 Hello again, everyone. Thank you for joining us today for a virtual meeting on the Extremely Low Probability of Rupture, or XLPR, Probabilistic Fraction Mechanics Code on Setting Up the Inputs. My name is Matthew Homiak, an NRC's lead from the Office of Nuclear Regulatory Research for the XLPR program. XLPR has been a joint venture with the Electropower Research Institute, or EPRI, and my counterpart from EPRI, Craig Harrington, is joining us again today as well. Supporting the meeting is the NRC staff, and the EPRI staff, and some of our contractors. The purpose of the meeting is to help new users to XLPR, get started by reviewing how to set up the inputs to the code. And to that end, I'd like to welcome all of our new users. We have several now. I'm pleased to report. So, XLPR version 2.1 was publicly released in early June. This is just a recap of the request process in case you haven't completed that yet or are interested in completing it. There's no fees for getting the code, but prospective users need to satisfy certain citizenship requirements and signs and user license agreement. And users can initiate the request through EPRI.com. We've partnered with EPRI to do the distribution, and then you have the product ID on the screen here. You can just search on that on EPRI's website, and it will come up, and you can start the process from there. And then after your request gets reviewed, we'll be provided with a copy of the code and supporting materials for electronic download. So, our meeting today is actually part of a series of meetings that we've been having. The last was held on June 3rd, and I believe many of you may have attended that. We have the summary of that meeting and also the video recording on YouTube available now, and that should be shared with the chat. And also, we have firm dates now for the rest of the sessions, July 29th and August 5th, sent out an announcement. So, please mark your calendars for those. I think the announcement had the links in there for the Webex as well. We've had a slight format change where we combined running the simulation and retrieving results into one session. That's the third one. And that opened this up for a different topic for the fourth session and final session on advanced methods. So, I think that was a good change on our part, and it will provide some more information to you all. So, here's our agenda here for today. We're currently in the introduction and opening remarks portion. Next topic will be resources for inputs, and we'll review the input set. Talk about some input uncertainties, review the inputs group report, we'll review the SIM editor. We've got a couple of demonstrations for you. We'll be demonstrating some of the challenge problems that are in your release package. And we'll set up the inputs today, and in the next session, we'll actually go through running those input sets. We'll have a break after that, and we're going to use that time to caucus the presentation team here and maybe do some other demonstrations or something like that to help answer any questions that come in. So, we've got a decent 30 minutes or so slated for questions and answers, and then we'll conclude the meeting. And I'll use the remainder of the introduction and opening remarks portion today to discover some more administrative items and also show you how to set up your software. So, this is an NRC category three public meeting again. It means the public is invited to participate by providing comments and asking questions throughout. We're using the Webex platform again to deliver the meeting, and I would encourage everyone again to participate through Webex. So, you can see the slide presentations and the demos. This is also how we primarily plan to take your questions. So, you're going to get the best experience using Webex. You can submit your questions at any time using the Q&A feature. So, you want to make sure that's shown on your screen. It may look slightly differently depending on whether you're using Webex in your internet browser or through your desktop client. For the web browser, you can access the Q&A by clicking on the question mark bubble. Located between the speech bubble and the three dots. And then for the web browser, I'm sorry, for the desktop client, you click on the three dots and then you can select the Q&A option here. That's shown on the right. Either way you do it, you should display a Q&A box on the right side of your screen. We use this back at the June 3rd meeting, and I think it worked pretty well. After the meeting concludes, we'll wish you a summary, which will be made available on NRC's public website. And like with all our other webinars, we'll also be recording this meeting for later viewing, and we'll make that available on youtube.com and share the link. I'll actually be doing most of the presenting today. Before I was leading the XLPR effort from the NRC side, I was also the lead of the inputs group for version two development. And assisting today with some of the presentations, we have Nathan Glunt from the EPRI staff. And then backing us up are our Q&A team. That includes Craig, of course. And again, we have Marcus Burkhart from Dominion Engineering under contract EPRI. Marcus did most of the presenting at the June 3rd meeting, and he was also heavily involved in the inputs group efforts. We also have Cedric Salabari from Engineering Canucks Corporation of Columbus. He's our resident mathematician under contract NRC. And we also have Marj Eriksen from Phoenix, Engineering Associates under contract EPRI. Marj was the lead of the models group development for XLPR version two. And Giovanni Facco from the NRC staff is helping us out today. He's a Webex host. If you're having any difficulties with your Webex or your audio setup, just feel free to reach out to him directly through the chat feature. I was adding the stuff from my head this morning, and I think that this whole team here is bringing about 50 years worth of combined experience with XLPR. That's quite a bit. Craig, did that sound about right to you? Sounds about right. Did you want to add anything, Craig? Well, just just real briefly, Matt, I appreciate everyone taking the time to join this session this morning. I know many of you have completed the process to download XLPR. Some of you have not yet. I know there have been some issues working through questions and issues with the end user license agreement. And we are making progress on that. This XLPR is a very unique beast in the way it was developed and the way we are now distributing it through EPRI. So that license agreement has some unique features because of our relationship with the NRC in distributing this code. And so we've had a few startup challenges and some different questions. We are working through those as quickly as we can. And please bear with us. We should have all of you through that process, I think, very soon. If you do have issues though, feel free to reach out to me directly either through my regular email address or the XLPR at EPRI.com. And I'll try to help however I can. So thank you for joining. All right. Thank you, Craig. And thanks, everyone, for being here today. Remember to use the Q&A feature, of course, to submit your questions. Okay. Now I'd just like to show you how to get set up with your code. Maybe you've just downloaded it and haven't done this yet or maybe your request is still pending. But I'll just review the process quickly. So you're provided with the code as a zip file like this. Here I have it on my desktop, but you can have it in any directory. The first thing you want to do is extract the contents. You can use any program that handles zip files to do that. Now, XLPR requires a certain folder structure for proper operation, which is explained in the user manual. So to maintain the required folder structure, I'm just going to want to extract the contents of the zip file into a single folder. I'm not going to show that here. But once you do that, you'll have everything in a folder here. And now this is important. And I highly recommend that you retain that zip file. At this point, if you end up changing any of the default inputs or anything like that, there's not really going to be a way for you to restore those, other than by going back to the originals that were provided. So don't delete your zip file. Keeping it will make sure you always have a pristine copy of the code to go back to if you need to. This is also a good time to install the SIM editor that's provided in the release package in the SIM editor folder. And you'll just want to run this install file right here. You should also install either the Gold Tim Pro or Gold Tim Player if you haven't already. Those are available from goldtim.com. You want to go to products, Gold Tim Player, go to the free download. And we're on version 11.1.7. That's an earlier version because that's what we've tested under our QA program. Let's click on previous versions and then find the latest right here. Download it. Doesn't take all that long. And then you just want to run that install file on your machine. Okay, so we'll be providing hints and tips throughout today. And on your setup, here's some tips there. Like I said, keep copy of your zip file. And make sure to maintain that original folder structure. Okay, so just a bit here. Another administrative item on your maintenance request. XLPR code is supported by an active maintenance program. And users can