Here's what we've been up to lately...
Developments: (Monday, October 16, 2017) There are new tools on the way. The following are in our DEV version, which is available to anyone who wishes to try out new features before the production release (our BraveHearts).
Again, if you wish an early copy of the development version (will become 1.17), let us know. If we can get the help file complete by the end of the month, the new version will go into production on November 1, so look for it.
APC: We are working to make the APC program work as a batch script for automatic processing following a Promod run.
Just noticed: (Thursday, October 5, 2017) If you are using the Flo/Con file tool in PAT 1.16, you have to be careful if you plan to reduce the size of the event (EVE) file with the "Mark non-zero constraints" button. This button is designed to mark all "non-binding" constraints as those that never come within the specified tolerance of their bounds. Here's the problem: It only processes the constraints within the scope, which unless the scope is the entire system, will get rid of far too many constraints as "non-binding". We'll fix this right away, but for now: Set the scope to "All" pools, so that the entire event file will be processed by the "Mark..." button mentioned above.
If you haven't tried this yet, it's a cool way to make the event file smaller so that Promod will be less slow. Be sure to use a conservative tolerance, say 70% or so.
Someone also pointed out that if the flo and con files are in a folder other than the PFF folder, the PAT will fail to find them - even when you point directly to them. Again, this will be fixed right away, but for now, put the flo and con files in the same folder as the PFF.
PAT 1.16 update. (Thursday, September 21, 2017) We found a few bugs in the latest round of training, so the latest 1.16 is on the server now. (Note: the zip archive will say 1.16.0, but the Title Bar on the program will show the sub version number of 1.16.1) Here is the list of fixes:
PAT 1.16 and APC 7.0 Released! (Thursday, September 7, 2017) Check the Message Board for the many new features included in the 1.16 PAT release, and the APC 7.0 release. All known bugs are fixed (including the Promod Area / Company bug mentioned below). The help files are up to date and ready for use. All licensed PAT and APC users can download the new version now. For those who are not yet using the APC, a trial version is also in your FTP folder for use through September. Even if you do not regularly compute APC, take a look at the help file (on the Products page of this website) to see what it can do. It's a real time saver.
The new FTR tool on the APC. (Tuesday, September 5, 2017). The APC can now read the MISO FTR award csv file, and provides easy bus mapping for the source and sink strings. You can compare FTR revenue against generation congestion charges to make sure your unit portfolio is sufficiently hedged against anticipated congestion from the Promod model. Currently the code is in the DEV version (let us know if you wish to try it out), and will go into production soon.
New Tool in the PAT. (Tuesday, September 5, 2017). On the Power flow tool, look for a new tab labeled "Flows", where you can view the "change" in flow on all branches due to any of the contingencies in the event file. The new feature is currently in the PAT DEV version (let us know if you wish to try it out), and will go into production soon.
Bug Alert! (Tuesday, September 5, 2017). We've found a long-standing bug in the PAT, but it's a rare one. If your Promod companies and Promod areas do not have the same numbers (see the SMA2 table in the PFF), then the PAT will get seriously confused in the load mapping. I believe we've fixed the problem in the DEV version (1.16). I decided not to repair version 1.15 (the current production version) because the situation is very rare. To make sure, check the SMA2 table of the PFF and verify that the area/company numbers are the same. If you need the repair, either download the DEV version or let us know and we will get you a 1.15 with the repair.
Big change in the APC! - (Saturday, August 26, 2017)
We have taken the first step in making the APC able to read an FTR portfolio and compute FTR revenue vs. Company congestion charges. The company congestion charge is the LMP spread between the generators and the load hub (for a specified company) multiplied by the generator's hourly dispatch. The APC computes this charge from the hourly LMP in the BUS file and the hourly dispatch in the UNT file. The FTR portfolio is read from a simple CSV file that lists source/sink/mw/type for each FTR. It's easy to use and very fast.
Bug Alert! (Monday, August 21, 2017)
In version 1.15 and the DEV version, there is an error occurring when an interface appears more than once in an event file. This usually happens when an interface is constrained in the base case and also in a contingency case. The bug causes the 2nd occurrence to point to the wrong interface. If the intended interface should be binding, then the solution will be affected, otherwise not.
The repair to 1.15 is on the server for everyone. The "DEV" version still has the bug, so if anyone needs a new DEV, let Rob know, and I'll put one up for you.
Meeting presentations. On August 16 & 17, Rob and Steve attended the MISO Promod meeting in Minneapolis, hosted by Xcel Energy. As usual, ABB did a superb job organizing the content and the activities. You can view our presentation on the APC here, and on the PAT here.
Regarding our development policy...
Recent PAT developments in the DEV version (to become 1.16)...
