for for this webinar. So I save my project. This is my example.

[00:18:53]for the webinar and once I have saved that I can create a variant. Now I have um prepared a variant so that it doesn't take too much time uh which has trackers on terrain. I will import that from from the other project into this one. So here we have a system a large system which is on on a terrain. Uh these are horizontal axis trackers and as you see now PVS detects that we have five minute weather data here and proposes us two types of simulations. The hourly or the subhourly simulation and the subhourly simulation will follow the time steps of the weather data. So I can just run the simulation now and you will see that uh for 1 month it's reasonably fast. Of course, if we make a simulation for the whole year or with minute time step data, this will be accordingly longer.

[00:19:57]But you see we have the results here.

[00:20:00]And I can look at hourly graphs. For example, just looking at the results of the generated um energy.

[00:20:11]And you see I have now um my my results here in in five minute steps and not anymore in hourly steps which were much much coarser.

[00:20:24]So this is one way to simulate and uh for the situation where you have minute step data or you want to simulate a full year of course this will take some time.

[00:20:35]If your weather data spends a full year, you might need to find more efficient ways to simulate your your project.

[00:20:46]So we have foreseen this as you've seen before. We we have here the option to just switch to hourly simulation and run the same exactly the same project now in hourly steps. And you see that goes much much faster.

[00:21:04]We have the hourly results. We can look at the report and if you are interested in in understanding where the subhourly simulation gets you more precision or gets a different result, we can compare the report of the two simulations right now.

[00:21:20]So I open the report, I compare it with my five minute data and you will see that the two results are quite similar uh and that it's mainly where the subhourly models needed to be adapted that we have a difference between the two results. So you see here on the left side the hourly results on the right side simulation with 5 minutes. You see that there is the difference between the two in performance ratio and also in bacial performance ratio does not differ too much which means that the hours simulation for for this example is is quite good. If we go to a detailed comparison in our loss diagram, I will not run through through all the differences, but I will just point out that the main uh differences between the two results come from the transposition which is the very first value you see here from the diffuse irradiance on the rear side which is uh defined by the subarly models that have been adapted and finally the clipping losses of the inverter which are quite high here because we have a big um AC/DC ratio uh a very large DCAC ratio for for this example. You will see that the differences are also mainly in in this part of the clipping losses. I remind you that the the hourly simulation of PVS cyst already had the possibility to account for clipping subhourly clipping losses in case you had subhourly weather data. So if you were importing subhourly weather data and this has now been is more automatic in in our case because we simulated hourly steps with a subhourly weather file and PVC automatically applied the model and we see here that we have a gain or or additional loss due to subhourly clipping of 0.88% on the left side. And if we compare the two values now of the detailed simulation we see that this this correction matches quite well. it's overestimating a little bit the losses.

[00:23:28]So of course the the detailed 5m minute simulation is better and more reliable.

[00:23:33]But you see that even even though you um you can run an hourly simulation or already get a very good approximation.

[00:23:42]And this brings me to to my last point.

[00:23:46]Uh I will just save this file hourly results brings me to my last point. So the subhourly simulations of course give you u more precise results and a and a better simulation but of course it comes at the expense of a longer simulation time. And we try to to cope with this first by giving you the option of quickly switching between hourly and subhourly. So in an early design stage you can run just the hourly simulations and then when you need the detail when you make your final simulations then you can easily switch to the subhourly simulation. And the other thing that is useful to remember in in these cases is for uh in the advanced simulation you have the possibility to make a simulation over a reduced time span. So here we didn't need to do that because it's only one month. But if you have a full year of minute data, you can for example choose to run only over the summer uh months where you have for example clipping to understand the clipping and the details of the clipping. And then of course your turnover of simulation and analyzing will be much more efficient than if you run always the full year.

[00:25:01]Okay. And with this I will close this part of the presentation.

[00:25:07]come back to our slides and uh this is the end of our of our first part of PVC 8.1. Thank you very much Orian.