submit their problem reports and also any requests for new features or enhancements. And those should both be sent to xlprnrc.gov and xlprepri.com. Like Craig and I can get them and process them accordingly. And section 5.2 of the user manual has the submission guidance and format. Also some further information on how you would do that. Okay, on to our next topic here. Resources for inputs. And this is starting the formal portion of our agenda. So here's some key references for inputs. Several of them are in your release package that includes the user manual for the xlpr, as well as the sim editor user's guide. And then there's also the training manual for the introduction training. We'll talk a little bit more about that. But I highly recommend that you do take that entire training. We'll be basically summarizing a bit of that today. And then we've mentioned some other technical reports, which we will be putting out soon. Just some administrative delay there, but those are listed here. And include the framework group report, the inputs group report, and as well as the subgroup report on the welding residual stresses. So these are all good references to have for inputs to the code. How to do them and actually, you know, what they are. Okay, now another resource that's available to you in the release package is actually databases of inputs. These include the weld types that you provided with three example input sets there for the property sheet for Westinghouse recta pressure vessel outlet nozzle, Babcock and Wilcox recta coolant pump inlet nozzle, and Westinghouse steam generator inlet nozzle. There's another database of materials. You have three different carbon and low alloy steels and three stainless steels that you can use on the left type or right pipe sheets. And you also have inputs for alloy 82, 182, and alloy 52 and Web 52, which you can use on any of the sheets actually. The other database that's available is the welding residual stresses, both for in the axial and hoop directions. And those have a number of profiles. For each of the three weld types, you have an unmitigated profile at three different repair depths, 0%, 15%, 50% depths. And the same goes forward for overlay, mzip, and inlay. So that's 12 profiles for each, the 36 weld residual stress profiles you provided with. Okay, so let's go ahead and actually review the input set at this time. So that's located in your XLPR main folder here. XLPR tag 2.1 input set. So this is the file that will be read either by the GoldSim model file if you're using GoldSim Pro or the GoldSim Player file if you're using the GoldSim Player when you run a simulation. We'll also be presenting the sim editor today. That's a tool to help you populate the input set. Let's first take a look at the input set file though. So this is an Excel workbook. It's organized into various worksheets, user options, properties, etc. General workflow would be to work from left to right through the sheets and then from top to bottom. So let's start with the first, the user options which we're on already. This is where you define the simulation that you want to perform, schedule things. That's also where all your settings are. At the top here you have a backwards compatibility statement. It's not really useful now but as we get into more maintenance and future release the code this will be helpful. Next is just a text box here where you can put some notes to describe your analysis. This is helpful because you can't actually rename this file. So here you can describe what the input that you've set up. Moving down everything else is organized into a table. The column B here, this is important throughout the input. This is the global ID. Most inputs have a global ID and these are referenced back in GoldSim. These are basically a unique number for the inputs. So you hear a lot about us referencing global IDs for various inputs. Column C here is just a brief description of the input itself. Then you have units for the input if there are any. Not all inputs have units. The actual input value that you would put in and then you have some for their description over here. And then again up at the top here we have a color coding. So you see these different colors throughout the input various inputs and these tell you what those inputs mean. For example this red one here means that you put that arranger invalid if I deleted the first input here. It turns red and you know that tells me that I need to do something there. Okay so let's just go through what's available here on the user options. First are some general options. Basically the simulation time, how many subunits you have and as well as your different crack orientations are here. Next we have some simulation settings for the probabilistic settings. Here's where you define your sample sizes for the valedictorian epistemic loops and some other settings as well. A few of these are just recorded here. Those are noted. You can record them here but you actually need to set them when you go to run your inputs and go to them. Next category are the weld type options. These are more just some miscellaneous settings here. I won't cover those. Next is where you would set your mitigation whether that be non or stress-based like SIP or you know synchrohydrogen mitigation. This is where you would turn those on and also where you might schedule them to occur throughout the simulation. Next is just some options for loads and stresses as well as you can define your different operating periods here. Up to three different operating periods are allowed. And also in this category is where you would set your activate your transients. You do that. There's up to 20 transients and you turn them on here and specify the different type of transients that you would like. Here you have your crack initiation options and as well as if you're using crack initiation models which model you'd like to choose. There's some options here for crack growth as well whether that is fatigue or stress-cursion cracking or both. Next is a few options here for crack coalescence or CERC cracks. And then here we have inspection and leak detection options. Most of these are for inspection or ISI various settings as well as in the 808 series inputs here. This is where you would be scheduling your inspections both before and after mitigation. And then you can specify your detectable leak rate in global ID 810 here. Next there are the leak rate options here. I'm going to really change these. They're more just for reporting purposes. Next COD options and here local options you would just tell the code what your different local thresholds might be so you can set those here. Down at the bottom we have some error handling. We have a leak rate jump output and this is where you can set the thresholds for that. And finally there's a spot here to record flat and hyper keep setting options. Okay so those are your flags and your bells and whistles and stuff like that. We're going to move on to the property sheet next. And this is very similar. I'm going to set up here with the global IDs. I'm going to see that continue. Up at the top here is where you're going to be setting your different materials. That works from the databases which we'll show you. And again we have a similar color coding here as well. This is a little bit different set up here than the user options which are basically just deterministic inputs. Here you're going to be specifying your probabilistic inputs. So you have more inputs to deal with. Here's the name of the input again. Units just like user options data source. This is where you would specify whether the particular input is deterministic or constant as well as you can assign it to either the alliatory or epistemic loop. I can specify important sampling here. If you need to for inputs on the alliatory or epistemic loop. Here's where you would enter a deterministic value if you're using a constant. And then the rest of the cells here are for your distributions. Specify the type here, column I. And then these parameters which we'll get into a little bit later. They're defined on the weld options. They tell you what these are but these will be very depending on what kind of distribution you're using. So everything here on the properties is basically things about the components that are not material related. So you'll have the geometry. Here's where you put in your the flaw sizes and then you have operating conditions. Some are just there for any operating period or you can change the pressure, temperature and oxygen for any of the three operating periods that you're allowed. Moving on later you have loads and stresses. You have a bank of inputs here for earthquake options here as well. And then you can enter either the loads or stresses for each operating period. This is operating period one. Operating period two is next. And same thing for operating period three. At the bottom of that this is where you would specify what loop you're going to put your WRS profiles on. And you'll set those later on some other tabs. And then everything else on this sheet is for ISI inputs. And those basically are divided between a set of inputs. Pre-mitigation and then post-mitigation. And you have depending on what kind of mitigation you're using is where you would put in your inputs. So here's the bank for overlay inputs, AMSIP, inlay. You have some miscellaneous options here at the end. For those you can see here we have