kjs55

P.S. Please note the 5xxxx3 MWh value also shows up in the Report on the Loss Tree, in the "Balances and main results" table, etc.

PVsyst v7: Sum over E_Grid 8760 is 5xxxx2.3 MWh, which rounds to 5xxxx2 MWh. NB: 8760 has night losses. PVsyst Loss Tree (copied to Clipboard and pasted into Excel) yields 5xxxx3 MWh (no decimal places provided). If you remove one row from the 8760 summation, you get 5xxxx2.5 MWh, which rounds to 5xxxx3 MWh. So this tells me the PVsyst Loss Tree is missing a single row from its summation (namely, it's summing 8759 rows, not all 8760 rows). Please confirm it. (NB: Exact values masked for confidentiality reasons.)

PVsyst v7 & v8: In the PVsyst Report (output PDF file), please note the ambient temperature of all inverter power ratings. For example, please include: Unit Nom. Power (~>50 oC) = 765 kWac where the "at 50 oC" is added. Similar to: Max. power (~> 25 degC) = 840 kWac Please see attached screenshot. Thank you!

Open PVsyst v8.0.8 / demo projects within it, get this pop-up: Please advise.

Sorry, I don't have time to do a detailed bug report. If you cannot reproduce this, please add a comment and I'll add more details. v7 & v8: Activate Power Factor on PF tab of Energy Management. Then go to next tab Grid Limit and activate w/ Apparent Power basis. You'll see the units of Grid Limit change from MW to MVA. Now return to PF tab and deactivate it. Now return to Grid Limit tab. Units of Grid Limit remain as MVA instead of changing back to MW.

I also noticed that the value entered into Grid Limit doesn't round, it truncates. Why? And the allowable significant figures is inconsistent. You can either enter 0.17 kW with max. two (2) sig figs or 999.17 kW with max. five (5) sig figs. Why? The relative (%) impact of the truncation will of course depend on the value entered. It seems like an easy enough fix to allow a fixed number of sig figs, e.g., several (more is better, e.g., so that we can exactly match utility contracts which can be defined as irrational numbers). It also appears that the units go from kW to MW depending on the project's power capacity, so ideally the increased number of allowable sig figs could also be consistent regardless of the units. Thanks.

@S.Faulkner: See PVsyst Release Notes for v8.0.0 ("Corrections"): https://www.pvsyst.com/help/release-notes/index.html#exec-1--corrections "Simulation: the bifacial mismatch loss is now proportional to the Bifaciality factor"

@S.Faulkner: See PVsyst Release Notes for v8.0.0 ("Corrections"): https://www.pvsyst.com/help/release-notes/index.html#exec-1--corrections "Simulation: the bifacial mismatch loss is now proportional to the Bifaciality factor"

bifiMMF = Bifacial Mismatch Loss Factor bifiCoef = PV module bifaciality coefficient ## v8 Help Menu: https://www.pvsyst.com/help/project-design/bifacial-systems/bifacial-systems-results.html v7 Help Menu: https://www.pvsyst.com/help-pvsyst7/index.html?bifacial_results.htm Suggestion #1: I suggest updating the v8 Help Menu loss tree to represent a PVsyst v8 loss tree. Suggestion #2: For this (v7 & v8) Help Menu example, I suggest reporting the assumption used for Bifacial Mismatch Loss Factor, for the benefit of the PVsyst user. Comment #1: It is difficult to reproduce the "Mismatch for back irradiance" value of -2.6% on the loss tree for this example without knowing the value from Suggestion #2. My best guess is the value of bifiMMF must be 16% when using PVsyst v7. NB: I don't think the example is accurate for the updated v8 calculation method. Comment #2: In general, I think this loss value was calculated incorrectly in v7, as = (bifiMMF*GlobBak)/(GlobBak + GlobEff). I say "incorrectly" b/c a statement on the v7 Help Menu page says, "NB: Here, the GlobBakEn is GlobBak * Bifaciality factor." I don't think bifiCoef is used in v7 (I'm running v7.2). Suggestion #3: If I'm correct, I would suggest fixing this comment re bifiCoef in the v7 Help Menu. Comment #3: I think bifiCoef is now used in v8 as (accurately) described in the v8 Help Menu. The difference in results between v7 (no bifiCoef) and v8 (bifiCoef) can be significant, esp. for large values of bifiMMF. Please confirm it.