[00:25:20]In the second part we will talk I will present you very quickly a new product that we have developed which is called PVC basic. It's in principle a simplified PVC simulation for small systems and it's really aimed at rural areas which are far from the grid and it's it's meant for um to be able to make an easy system design and sizing for uh people which have also low level of of training a limited training and simulation experience. So the original idea is really to help in in regions or areas where people have difficulties with access to water and electricity to give them means to reasonably design and size small PV systems uh and solar pumping systems that they can set up. Currently PVist basic will only simulate pumping systems. You see a screenshot example on the right side. But on the uh long-term in a future version, we will also add standalone systems to this. The tool contains also a set of technical tools that help with the understanding of of very basic electrical concepts. I will show you later on one or two examples of how this works. So let me show you then um how PVC basic looks like.

[00:26:41]I will pull it over to our screen.

[00:26:44]Minimize this. So as you see it reminds a little bit of what you know from PV cyst and that's no coincidence. The simulation engine that is behind PVIS basic is the same as PVC. The only thing is that we stripped a lot the parameters that you can define. So there's much less level of detail that you can define and we also simplified the way and or the approach how you set up a simulation. And I will just open uh an example that comes with PVC CLI um which is a pumping system for a well and principle for for running a simulation with this tool. You have a linear workflow on the left side and the simulation or guidance of what introducing of parameters on the right side. This is a uh complete demo project. So you already have simulations results, but you can navigate back here to the different steps of setting up your system. Or if you want to be quick, you can just click on on this button and you start by selecting weather data and the system type. Right now you can pump from deep wells or tanks or from a lake or a river. In this example, we have a well. Then you define the water needs that you have. You have different levels of detail. Here we have 40 cubic meters per day and you specify the reservoir that you want to to fill up w with your water. Yeah. The the dimensions. In the next step, you put the left side. And you notice that this image that basically represents the whole system what you're setting up has now passed the highlight from the water tower in the beginning to the bore hole and and the well here on the left side. And this helps just to to focus on the parameters that are being entered in this step. So you have to give the details of of your friatic levels. Then you have to specify the pump. The pump comes from a component database which is the same as the PVC database. So you have access to the exactly the same components. and PVsyst uh chooses uh or marks the the pumps which are adapted for the system and and match to to our previous choices. So I keep this choice. Then we have the same choice for the PV module.

[00:29:10]You see again find everything that is in the PBS database.

[00:29:15]Uh but you have limited way to define your strings of and and modules in in series and parallel. Finally, you set up the orientation. So there is no shading 3D shading model in in this tool. You just have plain PV modules which are oriented in a in a certain direction of the sky. And in the detailed losses, you see the only thing that you can define is a thermal coefficient of the PV modules and the losses in the in the cables in the DC cables.

[00:29:46]Um to give you an example, let's say for for this site we we choose a little less inclined panel at 10°.

[00:29:58]I come back to my detailed losses and to my results. I can now have everything in green here on the left side. I can run my simulation. If I do this, it goes very quickly. It's hourly steps. This will not work in subourly steps. And I have now my results which are shown here. And I can have a a report that resembles quite a lot uh the PBS report, the normal PBS report with just less pages. Yeah, it's it's just a concise way of of displaying the results.

[00:30:35]So this is basically functionality of of of this model. we you can save your projects and and then go go over to the to the next one.

[00:30:49]Okay. And this is basically what what PVC basic is doing.

[00:30:55]I'm coming back to my presentation.

[00:30:59]Um again the as it's meant for for regions and areas where the idea is to help people to have access to water and and uh electricity.

[00:31:15]the license uh for this tool which is separate from the PVS license is sold as as a at a symbolic price and as you as you've seen it's really meant for a very specific case of of rural areas and and small systems and with this um I end my presentation of PVC's basic I have now with me Michaela Olioi who will take over and um explain you say a little words on PBC CLI the command line interface please.

[00:31:48]>> Hi Bruno and thanks for the introduction. Um so PBC CLI is our third product we're presenting today. Um CLI stands for command line interface and it basically serves the purpose of running key PVC functions from the command line.