reserve for future use. And then finally you have a set of correlations. So some inputs in XLPR can be correlated to each other. This is a rank correlation. So this is a range from negative one to one. Negative one would be a perfectly negative correlation where a sampled high value. One of the inputs would make a sampled low value on the other one. One would be a perfect positive correlation and zero would be no correlation. So the next we have here are four sheets for material properties. Left pipe, right pipe, weld and mitigation. These are all pretty similar. So I'm just going to show you the weld. Very similar setup to the property sheet where you have the basically the same column structure. Except here we're dealing with material properties. The first set of inputs, just general properties. Those are your strength and fracture toughness inputs here. Next we have Tiffany properties. These are used on weld and mitigation materials, but they do not need to be used on the left pipe or right pipe. You'll see those actually grayed out on those other tabs. Crack initiation inputs would be specified here as well. And then you have various categories here for the different models. Weibull models. So all three models here is where you put the inputs for crack initiation. Crack initiation just occurs in the weld component or mitigation component. So you don't need to worry about these inputs on the left and right pipe, although they're there. Next you have crack growth properties for all the different models that are available on XLPR. Fatigue crack growth and PWSCC or SCC crack growth. We have some material flags, which you set here depending on what kind of material you're working with, and then again a set of correlations. The basic overview on how the material property tabs are set up. Four of those. I'll show you the two weld residual stress inputs here. So here we can define a WRS profile at 26 points through the thickness in both the hoop and axial directions. It can be either deterministic, you put your inputs in here, or you can define a normal distribution that's required. Pram 1 is the mean. We'll talk about this a little bit. And pram 2 is the standard deviation for the particular point. And then over here on the right here, these are the deterministic values to show you what that profile looks like. So on each sheet here you have a profile you can use both pre-mitigation and post-mitigation. Okay, next an axial WRS sheet is set up the exact same way. Next we have getting into some more transient inputs here. And here if you have a type 1 transient, you put in your transient pressure, temperature, time, history essentially. You would define time at which the changes occur. And you have a delta t, a delta p. These are the changes from the normal operating conditions. And you would be able to do that for up to all 20 transients. If you wanted to, you could have 20 time point transients, which you would define here. And then we have another sheet for other fatigue inputs here. And if you have a type 1 or just a thermal transient, you're going to be inputting some of the scheduling information here. You know, when the period of this relation, when it's active, you would define the start and end months here, whether you want to skew the occurrence towards the front or the back of that time, how often the transient occurs, and how many cycles per occurrence that there are. Those are for your type 1 transients. If you have a type 2 transient in conjunction of the type 1, you also need to put in the stratification change in my brain at bending stresses. This column you can use to include some notes. And this one will carry over the transient names from the transient definition sheet. If you have a type 3, that's just a mechanical transient. You didn't put that separately down here in this table. Set up a similar to the type 1. It's just not confusing anything from the transient definitions here, just the change in membrane and bending stresses. And then finally at the bottom here, you can specify uncertainties for any of the transients using various distributions. These are basically a multiplier on the stress intensity factor. So that's just a quick review of the input set. I'm going to cover the other tabs at this point, but we'll get into one of them a little bit later. So just a quick overview of the input set file. That was the one developed and tested under the QA standards. You have all the sheets here, 10, which I reviewed with you, and then the three support worksheets, which I did not. And there are quite a bit of possible input cells there. So there's a lot of information that's being fed into XLPR, depending on what kind of analysis that you're doing. So we move on to our next topic here, input uncertainties. So like I mentioned, there's different ways you can classify your inputs. We talk about uncertainty of XLPR. We have either input variable uncertainty or model parameter uncertainty. And those are the ones that are going to be in the input set and what the framework is going to be sampling to represent uncertainties. That's mainly the way that uncertainty is dealt with in XLPR. And although nominally we may divide them up like this, they're really just all inputs into XLPR. There's also model uncertainty, and that refers to basically the different kind of equations that you might use to represent a particular physical phenomenon. One of the best examples of how we do that in XLPR is with the pre-initiation models. We have actually three of those available, either Direct Model 1, Direct Model 2, the Wyble Model. This is a list of the various distributions that XLPR does support. These are all listed on the drop list options of the input set. And you can see here there's a good number of ones that are available to you. And we have shown here in red which ones we actually use most often. So just more of a subset. But you have a good range there of input distributions that you can use. So I talked a little bit about aleatory and epistemic uncertainties. If you have an uncertain variable or an input distribution, you have to assign it as either aleatory or epistemic. You would assign it to aleatory if it's more of a natural variation. And it's not something you could really reduce. So a good example there is maybe a material property, yield stress, for example. Epistemic uncertainties are due to lack of knowledge. And those could, in principle, be reduced with further knowledge. A good example there would actually be weld residual stresses. Good epistemic uncertainty. And XLPR handles those through a dual loop sampling structure. You can use the dual loop sampling structure. But if you can also just use one loop or the other, aleatory or epistemic, put all your variables on either of those. And that's going to give you a classic single loop Monte Carlo analysis approach. I talked a little bit about the different sampling options. Although you can record a few of those in the input set, you actually really need to set those in GoldSim. So watch out for those. If you make them in the input set, but don't make them in GoldSim, your analysis isn't going to run the way you want to. And then just another tip here that the sampled input values aren't going to change the route given realization unless you have different operating periods. If you do, they will change if you've specified that. Otherwise that sampled input value isn't going to change until the next realization. So you're not going to really have the ability to simulate changes in, for example, temperature throughout the simulation unless you're using the operating periods. Okay, so let's review the inputs group report now. For that, the inputs group was focusing on coming up with example inputs for the three different cases, which are aligned with the well type database. And for each of those cases, we had inputs for different scenarios, covering fatigues, stress corrosion, cracking, and different kinds of mitigation at different points. And those are all listed here. We'll cover them in much detail, but that's what we did. So the 33 inputs sets in total, and then we also came up with all the inputs for the materials as well. The information we got the inputs from were from primary from the software quality assurance documents for XLPR. There also come from some vendor information that we were provided American Society, mechanical engineers, boiler pressure threshold code, code cases, things like that, as well as from U.S. Nuclear Regulatory Commission licensing correspondence, NRC, New Regs, serious reports, every technical reports, and also just the open literature. The inputs group itself report. The main thing, the main information is in the appendices. This said, there's lots of input, so these appendices are pretty lengthy. PENIX-A covers the non-material property inputs, and appendix-B covers the material property inputs. And when we release the board, you'll be able to see that a little bit more. I'll show it here a bit. Important caveat with the inputs group effort here is that we were not trying to develop comprehensive set of inputs or recommend inputs for any generic or plant specific thing. So you just, you know, buy or beware, I guess. Don't blindly use any of the inputs. You need to make sure that you're looking into them and making sure they're matching up with the type of analysis that you're trying to perform. Some tips from the inputs group document source