Hello, Would you please point me to the PVsyst v8 Help Documentation for how to apply the subhourly clipping loss when all you have is hourly meteorological (aka solar resource) data for a given site (which is almost always the case)? I was just reading this PVsyst Help Documentation <https://www.pvsyst.com/help/physical-models-used/grid-inverter/subhourly-clipping-correction.html>, and I noticed this sentence: "The only prerequisite to apply the model is to have a MET file that has been generated using sub-hourly irradiance data." Are you aware that this "only" prerequisite is almost never available, in the standard PVsyst modeling use case? Therefore, I think I must be missing or misreading something. I'm hoping you can please explain this to me here and point me to the relevant PVsyst Help Documentation that I'm presently unable to easily find. Thank you.

Spatial Smoothing is the newly released supplement to Triple-C. CCC-modeled clipping losses can be reduced by 1% per annum with Spatial Smoothing with a sufficiently large PV array size. See proceedings of Sandia PVPMC 2024: Spatial Smoothing Reduces PV Clipping! by Tim Townsend <https://pvpmc.sandia.gov/download/7908>. Please implement CCC+SS immediately in PVsyst.

P.S. I forgot to include the title of the newly published CCC white paper by Townsend and Sauer: Full title: Triple-C: Clouds, Capacity, and Clipping A Method to Correct Traditional Hourly-Based PV Simulations to Account for Subhourly Clipping Loss Short title: Triple-C: A Subhourly Clipping Correction for PV Modeling

The Triple-C ("CCC") white paper is now available at the following URL: https://www.linkedin.com/posts/tim-townsend-93a4a711_triple-c-a-subhourly-clipping-correction-activity-7138341773319331840-kNWO/ The authors, and many other major PV industry stakeholders, would like PVsyst SA to please directly incorporate this specific modeling option into the PVsyst software program as soon as possible. Thank you.

My understanding is that a fixed var output for all timestamps would not require any grid data. Is this not the case?

I was reviewing Slide 115 of the presentation linked in Ref. [1]. I'm wondering which of the following four (4) operational modes listed on that slide (pasted below for convenience) can be modeled in PVsyst? Thanks. A couple other general, related Qs: 1. Is it possible to model fixed var operation (mode) in PVsyst? 2. Is it possible to model fixed var operation (mode) in PVsyst for BESS-coupled PV systems? (i.e., when modeling both the BESS and PV system in PVsyst) BESS = Battery energy storage system [1] https://www.nerc.com/comm/PC/IRPTF Workshops/IRPTF_Workshop_Presentations.pdf Excerpt from Slide 115 of Ref. [1]: 1. Constant power factor or VArset-point Configure inverter with fixed Varoutput. 2. Controlled power factor or VArvia comm’s) Send Varsetpointcommands to inverter via Modbus/TCP 3. Autonomous VArcontrol (depending on V) VoltVArcontrol based on curve 4. Autonomous power factor control depending on real power output VArsas a function of Watts

PVsyst v7.4.2: I'm generally curious why the units of energy are listed as kW in the PVsyst 8760 CSV output file rather than kWh. The below link [1] lists it as kWh. It seems the units of apparent and reactive power are listed as kVAh. However, a client wrote to me and asked if the units of EReGrid ought to be kVARh rather than kVAh. Please advise. Thanks. [1] https://www.pvsyst.com/help/index.html?power_factor.htm

A couple more observations in addition to what I listed above: - Change E_GridApp to EApGrid [1] - Change E_Regrid to E_ReGrid [2] [In general, I'm surprised these modeled data variable names, which appear in the PVsyst 8760 CSV output file, don't follow the existing, preceding "E_Grid" naming convention (e.g., "E_GridApp", "E_GridRct", "E_GridLim"), but O-K.] Thanks. [1] https://www.pvsyst.com/help/index.html?power_factor.htm [2] https://www.pvsyst.com/help/simulation_variables_grid.htm

PVsyst v7.4.2: I suggest adding a power factor PF column to the PVsyst 8760 CSV output file. It would save us a bit of time from having to manually calculate it from the existing columns (aka modeled data variables) provided. Thanks.

PVsyst v7.4.2: Today I also noticed the mismatch between the summed E_Grid column in the 8760 CSV output file and the E_Grid values listed in the PDF report output file. They don't round to the same value. Sure, it's not significant. But I also don't understand why there's any mismatch. (And it's one more detail to have to retain in one's memory and explain to people.) I think pointing out such mismatches is useful in general, because it can sometimes (potentially) indicate other issues in the codebase. Thanks.

Running PVsyst v7.4.2 Generate 8760 CSV output file using default settings. Open 8760 CSV output file in Microsoft MS Excel. Highlight first column (Column A). Data tab > Text to columns > Delimited > Next > Semicolon > Finish Warning: There's already data here. Do you want to replace it? Because there's some data in columns B and C. I don't think there should be data in any column except for Column A. Thanks.