[00:32:09]The idea and let me try to thank you get the slides is that um PVS CLI will support the automation of your whole modeling process and try to speed it up and so PVC without CLI itself um is actually quite good when you're in that iterative phase of design whenever you're changing electrical components or modifying your 3D scene or maybe changing the granularity of the simulation but once you have set up your workflow and you have to do multiple projects all those uh UI and UI interactions can be quite tedious and timeconuming and um well actually we don't want you to miss your Christmas or any other celebration because you're running BBC simulations and you have waiting for the next batch. So this is where BBC CLI comes in handy.

[00:33:08]It allows to set up an automatized workflow and try to remove those time consuming actions in the UI and so the idea is that you will then program it to perform the actions in your set. So let the machine take over the whole running the simulation spark.

[00:33:32]So we are presenting PVC CLI today also because there were there were some updates with PBC 8.1. So we also have the PBC CLI 8.1 update. The main functionality that was added was the possibility to create sites.

[00:33:48]So the SIT file is a very central part of a PBCS project. And so previously one could only work with weather and project data in CLI. So now it's actually possible to define allocation and ask PBCLI to create the SIT file. There is one um necessary input which is the monthly weather data.

[00:34:14]This means that this functionality will work best whenever you pair it with a weather database from which you can pull your weather data. And I'll show you some examples later, but uh I can already say that this opens up many possibilities. Um in terms of use cases um first you can uh if you're doing some O andM some monitoring you can kind of automatize uh the production of your expected values for your production and uh you can also try to search for new sites with prospection tool since we can create a new sites you can just plug yourself in in a specific location and run a simulation at that place And you can also uh use the batch mode since a few updates uh which can help you with system optimization. So basically these three different use cases are now available with CLI.

[00:35:11]So I want to show you some examples and I was uh thinking of the prospection in particular. Um nowadays it's really easy to cook up um many different interfaces and apps. Um, and I wanted to show you how to build an app out of PBCCI. So, here's the pipeline I wanted to use. Uh, the idea was fairly simple as I explained before.

[00:35:36]You can take a weather database. Um, the first step would be then to make an API call and gather some weather data, ideally monthly weather data. Then you can ask BBC CLI to do two steps. The first steps would be to create a site and the second step would be to run a simulation with an existing project based on that site. And uh on that note, you don't need to import more weather data. Uh there is an option within the run simulation to generate some synthetic hourly or subhourly weather data so that you don't can skip that step basically.

[00:36:15]So I want to show that um actually in in practice. So let me just escape here.

[00:36:23]And so there you have it.

[00:36:27]And so basically this is uh works on the pipeline exactly as as mentioned. So let's try to look up for for a location.

[00:36:35]I'll try to aim for Geneva roughly. And so we can add some information but all of this is uh let's say fluff here. The idea uh is that you already have worked in within PV assist on your project. And so I have here a project that I worked on before called default project and some varants. And here I'm interested in a row based system with a 10 degree tilt. So I will choose this one and simply click on run simulation. And so now my app is asking PVC CLI to create the site based on the weather data it fetched and then run the simulation. So you can see now the step is running simulation.

[00:37:25]And so this has allowed me to do something that would have taken maybe five or 10 times more uh minutes within PVC. So now I have some valuable KPIs for that location. Um, this is all programmable of course. So I mean you could imagine anything you want with this tool.

[00:37:47]And there you have it. So going back to the the slides we had before.

[00:37:56]>> You have to get the focus.

[00:37:57]>> Yeah. Okay. So getting back here, our wish is that you let the robot um cook up many different apps for you and uh ideally that you try to be creative with all the whole possibilities that BBC CLI has. And um actually we're really looking forward to um you coming up with many different use cases and we will definitely help you set them up. Um so feel free to get touch uh get in touch with us and uh ask us uh about your dreams with PVCist.

[00:38:37]Um now let me just end with a small reminder. So if you want to get all the products we've shown today so either PVC 8 uh 8.1 PVC cli or even PVC basic you can just go to our web page uh which is www.pvys.com.