of the inputs. I'll show you how we did that with a demo here quickly. We also recommend that you provide both deterministic and probabilistic inputs for each global ID. And I will demo this, but this helps you avoid any errors. It's better to use input truncation if you can. That will avoid unrealistic values. And just be aware that GoldSim is going to look for some minimum values to read in no matter what you're doing. You need to provide always a deterministic value, a distribution type, and the minimum required distribution parameters. If you don't fill those out, GoldSim is going to be looking for them and it's not going to know what to do otherwise. Okay, so I'd like to just show you a quick demo here on researching an input input using the inputs group report, the databases. And the example we're going to consider here is the plant operating pressure. Okay, so if I have a ragged and pressure vessel outlet nozzle problem I'm working with, I don't want to start from a blank page. So I'm going to look to an already prepared input set to help me set up my problem. And maybe even save me some time too. So I'm familiar with inputs group report. So I know there's some example inputs there for the RPV outlet nozzles. So it's actually case one, an inputs group report. And I'm going to, like I said, investigate operating pressure. I'm going to start by looking through the databases available to me. Look at the weld type here. And then here's case one, RPV outlet nozzle. I'm just going to search on pressure. Okay, here's a hit right here. Operating pressure for operating period number one. That's global ID 3101. And it's recommending a value of 15.41 megapascals. So I could maybe stop there. There's a number that I can use. But I may also want to know where that number comes from. So to do that, I'm going to use the inputs group report. And I'm going to search on that global ID. I do get a hit in the beginning here on that particular global ID. But what I really want is later on in the appendices here. This is appendix A operating pressure. See here tells me the global ID and what the description is. And then it's basically the inputs from the Excel spreadsheet except they occur in a list except for horizontally, like in the table. So I'm just going to look for that value, 15.41 megapascals, which is right here. It's telling me that we use that same value for all 33 of the cases and scenarios. And the basis was from this particular NuRAG report. So let's see if I can find out a little bit more about that. That was NuRAG CR 2189 volume three. The inputs group report told me that number came from page 10. And that's actually page 16 of the PDF. I like it's having a little trouble downloading this here. There we go. 16, okay. And this is page 10 here. And then it's showing me that value from the inputs group report 2235 psi. If I wanted to know a little bit more about that, I could read the context here in this particular report, okay? So that's just a quick example of how you can use the inputs group report to research and input and also the databases, okay? I'd like to turn it over to Nate to talk a little bit about the SEM editor. All right, thank you, Matt. Hello, everyone. So this portion of the presentation, we will learn a little more about the XLPR SEM editor. The purpose of the SEM editor, it basically was designed as a user-friendly graphical interface to help the users, especially the new users, in creating those input sets and to leverage those input databases that Matt spoke of earlier. So the SEM editor actually functions in two different modes, simulation mode and database mode. The simulation mode focuses on creating those input sets and modifying them while the database mode is used for the databases for modifying and creating the databases, all right? The SEM editor workflow is shown here. In it, you can see that it works directly with the Excel input set, but also can pull directly from those input databases and it's where you directly enter your inputs as well. So there's a couple nice features about the SEM editor that make it very useful for new users. First of all, the SEM editor hides parameters that you don't need. As Matt just spoke of, you saw there's many, many input options available in the Excel sheet and the SEM editor will help filter those down a little for you. It also organizes all the information in a little bit more user-friendly manner. The SEM editor also includes error checking for input value ranges and we'll show you examples of that later. And finally, you can import from those databases directly with just a click of your mouse. So those are a few of the nice features. I don't want to spend a lot of time on the setup since it's explained in the documentation included with your zip file. There is a SEM editor user's guide, as Matt mentioned earlier. You simply extract the files from the zip file and run the SEM editor installer. You can find more details about that in the SEM editor user's guide. So once you get the SEM editor started, this is the screen that you're going to see. There's a simulation mode that you see at the top. As well as the database mode. And finally, there's a database configuration. The first thing a new user should do is to set up the database configuration. We'll show you that on the next slide. So the purpose of this database configuration is simply to link a path to your database files. So you want to make sure you save those database files in a spot you can easily remember. You will be modifying them and you want to make sure you know where they are and you can easily find them. So the point of this is simply to map a path to those so XLPR can utilize them. The XLPR SEM editor does come with the fault database files as well. We suggest you use the ones that come with your SEM editor though. So next we'll discuss the simulation mode. Simulation mode is basically how input files are created. You can use the SEM editor if you want or you can use just the Excel files. It's totally up to the user. The inputs will be organized based on model settings. You see here at the top the different icons on the top left. Within each of those icons there will be a set of panels that are an example of shown here in green. On the right you can see how each icon fits into the inner workings of XLPR as well. So in entering simulation mode this is what you'll initially see. You'll see simulation shown at the top. We're looking at the general model inputs. Next to general you'll see the other global model settings. And to the right of that you'll see the material property settings. Way to the left you'll see a pencil. Please keep this pencil in mind. This is the edit button. Before you change anything in the SEM editor you need to make sure that you hit the edit button. If you're ever having trouble you can't edit something for some reason. Just make sure you have selected edit. It's a nice little reminder. So to navigate through here you can either select each individual panel shown in gold here or you can simply scroll down and you'll be able to access all the different inputs. One of the nice features is that the data is that you're seeing is filtered in the SEM editor. An example of this is for effective full power years. You see here that the data source is constant with a value of 60 years. However there are distributive properties if you choose. So if you choose to change the data source to aleatory, for instance, show here. You see that the distribution parameters appear. They were hidden before because you didn't need them. Once you select it they will appear. And then you can choose your distribution. Unlike the Excel sheet where everything is per M1, per M2, and so on for distribution, the SEM editor actually provides you the labels for each of those rows. So here we're looking at mean, standard deviation, min, and max. So another nice feature is that it labels things a little different. Gives users the ability to understand a little easier what they're looking at. Next we are looking at the weld options. You see there's four panels available to the left in blue. On the right side there are drop-down menus where data can be imported directly from those database files. So for each of those the user simply selects the drop-down menu, selects the options that they want, and everything associated with that material or that weld or those welding residual stresses will be directly imported into your input file, just a click of a button. So it's very easy and very nice. Keep in mind what we're showing here at the bottom. You see post-mitigation welding residual stresses. That means that mitigation has been selected. If you do not select mitigation, then you will not see that value. It will be hidden from you. A little tip to just help everybody out. So it's important that any weld options or materials are imported first, or it's close to beginning of your analysis as you can. Keep in mind that when you pull that information from the database, it indiscriminately will replace all of those inputs. So if you've already gone through and changed a few things, it will directly overwrite them. Also, when you switch between model settings, the page does reload. So it's always a good idea to save your progress before switching model settings. The welding residual stress panel is shown here. It's nice that all of the welding residual stress options are in one centralized location. The user can plot the