Recent APC developments:
Version 6.0 is out there. (August 22)
Here's what's new in 6.0
- New Monte-Carlo Editor for Generation. In the old PAT, we had an editor for modifying the outage library (binary .lib file - the one everybody loves). It was OK, but clumsy and difficult to use, especially to see the big picture. In the new PAT, we are about to release a new way. On the "Gen Outage" tool, where the PAT displays the forced and planned (maintenance) outages, a new button will save a special CSV file with a very simple format. The unit names go across the columns in the first row, followed by the total number of hours out for each unit over the course of the year. The rows represent the 8760 (or there about) hours in the year. Each cell in the spreadsheet is either blank (in service), or has the word "out" (for out of service - really). To modify any outage, just edit the cells to place the outages wherever you like. Note: Only the units that are in the scope and checked in the box on the left side will appear in the CSV. The same file can then be read back in (another button) to modify the outage schedules for the units (and only the units) in the CSV file. All other units that are not in the CSV will be copied to the new LIB, unaffected. Yet another button will write out the LIB file for Promod. We think this will be a much easier way to visualize and modify any forced generator outages. (Thanks Charles - great idea).
- The new feature to save events into CSV templates for porting to PowerBase seems to be working well. I recently changed the default filename to include the "_1", to fix an annoying inconvenience.
- If you're writing bus templates for PowerBase, you may notice that the PAT does not write the area name. This is because the "Long" area name is required, which does not appear in the PFF (surprise!). If anyone can think of a way to get the long name (like a special Promod file), we will consider it.
- The event warning that appears in the light-red warning box at the top of the program was falsely reporting duplicate interfaces when an interface appears both in the base case and in an event. Fixed.
- The errors noted below are all repaired.
Again, if you wish an early copy of the development version (will become 1.17), let us know. If we can get the help file complete by the end of the month, the new version will go into production on November 1, so look for it.
APC: We are working to make the APC program work as a batch script for automatic processing following a Promod run.
Just noticed: (Thursday, October 5, 2017) If you are using the Flo/Con file tool in PAT 1.16, you have to be careful if you plan to reduce the size of the event (EVE) file with the "Mark non-zero constraints" button. This button is designed to mark all "non-binding" constraints as those that never come within the specified tolerance of their bounds. Here's the problem: It only processes the constraints within the scope, which unless the scope is the entire system, will get rid of far too many constraints as "non-binding". We'll fix this right away, but for now: Set the scope to "All" pools, so that the entire event file will be processed by the "Mark..." button mentioned above.
If you haven't tried this yet, it's a cool way to make the event file smaller so that Promod will be less slow. Be sure to use a conservative tolerance, say 70% or so.
Someone also pointed out that if the flo and con files are in a folder other than the PFF folder, the PAT will fail to find them - even when you point directly to them. Again, this will be fixed right away, but for now, put the flo and con files in the same folder as the PFF.
PAT 1.16 update. (Thursday, September 21, 2017) We found a few bugs in the latest round of training, so the latest 1.16 is on the server now. (Note: the zip archive will say 1.16.0, but the Title Bar on the program will show the sub version number of 1.16.1) Here is the list of fixes:
- The LODF constraint builder (a tool that builds base-case & n-1 constraints around a list of buses) was not properly including the base case branches when saving events. The event builder also now ignores any potential monitored lines that have an LODF of -100%, which means the resulting monitored flow goes to zero MW.
- In the PDTR island tool, the PAT would not allow saving a csv file that did not previously exist, making it difficult to save the non-islanding transmission maintenance records.
- Also with the PDTR, the search filter was not able to find buses with lower case letters. Now it can.
- In the contingency evaluator, it is no longer necessary to first solve the SCED LP before computing contingency flows for event-building. After new events are added by the evaluator, all you need do is press "solve" again on the evaluator to see if the dispatch corrects the overloads. No need anymore to go back to the LP. The same is true in multi-hour calculations. Just create or read the contingencies, go to the evaluator and press "solve". Repeat if necessary.
- A caution, though: it is always wise to examine the LP's behavior with the current set of events to make sure the case is well-behaved. No sense in adding more constraints to an already goofed-up system.
- When saving PowerBase event templates on the event tool, my official tester found a few problems that prevented a clean upload to PowerBase with the new events from the contingency evaluator. All fixed now. (thanks Rob C.)
- If you are using outage files with an otg extension, the PAT can now read these files - just drag them in together into PAT's main window after solving an hour. The same is true for con contingency files.