[00:38:55]Um the installer for PVC and PVC CLI is the same. So once you press download you will get basically both products. Um but you need separate licenses for them. So if you want to use PBCCI beyond the trial that lasts two months then you need to go to this website and purchase a different license than the PBCist one.

[00:39:20]And there there we have it.

[00:39:23]>> Okay. Thank you very much Mikille for for this nice nice presentation and for this example which I think shows clearly the the potential that you have with with PVC uh CLI. Yeah. and and what you can do with that. And we are looking forward to to put more functionality into CLI and and to see this develop because we believe that the automation of workflow is important nowadays especially in in larger projects and we are making efforts to get this accessible and then working also with the PVC simulation.

[00:40:00]And with this we are basically closing the part of the presentation that we wanted to give. we come to our Q&A part.

[00:40:09]Uh I remind you that there is a special tab in in the webinar for creating uh for for asking your questions. We should now have um already our team in the background who had has picked up your your questions and is we will get them here on the screen soon. Just hold on a second and we will try to answer them.

[00:40:36]Okay, good. Uh, see if I get it.

[00:40:54]Okay.

[00:40:56]So, um the first question I see here is uh related to PVC 8.1 and it says do we have an option to import solar giz data directly not ty data but only data that we can import with solar giz prospect.

[00:41:13]Uh so I think this is a question to who presented to us the weather data import wizard. So could you try to answer that question?

[00:41:24]I'm sorry. Could you repeat the question?

[00:41:27]>> But we have an option to import solar giz data directly not ty data but only data that we can import with solaris prospect. Solaris prospect the the product.

[00:41:38]>> Um so right now I think we have uh two ways of importing solar data. There's the site I think it's one of them available in the API.

[00:41:49]>> Sure. Do you want me to answer the question >> is if you know >> well I can try. I'm not sure I will answer you know correctly. uh but there are you know two ways of you know importing you know sele data at the moment you know through site as you said so you can import you know PMY so is you know space on the web you know API uh and then you know you can also you know import it you know via uh the non format so you basically you know download you know the CSV a CSV file you know from solist website and then import it into pist so there's you know two ways of you know doing it >> and maybe I can add to This as Orian mentioned in in her presentation. Right now this uh the known formats always generate hourly weather data files for for PBS but in the future if you have a finer uh granulation of your time steps we will also account for that >> and I can also add something that um also the API calls themselves will be gradually expanded because right now we're just uh generating the metronome data from the site creation and the other features will also come along u in on the later batches.

[00:42:59]Okay, the second question is about the the sun position. How are the sunrise and sunset adjusted to the subhourly time interval? The sun doesn't rise or set at specific time intervals. So, how is PVC dealing with accurate time adjustments? So, um I I can answer to that. The sun position is indeed very important to do get the right simulation. You have to have a representative sun direction for the simulation step you are performing. And in PVC, we place the sun in the center of our time interval. So if it's an hourly time interval, it will be half an hour from the from the borders. If it's a 1 minute interval, it will be 30 seconds from each borders, but it's always centered on the time interval we are simulating.

[00:43:43]There are some special cases if the sun gets over the horizon line where we only take the part where the sun is above the horizon. So then you are not exactly on the time step center but slightly uh uh apart from that. But this is how we we do this and and with this we get really a sun position which is representative for the time step we are simulating.

[00:44:09]>> Should I ask question three?

[00:44:11]>> Yes.

[00:44:12]So the third question is um is the grid limitation calculated with subari time stamp if you're doing a subari simulation or is this loss always only showing at inverter level? I think the question is more about uh whether the grid limitation also works simply with subhourly simulation.

[00:44:31]>> Yes. So um yes indeed I mean we in the the subhour simulation really goes in subhourly steps. All the the simulation stages that we do in hourly simulations we do equally in subhourly steps. Now uh the the grid limitation in PVC up to now was limited to to monthly values but this has been changed already in in version 8.1 the current version 8.1 where you can import a time series and this time series can again go down to one minute time steps. So of course then if you make a 5minut simulation with 1 minute time series we need to average over the 5 minutes but in principle you have the granularity of your simulation also for the the grid limitation and this is true not only for the grid limitation but also for load profiles for example and soiling profiles that you can now import in subhourly steps and PVC will will match the best possible these time series to the simulation step that is being performed.