welding residual stresses if they wish to view them. There's a few different places in the SIM editor where the user can plot, such as transient time histories, and as you can see, welding residual stresses. So it's just another feature available. The geometry panel is shown here. Notice that the units to the left, or excuse me, to the right, are grayed out. If something is grayed out in the SIM editor, that means it cannot be changed. It is subtle, but you will notice it. It is the same way in the input Excel file. So anything that can't be changed will be grayed out. Also, notice the validation ranges. So for pipe outer diameter under the value, you see that there's validation ranges of 0.4 to 1.4 meters. That means your input has to be somewhere in between those values. But let's say the user enters something that is outside of those values. Let's say 1.5 meters. In that case, the user will get a value out of range pop-up. The pop-up is just there to tell you that you did something wrong. If you hit cancel, it will revert it back to the value it was before, or the user can change it manually. There's also a set value to drop-down. It allows you to select the minimum range or minimum from the range, the maximum from the range, or an average in between. So it's part of the error reporting that the SIM editor does provide the user. The panels available in operating conditions are shown in the green. Please notice that the yellow caution signs next to some of the panels, these indicate that the panels are not needed in the analysis. So they will be hidden. So from what we're looking at here, operating periods two and three are hidden. That tells me that this analysis only uses a single operating period. In addition, what we're seeing here, the Tiffany parameters and transients are also hidden. That tells me that fatigue is not selected. The user will not be able to view those panels. It will not clutter the SIM editor. So it really limits what the user has to focus on. Next is the mitigation options menu. It allows the user to select whatever mitigation type they choose. Once the mitigation type is selected, the applicable mitigation options will appear. Again, the user can import material from the database in case they have inlay, onlay, or overlay. So they can pull those material properties directly with just a click of a button. The cracks menu is shown here, along with all inputs associated with it. Note that what's shown in the cracks menu is very high-level. It's basically what type of initiation you're looking for, what type of growth you're looking for, and whatnot. Leopold inputs all in one centralized location. For individual crack properties for each material, you want to make sure you go over to the material properties buttons. Again, this is high-level, whether you want fatigue, SEC, or initial flood density for crack initiation, that sort of thing. The inspection and leak detection panel is exactly what it sounds like. Note that you can select pre-mitigation, inspection options, post-mitigation, as well as during-mitigation, so all the panels are shown here to the right and purple. All correlations that are available are in one centralized location under the correlations menu. You can select correlations for your different materials or your inspection and leak detection properties. So, each material also has its own options menu. These match the spreadsheet tabs, so you'll see a lot of similarity. Keep in mind, just as Matt pointed out, there are less panels for left and right pipes than there are for weld and mitigation. That's simply because the left and right pipes exclude the initiation panels. There's also a couple of helpful tabs that the SIM editor comes with. First of all is tools tab. The tools tab, you can screen capture. There's a helpful units converter. You can view all your important sampling in one list. There's also instructions for running Leopold and Tiffany. Those are done outside of the SIM editor, but it still gives you some instructions. Well, we will touch on those on the next seminar. Next is the help tab. The help tab includes keyboard shortcuts, a document that explains entering distributions a little more. Finally, it's important to note that there is a GoldSim units database, so you can learn the abbreviations for the different units that GoldSim does allow. Notice that there's also submit feedback button. We ask that you do not use that submit feedback button. Any feedback, we would appreciate if it got sent to xlpr at nrc.gov or xlpr at epri.com, just as user manual states. So, now we'll briefly discuss the database mode. So, the SIM editor database mode just allows you to view, access, edit those databases that we talked about earlier. So, the available databases are well-typed material and the axiom hoop welding residual stresses. Note that we pre-populated them that you saw earlier. Those are predefined. Those predefined are password protected. We don't want you to change those. However, we do give you the ability to select new and create new database options or you can duplicate an existing database, rename it, and then you can change whatever you please. Also note that within the database options, there is no filtering. In the simulation mode, we filtered out to make it easy, but in the database mode, everything is left visible so that the user can set up as many of those options as applicable. And that was just a quick summary of the xlpr SIM editor. Please note that there's far more information, including the training that comes in your zip file. We'll also use demonstrations here to discuss a little bit further. And I'll take it back to Matt. Okay, thanks, Nate. Okay, great. All right, so now we'll end the presentation portion of our agenda today with a couple of demonstrations, which we're going to show two ways. The first, we're going to set up a problem using the input set. And then after, Nate's going to show us how to set up the exact same problem in the SIM editor. Today, we're going to set up the inputs. And then when we have the next webinar, we'll actually run those inputs and show you how to look at the results. So you want to stay tuned for that. Okay, so what we're going to do today is demonstrate the solution to challenge problem one from the xlpr training materials. That problem covers a lot of the key skills. And by recording the solution here today, you'll be able to reference this video later on if you need help completing it on your own. You'll find these training materials in the xlpr release package, the training folder, and we want the introduction training. And here you have the training manual itself, and then the folder containing all the files that you'll need to complete the various hands-on exercises included throughout that training. So let's open the training manual. Now, I highly recommend that you invest the two and a half days of your time to go through the entirety of this introduction training. That is the first thing you do after getting your copy of the code. And right here in the syllabus, you have a suggested schedule for doing just that. And then you're also provided with some links here to a prerecorded version of that particular training in YouTube, like all our other videos. So you can watch those, and then the slides for the training are basically provided throughout the rest of the training manual here. So you can follow along the videos of the slides in the training manual. So you'll complete several exercises as you work through the various modules in the training and sequence. You'll be building skills and then working up to the challenge problems at the end. That's what we want to demonstrate today. So let's go ahead and skip ahead there to the last module and the first challenge problem. Okay, so the challenge here is basically to modify an existing set of inputs. And they tell me where I can find all the files that I'll need to do that. Now let's review the changes I'll need to make. First, they want me to change the operating temperature midway through the simulation using a linear triangular distribution. I also need to make some changes to the pipe wall thickness. I need to update the material properties on a white right pipe sheet. And then I also need to check some sampling settings. Okay, so let's go ahead and implement all these changes. I'll navigate to the files that I need to use. Those are in the exercises. Day three, challenge problem one. Now I can go ahead and work the problem right here in this folder. And if I mess up and I need to restart, I can always go back to my zip file. Then I keep it handy to retrieve the original files. Let's open the input set. So I was instructed to update the temperature midway through the simulation, like I said. And the only way I can do that is if I have multiple operating periods. So let's see how those are set up here on the user options. The overall simulation here is global ID one of 720 months or 60 years. And the duration of the operating periods that would divide. So I see operating period one here set to end at 720 months in global ID 402. This occurs after the 720 month operation time. That means it's currently set up. There's only one operating period. So I'll introduce another by making operating period one. And at the time they gave me, which is 35 years. The basic time step in XLPR is one month. So that's what I like to work with. I'm going to enter 35 years in terms of months. Now an advantage of working with Excel input set is that it supports formulas. So I'll multiply it 35 by 12 to get 35 years in terms of months. 