Speed Issues with versions 6.2+ ?

[Iain_Kelly]

Hi there,

Until recently I've been working mostly with the 6.1.7 and 6.1.9 versions of PVsyst, but have upgraded to 6.2.2 to take advantage of the newer features (being able to specify PV fields in terms of number of modules in the 3D modelling view is particularly useful, also the resolution of the report-printout bug, thanks for that).

However, I'm experiencing quite serious slow-down when working on larger projects with the new version- this also applies to 6.2, which I tried a month back or so. Even viewing a tab such as 'System' or 'Horizon' and then returning to the main screen with 'cancel' (which shouldn't require any processing) causes a freeze of about 30s. This same slowdown occurs with smaller projects, although to a lesser extent.

Is there a particular reason for all the extra processing? I want to be using the most up-to-date version of the software, but having to cope with the delays is a major issue.

(I'm using a not-particularly-powerful acer laptop to run the software. However, it coped with pre-6.2 versions with relative ease.)

Iain

*EDIT* It's possible what I'm seeing is caused by having created the project (along with a large 3d model) in an earlier version. A new project file I'm working on today isn't experiencing the same problem. Does this sound about right?

[Bruno Wittmer]

With PVsyst V6.25 the handling of large and complex 3D scenes has been improved.

The loading and saving of the different variants has been made faster.

The editing of the 3D scene is much smoother.

This makes the handling of large projects now much easier.

[denizurla]

Dear Bruno, I find the new functions of the PVsyst V6.25 very good and also the speed of swithcing interfaces is increased considerably in comparison to V6.2+.

However, I still find the simulation process much slower in comparison to V6.19. I am working with very large Helios3D files. I should say that the "optimized shading calculation" helps to create the shading table much faster but when I move to the simulation, it feels like the PVsyst is running again a shading table calculation, which takes too long (5-6 hours for a project with around 1500 objects)

I tested the same simulation without the Shading Scene and the simulation took just couple of minutes. (Projects have central inverters)

For the simulations, I set the parameters in the Near Shading definition window as follows: Use in Simualtion --> "According to module strings" and Calculation mode "Fast (Table)"

Why do you think that the simulation takes too long even though I created the shading table beforehand?

I would appreciate your feedback.

[André Mermoud]

It should not recalculate the table if this has already been calculated and nothing has been changed in the shading scene.

I will check in the program.

[denizurla]

Hello Andre,

Thank you very much for your answer. Were you able to check if the simulation recalculates the shading table? I saw that you have a new update V6.26 and I read in the "Software Development", you have a development called "Big shading systems: optimized a time-consuming verification.". I didn't update my V2.25 but will the update make simulation with big shading scenes faster?

Regards,

Deniz

[denizurla]

Hello,

Unfortunately installation of V6.26 didnt solve my simulation speed problem as I was expected to.

Does anyone else have the same problem with large shading scenes? I have uploaded a picture of where I am having difficulty. Shading table creation takes around 45 minutes, which is reasnable but "Hourly Simulation Progress" takes around 5-6 hours for a system size of 20MW with 10 central inverters. Shading scene includes around 2000 objects (module tables and trees).

I would appreciate any feedback on this issue.

Regards,

Deniz

slow simulation process

[Fredy]

Hello

I've just updated my PVsyst from 5 to 6

I have exactly the same problem

i work with large scale palnts and on that screen "Hourly simulation Progress" it is very slow

Is there any option to simulate faster?

Or is it a problem from the program?

Look forward for your feedback.

[Ninja]

On Monday I updated to 6.28 since that time it isn't possible to run a simulation.

Today I updated to 6.29 but there's no use.

When I start the simulation of a 21MW plant the software needs half an hour for calculating the january and than it always stops because of an error.

How could it be that there isn't enough memory if my camputer has 8GB RAM?

I think there is be a bug in the software!

I look forward for your feedback.

[Fredy]

Any news regarding this problem?

I've updated to 6.30 and on this part like "Denizurla" referred:

The simulation it’s too slow it takes like the same time as the Near shading table calculation, or more..

Is it normal to take all that time?

Previously it was like “instant” in PVsyst5 let’s say 1 minute in the worst cases.

Why does it takes hours now?

[André Mermoud]

If you have a big system (big shading scene), please check that

- You are not using the calculation mode "Slow", which calculates the shading factor at each simulation step.

- You are not using the "Module Layout" option, which is not suited for big systems (see Big systems calculation in our FAQ)

Please also check that in the 3D editor, the menu option "Tools" / "Optimized calculation" is checked.

The simulation may also be slow if you have a very big number of inverters.

[Fredy]

Dear André

I'm using all that options:

Near Shading Calculation mode -> Fast (table)

Not using "Module Layout"

On 3d editor i have that option checked "Tools" / "Optimized calculation"

My system is about 896 objects tables (on 3d model), and 8 central inverters.

it takes almost the same time as the Near shading table.

This is stressfull when you have to do several simulations where you change one loss or one other parameter you have to wait all that time again.

do you know what can be causing this?

Best regards

[André Mermoud]

I have just found a routine which is executed very often for a same calculation during the simulation. I never suspected this routine as it is a very simple one.

This expecially affects the simulation of systems where the numerous tables are not grouped into arrays ("fields"), like in the Helios3D scenes.

Correcting this substantially reduces the execution time of the simulation when using the Shading factor table.

This correction will be included in the next version 6.31, to be released soon.

[denizurla]

Hello André,

This is great news and I am really happy that you looked into this issue. When should we expect a new release 6.31 with the correction?

Best Regards,

Deniz