[00:45:34]Mhm. So maybe we have uh a question on PVC CLI now. Uh do you need a license for each user of CLI or is only one CLI license on a Windows server required?

[00:45:50]>> Well, the the answer is that you just need one license on a Windows server.

[00:45:54]It's per installation. So the idea is also to try and run it on a different uh infrastructure that is not your personal uh PC or work PC but something with a bit more power and that could be shared.

[00:46:09]>> Okay. And the second question is is there a manual for for PVC CLI and is there a separate license required if there's already a license for PVC state?

[00:46:19]>> So separate licenses as mentioned. So you need uh I mean two licenses if you want to work with both and the manual.

[00:46:28]Yes, there is also a help um page documentation on the web just as the PVCist documentation is on the web. You can find it um rather easily. Uh let actually let's show it directly. So, so here you have the support tab and we get to PBCCI help and this is really looking like the PBCS help. Both are connected. Here you can find all the information you need to start um working with PBC CLI. I have to add that um there are also some indications on how to add some example of scripts. So you can find all of this in the help and um we hope uh yeah if you have any more questions of course feel free to ask us directly.

[00:47:29]>> Okay.

[00:47:32]>> Yeah. Let's go back to PBCist. Um I think um questions four and five are go kind of two together. The question is whether you can convert uh either monthly data or hourly data into subhourly data and then run the simulation with with PBC of course. Um maybe and Eric can answer this question.

[00:47:56]>> Sure we can obviously well we can try again. Um it is possible indeed you know to convert you know monthly uh data uh to you know uh hourly you also subly data you know via you know synthetic uh data tool uh which is available in pist uh so this is based you know on metronome uh data uh just model I believe >> but I might jump in and add that um since we're talking about a stoastic process if you generate uh two different file, one hourly, one subourly from the same monthly values. Uh they don't necessarily match day per day. If you really want to have a direct comparison, you should generate a one minute data file and then when you run the simulation, you should choose the hourly version to really have something matching your sub data.

[00:48:55]So yeah, I there is a question about timestamps. So we can continue with Orian and Eric. So is the convention of uh subarly data beginning of the interval for the time steps?

[00:49:08]>> Um yes. So by default that's the convention. So in PVSYST when we say the time step is uh 12 hours uh that refers to the beginning of a time step. So if you are in five minute data your 12 uh hour is from 1200 to 1205. That's our convention. And then when you import uh data through through the wizard, uh you can choose that like there's actually a setting and you can choose if your actual uh CSV file has the same reference or if it uh uses a different uh reference, for example, the end of a time interval and if nothing works, you can also like finally tune your import with a time shift.

[00:49:51]>> Okay, thank you. There is another question concerning the subarly simulation that says did you study the impact of the uncertainty decrease? What percentage of improvement can we expect with a minute or 5 minute data compared to 1 hour? I mean this is a very vague question because these several steps have uh impact on many different parts but I think the most important part will be the clipping and maybe Michael you can uh give us a comment on how it uh changes the precision of the clipping corrections.

[00:50:23]>> So so indeed if you use hourly simulation without the clipping correction model that we were using in 8.0 O um then depending on the DC to AC ratio you had um different undervaluations of the clipping loss in the hourly simulation. Um if I recall correctly that undervaluation of the loss so overestimate of the production was uh around from 0 to 5% depending on the DC toac ratio. Now with the sub simulation that is cleared.

[00:51:02]However, I have to add that the main uncertainty of the simulation is usually not the models but rather the weather variability and that doesn't change much with um the going to subarly scale and also the user input. I mean there is just some parameters that you don't know very well and uh you cannot be sure that they will be implemented as they are in the field.

[00:51:25]So that also is not really improved on.

[00:51:28]But regarding models, yeah, I would say um at least the transposition, the clipping um temperature could be a couple percent gain in accuracy.

[00:51:43]So thank you, Mik. We have a question on PVC's basic I see.