420 months calculated for me. Okay, the next thing I need to do is update the operating temperature itself. With operating period two now active from 35 to 60 years. I'll need to make sure I've completed all the applicable inputs for that operating period. I do that on the property sheet. Okay, scrolling down. See I have a category of inputs here for operating conditions. And those are defined for each operating period. I need to update the ones in operating period two now. Specifically global ID 3202 operating temperature. It needs to be on the aleatory loop. So we'll change the data source from constant to aleatory. I do that at darkens the text for the parameters I need to input because now I'm working with the distribution instead of constant. Currently there's a normal distribution defined with some dummy values. I need to input a triangular distribution. We'll change the distribution type in the cell to triangular using the drop down menu. Now I need to enter the parameters for the triangular distribution. I can determine what those are by referencing the drop list options tab. This sheet lists all the available distributions which we explained before here on the left. And then what all the parameters that those distributions need the columns. Let's find the one for triangular. Okay, so for a triangular distribution, param one is a flag indicating whether it's linear or logarithmic. Parameter two is minimum. Parameter three is the mode. And parameter four is the max. So now what I know what the input parameters are for triangular distribution we'll go back to the property sheet and enter the values I'm given. Okay, so versus it needs to be a linear. I'm told it needs to be a linear distribution which would indicated by zero. Param one. Param two is the minimum. The given value is 310 degrees Celsius. That's what I enter here. Parameter three is the mode. That was 325. And parameter four was the max. That was 340. And following the inputs group recommendations, I'm going to set a deterministic value as well right here. And I'm just going to set that equal to the mode. That completes the inputs for operating temperature. However, because I activated operating period two, I need to make sure I complete all the inputs for that operating period. That's going to also include the operating pressure global ID 30201. I'm just going to assume that it's the same as from operating period one. I use a formula to help me do that. Dissolve oxygen is really just for fatigue, so I don't need to worry about that here. The other group of inputs I can change from operating period to operating period are the loads and stresses. I'll scroll down there for operating period two. And again, I'm just going to assume that these are equal to the loads from operating period one. Using formulas. And then just like any Excel sheet I can direct to complete. And now I don't need to worry about the stresses because back on the user options I defined that I would be inputting these inputs by loads instead of stresses. I was also instructed to update the pipe wall thickness. That's up here at the beginning of the property sheet. Global ID 1102. I want a constant value, so I'm going to change the data source for map as the constant. And then in deterministic value, I'll enter the provided number, 0.067. I'll check the units. It's meters. Meters is what's required. And in fact, I cannot enter anything else. And I know that because of the color shading here, that red color corresponds to inputs that can't be changed by the user. I'm also going to follow the input recommendations here as well. And back to my global ID 1102. And I'll define a distribution as well. This is not required. I'll just show you how you can do it. It's a good practice to avoid errors. Discrete distribution, you need to say the number of pairs in the first parameter, the probability, 1.0 or 100%. And then the actual value itself, which we'll just take from the deterministic value. Okay, the next thing I need to do is change the write pipe material to 304 stainless steel. So I'll go to the write pipe tab. Up here at the top, I can see it's currently 316 stainless steel. I can change that back on the properties, but I'm not going to show that today. So what I'm going to do to make this 304 stainless steel, I'm just going to copy from the materials database that's provided in the release package. This is how you can do that. This is how you get there. Go to the database folder. I have the one for materials. So each material has its own sheet on the bottom here. I'll find the one for 304 stainless steel, which is right here. And what I'm just going to do is grab everything that I can. I'll have to do this in two copy paste operations, actually, just because the way the sheets are formatted. But I can get most everything in this first operation. I've got pretty much everything up until the correlations. Copy back to the input set, write pipe tab. And I'm just going to paste it in here. Now, this is important when you're working with the input set. And copying and pasting only paste the values. And this ensures that you don't overwrite any of the formatting that's in the input set. That's the first operation. Let's go back to database here. I need to grab these correlation inputs. Copy those. And then back to the input set, and I'll put them in the corresponding spot in here. Again, just pasting the values. Okay. So that's material properties done, updating. So now the last thing I need to do is just check the sampling parameters. Those are back on the user options up at the top. Here in global ID 101 is the epistemic sample size. That needed to be 10. Looks like that's what's here. I'm just recording that here. I can't. She said that needs to be done in GoldSim as noted. The random seed for the epistemic loop needs to be 1. That's global ID 102 here. That's also set correctly. The aleatory sample size, global ID 107, 20. And global ID 108, the random seed for the aleatory loop. That's 5. That's already set here as well. Okay. So no changes needed there. And lastly, I just need to specify Latin hypercube sampling. That's at the bottom. Just recording that here as well. It needs to be turned on, and that's indicated with a 1. And that is the case here for both the epistemic and the aleatory loops. Okay. So that's what's here. So I was able to confirm all the sampling options are correct. And now I'm ready to run the problem, we'll show you in the next seminar here. That concludes demonstrating the first part of challenge problem 1 using the input set. So now Nate's going to show you how to set up this exact same problem using the SIM editor. All right. Thank you, Matt. So let's get into the SIM editor. As Matt said, we will do the exact same thing just to show you the differences. So here I'm sharing my SIM editor for you. I've already opened the file just to make things go a little quicker here. And the simulation opens by us looking at the general settings. So the first thing I want to make sure you notice is that there is an error ribbon, bright yellow here at the top. And this is really to help users out. You see here we have 0 errors to resolve in 16 warnings. So the first thing we always want to do is to understand what those warnings are, seeing if there are anything that we need to be concerned about. So view warnings, we can go down. Sulfur content, since fatigue is not selected is the first four. Well, this simulation does not have fatigue selected, so that's acceptable. Again, there's no fatigue initiation. That's acceptable as well. We don't have a problem with that. Load stresses and operating conditions for periods two and three. Well, we know the point of the exercise is to add a second period. So we fully expect those to be turned off. So those again are no issue. And the rest of the warnings are all to do with transience and fatigue. Again, fatigue isn't on. So all these warnings are fine. They're helpful, but always make sure to look at them whenever you start. Now that we understand what warnings we have, we can now continue on. So the first thing we want to do is add that second operating period. You do that at the top by going to operating conditions. Select that and load the panel. You can see there in loads and stresses and operating conditions for period two that it's got the caution symbol, which means it is hidden. So that's as we expected. So we want to make sure we add a second operating period. So we'll scroll down to this operating period. Make sure we hit edit. Now we hit edit. We can enter the end of operating period one, which was 35 years, 35. The abbreviation for years is just YR. And remember, if you go into the STEM Editor Help, it'll provide you with those gold STEM abbreviations. So now we've added our second operating period. You see it highlights. As long as the second operating period ends after that's acceptable. So our analysis is 720 months long. So the second operating period ending at 722 is acceptable. So now you see that loads and stresses for operating period two are now available to us. Also note to the right here, operating period one used loads. So we're just going to assume the same thing just as Matt did and leave period two as loads. So our next step is to go down to the operating period two, loads and stresses. You