PAT 1.16 and APC 7.0 Released! (Thursday, September 7, 2017) Check the Message Board for the many new features included in the 1.16 PAT release, and the APC 7.0 release. All known bugs are fixed (including the Promod Area / Company bug mentioned below). The help files are up to date and ready for use. All licensed PAT and APC users can download the new version now. For those who are not yet using the APC, a trial version is also in your FTP folder for use through September. Even if you do not regularly compute APC, take a look at the help file (on the Products page of this website) to see what it can do. It's a real time saver.
The new FTR tool on the APC. (Tuesday, September 5, 2017). The APC can now read the MISO FTR award csv file, and provides easy bus mapping for the source and sink strings. You can compare FTR revenue against generation congestion charges to make sure your unit portfolio is sufficiently hedged against anticipated congestion from the Promod model. Currently the code is in the DEV version (let us know if you wish to try it out), and will go into production soon.
New Tool in the PAT. (Tuesday, September 5, 2017). On the Power flow tool, look for a new tab labeled "Flows", where you can view the "change" in flow on all branches due to any of the contingencies in the event file. The new feature is currently in the PAT DEV version (let us know if you wish to try it out), and will go into production soon.
Bug Alert! (Tuesday, September 5, 2017). We've found a long-standing bug in the PAT, but it's a rare one. If your Promod companies and Promod areas do not have the same numbers (see the SMA2 table in the PFF), then the PAT will get seriously confused in the load mapping. I believe we've fixed the problem in the DEV version (1.16). I decided not to repair version 1.15 (the current production version) because the situation is very rare. To make sure, check the SMA2 table of the PFF and verify that the area/company numbers are the same. If you need the repair, either download the DEV version or let us know and we will get you a 1.15 with the repair.
Big change in the APC! - (Saturday, August 26, 2017)
We have taken the first step in making the APC able to read an FTR portfolio and compute FTR revenue vs. Company congestion charges. The company congestion charge is the LMP spread between the generators and the load hub (for a specified company) multiplied by the generator's hourly dispatch. The APC computes this charge from the hourly LMP in the BUS file and the hourly dispatch in the UNT file. The FTR portfolio is read from a simple CSV file that lists source/sink/mw/type for each FTR. It's easy to use and very fast.
- The single-case version is complete and undergoing verification now.
- The multi-case version will follow soon, so you can compare congestion charges and FTR revenue changes among cases. Should be a huge time saver and an excellent tool to help judge the value of comparative cases.
Bug Alert! (Monday, August 21, 2017)
In version 1.15 and the DEV version, there is an error occurring when an interface appears more than once in an event file. This usually happens when an interface is constrained in the base case and also in a contingency case. The bug causes the 2nd occurrence to point to the wrong interface. If the intended interface should be binding, then the solution will be affected, otherwise not.
The repair to 1.15 is on the server for everyone. The "DEV" version still has the bug, so if anyone needs a new DEV, let Rob know, and I'll put one up for you.
Meeting presentations. On August 16 & 17, Rob and Steve attended the MISO Promod meeting in Minneapolis, hosted by Xcel Energy. As usual, ABB did a superb job organizing the content and the activities. You can view our presentation on the APC here, and on the PAT here.
Regarding our development policy...
- We get our development ideas from you, the user.
- To send us an idea for development, register on the message board and post to the development thread, or create a new thread if you think the new idea might spawn a lot of discussion.
- We will respond to the post and let you know if (really, when) the suggestion will be implemented.
- In a few days, when we've taken our initial attempt to implement your idea, we will post a "DEV" version of the software for you to download and test. When you give the OK, we will roll the result into the production version.
Recent PAT developments in the DEV version (to become 1.16)...
- The Event Output tab now creates CSV Templates for uploading new PAT events to PowerBase.
- There's a new Event diagnostic to identify Promod events that have conflicting names when truncated to 21 characters (this isn't a problem with PAT-generated event files, because each event name is prefixed with a special string (!E:xx, where xx is the event number), so each event will have a unique name.
- Event compare tool - you can read a 2nd event file and examine the differences, and merge. You can easily merge events with common contingencies, even with single event files.
- The scope tool now considers power flow areas along with pools and companies.
- New PDTR islanding tool. Use this to search and repair transmission maintenance records that cause islands.
- Coming soon: New information in the HRX file that will allow you to compute all LMP very, very fast for all hours, without solving the hourly SCED problems. This will be HUGE when ABB releases the version that sends the new data.
- We've received many comments recently about the need for training videos. We will be working to add selected topics to the website soon.
Recent APC developments:
Version 6.0 is out there. (August 22)
Here's what's new in 6.0
- LMP on and off-peak and RTC averaging by company.
- LMP spread calculations between any two points (hubs, zones, buses in the BUS file).
- Fuel burn in mmBTU.
- SPP method up and running.
- Help file re-written and all up to date.
- It's on the DEV server. If you're licensed with the APC, it should also be in your folder.