[00:51:50]>> Right. Um yes. So is PBCS basic included in standard PVC license or available separately?

[00:51:58]>> It's a separate license indeed. Yeah.

[00:52:00]PBCS basic is also a separate license.

[00:52:03]But as I mentioned since it's aimed at at situations where where people have difficulty and access to to water and and electricity, we really made a very very low price which I would call symbolic. So it's it's not hard to to to get a PVC basic license as well.

[00:52:25]So maybe another another question on the command line. Can uh can we run PVC CLI with one day data only?

[00:52:36]>> Uh yeah that's no problem actually. So um PVC CLI can be run uh with any weather data. So file that you create. So if your date uh file has only one day then the simulation will run for just one day and so that's definitely possible and um yeah I think this extends to any period within a year basically we don't have yet the possibility to run more than one year but then you can run multiple simulations.

[00:53:08]>> And then the next question is a little bit similar. Can we run the subar simulation with PVC?

[00:53:13]>> Yes of course. So this was updated uh with a new option in the run simulation command which is the time step of the simulation.

[00:53:24]And um the next questions if I already have a large number of site and met files, can I use CLI to run simulations of multiple variants on each of these sites? For example, I could create these variants beforehand using Python by processing a text file. Uh yes, that is possible. Um well, as I've shown, I just had some VC variants in in the app I showed before and then generated this SIT and ME on the fly. So even though you don't have SIT and ME files, you can do that. You can do it with existing SIT and ME files of course. And um I have to add on the on the idea of doing varants, you can also use a batch file. So either you can use your own scripts to modify um the the projects as as you want but ideally you also want to use the batch file which goes through PVC to modify those those variants.

[00:54:26]>> Uh next question what is the purpose of weather data conversion with PVC CLI?

[00:54:31]Uh, good question because I haven't used it in in my app. But the idea is that if you have um some actual weather data that you want to use for the simulation, you need a way to turn it into an ME file to then use in the simulation. So the weather data conversion serves that purpose to take that data and feed it to PVC.

[00:54:58]Basically, it automatizes what Orana showed you in her example. You have the meth file which describes the format and you have a CSV file with your measured data or your weather data. And from these two together, you can create the ME file that can be used then in the simulations. And this you can do if you have the MEF file and the the CSV data, you can do this then automatically with PBC CLI. And as a use case that comes to my mind would be for example you have a site that you are monitoring and regularly new data is coming in that you measure from your data loggers. You need to convert this data from data loggers into me files for the PVC simulations and this could be fully automatic with PVC CLI for example >> and um is there a limitation on the number of simulations with PBC cli per month? I don't think so. Right.

[00:55:52]the the license just allows you to run as many simulations you want as long as your machine is capable of >> it can run only only one machine.

[00:56:02]>> Yeah.

[00:56:03]>> Yeah.

[00:56:06]>> And PVC CLI commands are inbuilt or we need to enter those commands every time when we wish to run this. Can we save the set of commands we use frequently?

[00:56:16]>> Oh, that's a great question. So um the way I personally use it and would recommend to use it is um just to write a script that runs um like a command line subprocess and basically that generates those commands for you.

[00:56:32]However, it is possible if you have a complex um com like um chain of commands that you need to give or a complex chain of options to put that all into an options file and just give PVC CLI the options files directly. But I would say this is kind of um an alternative to writing a script.

[00:57:00]Okay. And I think with that our time is also coming to an end. We've answered many questions. I see that uh you're interested and it was was you know interesting of what we were seeing. I hope you took some some uh interesting information home with you. I thank all the uh presenters Orana and Mikail who were here with me to present this but also all the team in the background who prepared and and this uh this webinar and helped with the technical aspects and is now sitting right there uh reading your your questions and and moderating the chat. So it was a team effort and I would like to thank all of them for that. Thank you for for for joining this and I hope uh you will have success with your PV simulations and hope that you enjoy working with PVC.

[00:57:55]Thank you very much and have a nice day.

[00:57:58]>> Thank you Bruno for your presenting.