can just scroll down or we can click. It'll take us directly there. Notice that there's a second scroll bar available for us. Also we could pull this menu down and adjust it as we please. So keep those in mind. Now one of the things that's a little more difficult in the STEM Editor is copying large amounts of data. So you saw Matt just put in formulas and copied it. We could copy as well, but we'd only be able to copy one value at a time. So for these inputs, it's simply easier to edit them manually. So I'll just type these in. They're exact same as operating period one. So if you just bear with me, I'll make sure we get all these entered. Now, like I said, you can copy individually, but you cannot copy several different inputs. At once, so that wraps up the dead weights. So now we'll move on to the thermal. Do the same. Now you can choose to do some of the analysis or some of the setup in Excel, some of it in the STEM Editor. As you please, you can go back and forth as you want. It's always easier to kind of stick with one, but it is up to the user. Now those have all been entered. So let's continue down into our operating conditions for periods. So if we scroll down, the operating conditions for period two is now available. The whole purpose of what we want to do here is to change the operating temperature to a triangular distribution. So the first thing is to change the data source to Alliatory, and you see it populates our distribution. It comes in as a normal distribution. We change that to a triangular. We now see that we have our limb log, men, mode, and mesh already labeled here to help the user out. So this is a linear. So we want to enter zero here, as it says. The minimum is 3D10, mode of 325, fine and maximum of 340 degrees. So that's all that takes, but notice up here that when we created a second operating period, we also need to make sure we include pressure. So we'll go back and look at what the pressure was for operating period one. We'll assume it stays the same. Talk to you and paste that in. So now we have our operating period set up, and everything filled out appropriately. So we'll make sure we save. Save, choose either one. It's up to you. Saving takes a couple seconds, and now we can move on to the next step. So the next step is to change geometry. So we find that in the weld options, settings here at the top. So click on the weld options setting. As you can see, it does reload the page. That's why we want to make sure we always save. Okay, so now we want to navigate down to geometry. So if we go to the panel, click on geometry, it'll take us all the way down to where we need to be. And our goal is to edit the pipe wall thickness to a constant value. First thing we have to do is select edit. After we save, the edit disappears. So we need to make sure we select it again. All right, we want to change the data source to a constant. So we'll click on epistemic, scroll down to constants. You can see the distribution information disappeared. And now we're left with a simple view that allows us to input our value. So our value is 0.067 meters. As you can see, the meters are scraped out. We cannot change that value just as the cell sheet was. So now that we've changed the geometry, now it's time to change the right pipe material. So we can either again click on the weld properties or we can simply just scroll up. Takes you all the way back to the top. Currently, our right pipe material is listed as 316 stainless steel. We want to change that to 304. So if we click on the dropdown list, it provides us all the different options we have for materials. We want to select 304 stainless steel. It will basically ask you to confirm if you want to proceed. It's letting you know that it's going to pull all those materials from the database. Yes, we want to proceed and that's it. So instead of copy and pasting a large amount of data, it was simply clicking on that dropdown menu and now it is all included. Okay, so the next step we want to do is we want to make sure we save. Notice that when we saved, we now have an error. So we want to investigate what that error is. So click here to the error has an X next to it. Tiffany parameters, weld thickness value is out of range. Well, we investigate what happened here. When we changed our geometry, the geometry no longer matches the Tiffany parameters. However, we do not have fatigue selected. So Tiffany is not a concern for us. If we wanted to get rid of the error, we could go to operating conditions, go down to Tiffany parameters. You see this is where the error is, the program lets you know. However, if we click on it, it's hidden. It's because symmetry hides anything that you don't need. So if we did want to change the input file and get rid of that, we could turn fatigue on, change the Tiffany parameter, turn it back off. But we know that we're not using the Tiffany parameters. So we're okay to leave that error as. So the next step we want to do is we want to go into general and look at our sampling. So the simulation settings, sampling approach epistemic, sampling approach, aleatory are all located here. We want to make sure that our epistemic is set to 10. It is a random seed is set to one. It is as well. We also want to make sure that light and hyper-deep sampling is turned on. Can confirm that it is. Next is to go to aleatory sample size. I'll confirm that the sample size is 20. It is random seed is five and usually hyper-deep sampling is turned on, so it's yes. So those are already set. We've just confirmed it. So keep in mind, there's also notes out to the side of a few different places. For instance, sample size epistemic, it says this input must be entered into GoldSim. The value being entered into both the Excel sheet and the SIM editor for epistemic sample size is strictly for record-keeping purposes only. We do not need to, the program does not actually read these values from the Excel spreadsheet. So we just want to make sure we have good record-keeping here. Once we talk about the seminar for next week, we'll go into depth on how to enter the sample size and the random seeds. But keep in mind that the aleatory sample size is read from the input file. So this is the only input that we confirmed here that will be read from the input file. You notice it does not have a note out next to it, so please keep that in mind. So now that we've made all those changes, we are good to go and run the analysis, which again we'll talk about next week. So that's how to use the SIM editor for that part of the seminar. For that practical example, and I will go ahead and turn it back over to Matt now. Okay, thanks, Nate. So as you see, same problem. You can work at two different ways depending on what you're comfortable with. Some ways are faster in the input set and some ways are faster using the SIM editor as you saw. So there's pros and cons of each, but both are useful, I think. Particularly the SIM editor when you're first getting started because it's a little bit easier to use. So let's see. This looks like we're actually all caught up on the Q&A we've received so far. I had planned for us to take a short break here so the team could caucus and maybe determine what other things we might need to show you to help answer your questions and things like that. So I don't see any other questions but let's just go ahead and we'll take a five-minute break here and see if there's anything else we might want to just show you with the remaining time here. So I've got half past the hour here. We'll let's all meet back up here at 35 past the hour and you may see your screen here change a little bit. Don't worry about that. We're here and we'll be back in five minutes. Okay, thank you. All right, everyone. Hope you had a chance to grab a beverage or something like that. The short break we had there. We spoke a little bit with the team here about how we might use the rest of the time here. There was a question on aleatory and epistemic uncertainties and we wanted to just provide a little bit more explanation on that. So I'm going to have Cedric help us answer that. Thank you, Matt. Okay, so as we discussed, there is a theoretical definition of epistemic and aleatory uncertainty and usually as it was written in the Q&A, the aleatory uncertainty is a random something even if you have more data it's still here. So generally to illustrate this we use the concept of rolling a die or several dies unless you cheat you don't know what the result will be. So it can be a one or two and it's not because you will roll 10,000 times that you will know what will happen on 10,000 and one. So that's aleatory uncertainty. We don't know it's randomness generally in future events and we don't know what's going on. For epistemic uncertainty it's more lack of knowledge. It's something that if you have more data, more tests, you know better and if you take the rolling dice example it's just saying what's the probability to have one, a two, a three and a six and so on. You expect this to be one thing but maybe the die is biased on something. So the more you roll the more you can confirm or infirm this case. And in our world epistemic uncertainty will be more of values that you expect you know the uncertainty could go down like a mean value so the more tests you do you have a better estimate of the mean value while aleatory uncertainty is still something you don't know like when the the crack will happen and something like