No problem! Thank you! I just happened to notice "Syst ON" at the same link above -> "Syst_ON" Thanks again!

P.S. I forgot to include the title of the poster (PVPMC link above) which I'm adding now partly for my own future copy/paste benefit: The Triple-C Method for Correctly Simulating PV Clipping Loss by Townsend & Sauer et al.

Slight editorial suggestion to add the underscore "_" to the IL variable names listed in the PVsyst Help Menu, e.g., here: https://www.pvsyst.com/help/simulation_variables_grid.htm In all cases where PVsyst variable names are listed in the Help Menu, I hope the variable names are listed in the exact same syntax (spelling, capitalization) as what's used in PVsyst (e.g., what's visible to the PVsyst user in the software program, in the output files, etc.). Thanks.

I watched your recent YouTube tutorial on fitting the PVsyst one-diode model to measurement data (measured I-V curve data as a function of irradiance and temperature) [1] which said something to the effect of optimizing all five (5) PVsyst one-diode model parameters simultaneously won't give good results and so PVsyst doesn't advise this. I'm wondering if you have a citation for that (?). Ref. [3] of my previous post above, which is available to download for free [2] and which builds off the foundational Ref. [2] of the above post, suggests the opposite conclusion in the event, of course, of having well-calibrated data from a reputable laboratory and following the procedures of handling that data as described/prescribed in the paper (self-reference method, deviation in efficiency relative to STC, etc.). At the same time, of course, I'd appreciate a reply to my above post, thanks. [1] PVsyst 7 _ Series resistance determination, https://www.youtube.com/watch?v=5MJo_9dqbBk/ [2] https://www.osti.gov/pages/biblio/1184501/

I spoke to Tim Townsend who historically provided PVsyst with the recommended Uc and Uv values of 25 and 1.2 (which PVsyst attributes to PVUSA) derived during his time working at that same location in Davis, CA where PVUSA-related testing activities also took place. He confirmed to me that the height of the wind speed data used to derive those values via regression was 10 meters. (He said: “The pole was next to my office and plainly visible out the west windows of our building.”) I think this (the 10 meter height) is a critical piece of information to specify in the relevant section of the PVsyst Help Menu [1], so this is fundamentally my suggestion (to add this single bit of information to the help menu) and the reason for this PVsyst Forum post. Note that since NSRDB PSM v3, that particular weather data set suddenly changed the wind speed height to 2 meters (PSM v2 and prior NSRDB data sets used the 10-meter height which is more standard aka traditional for TMY aka TGY files given that the particular ground surface coverage doesn't significantly impact the wind speed at that higher up height). Unfortunately, I can’t find any formal documentation of this change from 10 m to 2 m apart from this NSRDB webinar's Q&A [2] and a YouTube video (also in the Q&A of a NSRDB webinar) [3]. In such use cases (of 2-meter wind speed data), I don’t think the thermal parameters that were derived using 10-meter wind speed height would apply. I’m pointing this out because many PVsyst users are using {Uc=25, Uv=1.2} regardless of the weather data set in use (i.e., incl. NSRDB PSM v3), and my feeling is that's not correct. I suppose if the 2-m wind speed data was scaled to a 10-m height, that's one way to continue to apply those same thermal parameters historically provided to PVsyst by T. Townsend. I'm not aware of anyone who's doing this, though, and my understanding is this wind speed height translation is not happening internal to PVsyst (in the PVsyst algorithm) in such use cases of 2-m wind speed height data in the TMY aka TGY weather (meteorological) data set. Either way, the situation (state of affairs) needs to be clear to the PVsyst user which is why I'd like to see it stated in the PVsyst Help Menu that the values provided by T. Townsend are specific to a 10-m wind speed height (and then a description of any such wind speed height translations, e.g., to a standardized height of 10 meters if that's to occur within PVsyst OR the lack thereof of any such translations thus leaving it up to the user what to do with the wind speed height external to PVsyst, prior to importing the TMY aka TGY data, for the purpose of applying/pairing with a specific set of {Uc, Uv} within PVsyst). [1] https://www.pvsyst.com/help/thermal_loss.htm [2] https://webcache.googleusercontent.com/search?q=cache:dqpgedl3nbgJ:https://nsrdb.nrel.gov/nsrdb/NSRDB_Webinar_QA_-_Oct_6_2020.pdf [3] https://www.youtube.com/watch?v=BzNoYACXCOY&t=4220s