this. Now another question we were talking during the short break was you may have which is why do we want to differentiate from epistemic and aleatory uncertainty? After all as we say we can run a single loop and we can just make one uncertainty and give the result and sometimes it's a correct way to do. The reason we want to differentiate is when we think in term of risk. Aleatory uncertainty really represents our risk. It's what can happen in the future and how likely it is to happen. So we don't know and we want to see what the risk is and it's represented most of the time as a probability distribution. Epistemic uncertainty is not really a risk as we say it's a state of knowledge uncertainty. So epistemic uncertainty is more our uncertainty over the risk. What we don't know about the risk variation so it's more a band around the probability and we have to be able to reduce it when we increase our knowledge. So in this case if you want to differentiate the risk versus your lack of knowledge that give us a band around the risk then you can use this double loop and separate these two then you can have cumulative distribution function and a range of cumulative distribution function to tell you how much uncertainty you have in this risk and what you can insurely reduce to have a more accurate risk. So I hope this definition give you a better concept of why we have included the double loop in XRPR and how we differentiate aleatory from epistemic uncertainty and on that note I will give the talk back to Matt thanks. Okay thanks Edric for that additional explanation on that question. I don't see any other new questions but I know we were moving through some of the demos and stuff I think good pace but you know you need to pay attention there to make sure you got all the details so if you maybe thought you missed something or something like that let us know. We could redo that part of the demonstration or you know any other questions of course on any other slides or anything like that would be happy to help answer right now. Maybe what I should also ask is you know if you could just let us know the chat here maybe or even by email after the the meeting here is you know kind of how far you've gotten with the code if you've tried to run it yet on your own just kind of started to crack it open a little bit or you're just kind of stuck. So let us know that'd be helpful for us to gauge where you're at so we can help make these webinars better in the future. We'll see if we get any other questions here. Also if you wanted to ask a question verbally I'm sorry I forgot you could raise your hand and we can unmute you and that may be easier either way through the Q&A or if you want to raise your hand and we can take a question verbally that may be easier given some of the more complex topics we've been talking about today. Okay like we got another question. Sayed go ahead. Oh sure hi I was wondering if you could elaborate on the variables of epistemic versus aleatory how you would characterize them. You talked about the variability and also a set of knowledge but I was wondering how you basically characterize those in terms of the for example when we have several variables I heard that the yield of stress you mentioned as a aleatory variable I was not sure how you basically kind of convince yourself that is a aleatory variable if you could elaborate on this point that would be great thanks. Yeah on this point I would say if you consider a specific weld to you know for a specific power plant then it would be an epistemic variable because you represent just one weld you know there is a unique yield strength and ultimate strength for this weld but you don't know it. Now some analyses we perform may be to look at the you know a set of plants or a fleet of plants and welds in general and in this case when you represent you don't know which specific weld you represent you just represent a family of welds and so on in this case it may be considered aleatory just because you you you don't specifically represent one plant or one weld so it could be any of them. Now in many cases we have seen that aleatory and epistemic uncertainty separation are quite complex the first reason in many of this input may have both an aleatory and an epistemic component and you have to see which one you want to represent and which one generally is the biggest of the two uncertainty and as we say in this example it really depends on the analysis you do do you want to perform the analysis for a specific weld do you want to perform the analysis for a generic weld do you want to perform the analysis for a fleet are you interested in using a mean value are you interested in representing the all possible variability so it can be any of these values and you don't well you have no reason to prefer one versus another. One important point is even if we separate aleatory from epistemic uncertainty we have a double loop this will not change the mean value if at the end you represent a mean probability of having you know a crack occurring a leakage and so on the mean will not be affected because when we think about it the separation is just splitting the Monte Carlo loop into two loops so instead of having one integral we have two integral but the mean over two the two integral will be the same of the mean over one integral. So we know that why the intermediate result as we say when we talk about risk and about uncertainty of the risk may be affected by the choice of aleatory and epistemic and the distinction is not always obvious and may depend on the analysis we know at least that the mean should be consistent once you look at the end of result does it answer your question? Thank you for your response I appreciate probably it means that the when we want to communicate our findings with the other groups for a specific problem we would need to basically clarify this point in our understanding and the type of analysis correct absolutely yes thank you when you present the result you you will have to say why it's conditional on considering this aleatory and this epistemic and if you change this you may have a different answer not for the mean but for the other value sure yes thank you you're welcome thank you for your question anybody else has a question go ahead and you can switch it to the q and a or the raise your hand all right I think that we have answered all the questions at this point I double check yeah okay all right I'm not hearing any more questions here and we are coming up towards the end of the meeting I'll go ahead and take us to the closing remarks Giovanni would you mind just making me the the host by the presenter I mean here we go okay questions and answers we just wrap that up all right so today's meeting we we reviewed how to set up the inputs in xLPR using both the spreadsheet and the simulator simulator and we talked about various topics about inputs and we also reviewed some of the other databases and things like that other tools that are available to you so let's look ahead here like Nate mentioned earlier that our next meeting is in is in a week but it's actually in two July 29th same time 10 10 to noon on the east coast here in the the United States what are these same local time for everyone else so the next topic we'll be talking about then we'll be running the simulation and retrieving the results we we are able to combine those two topics into one session and we suggest that you review your use manual topics before that session primarily in section three and we have listed here a couple of the subsections here that you would want to maybe look at specifically and then we're going to have our August 5th we're going to cover advanced methods so that's a variety of topics here for using some of the more advanced options in xLPR like multi processor and and things like that so that's August 5th so those are the last two webinars we have planned and like I said if you have any feedback would be we'd love to have that we'd like to make these sessions as useful as we can for the new users so please go ahead and send that to the xlpr at nrc.gov and xlpr at epri.com so I want to thank everybody again for their time for for joining in welcome the new users again after the meeting ends if you want to provide feedback so we can help make these meetings better just in general you can do that from the nrc website you just find this meeting listed there the recently held meetings and you can click on it and there'll be a forum where you can fill that out separately just on xlpr topics you can use the email addresses here to reach out to to craig and I and we'll help you have the best of the weekend craig anything else to add here as we wrap up no I think you've covered it Matt and as I said earlier if you still haven't gotten the code and you have any questions or issues I'll also try to help with that if I possibly can so just just let me know and thanks for participating today I hope I hope you found this useful I hope you find this series okay great thanks craig and and we'll uh we've recorded this session again we're gonna do the transcript and make this available on youtube.com as well and we'll have a summarized meeting available as well so we'll get that out to you as soon as we can hopefully before the next webinar here so with that I think next again everybody for joining us and we'll go ahead and turn the meeting and we hope to see you at our next one thank you very much