4. Simulations

As scientists and engineers, we all know the importance of simulations in solving complex real-world problems safely and accurately. Simulations make information and data conveyable for better decision-making. d3VIEW provides the ability to manage, create, track, compare and share your simulations and their data more effortlessly and efficiently. The Simulations application is divided into four main parts: Preview, HPC Job, Files and Responses. In this tutorial, we’ll review navigating the home page and these sections, so you can start managing your simulations more effectively.

What Will Be Covered

  • My Simulations Page
  • Preview and Simulation Details
  • Errors and Warnings
  • HPC Job
  • Simulation Files
  • Simulation Responses

4.1. My Simulations

From the d3VIEW home page, click on the Simulations button in the left side panel to go to the Simulations main page.

Figure 1: Accessing Simulations


Here, you can submit a new simulation (1), sift through your simulations by using the quick filter buttons (2) or using the drop down or search bar in advanced filtering (3). Review and restore any of your deleted simulations at the bottom by clicking on Recycle (4).

Figure 2: All Simulations


We can utilize advanced filters when we want to look for specific simulation qualities such a solver type or termination type as demonstrated in the following video:




4.2. Simulation Preview

Click on a simulation in the list to open and see its contents in a new tab.

Figure 1: Open Simulation Viewer


The first section at the top, ‘Home’ provides a real-time status of the simulation while it is running. The panels allow you the visualize different elements of the running job. Some useful information to examine are energy balance, minimum time-step history and time remaining for completion. Energy balance provides incite into detecting any abnormalities in the simulation and, if required, kills it before the end time. You can also click on the last section ‘Errors and Warnings’ to identify any issues with the simulation.

Figure 2: Simulation Details


Simulation Processing Statuses

Here is a graphical representation of the processing statues a simulation runs through while solving:

Figure 3: Simulation Processing Statuses



4.3. Simulation Errors and Warnings

View all the errors and warning from simulations files under this tab in your simulations. This reduces the need of searching for errors in your simulation files.

Figure 1: Simulation with No Errors or Warnings


Email Notifications

If you have an email linked to your account, you’ll also receive warning and error messages once the simulation is done solving via email. To update your email, click the down arrow under your name at the top right corner of the screen (1) and choose ‘Settings’ (2). Then under ‘Info’ (3), input your desired email in the space provided (4).

Figure 2: Update Email


Here are some examples of email notifications we may get of errors in our simulations

Figure 3: Email Notification Examples



4.4. HPC Job

In order to run a simulation on the HPC, you have to submit a job. To start a new job submission, click on ‘New Simulation’ in the upper right corner of your Simulations page and follow the instructions presented in the previous tutorial section: Job Submission. Submitting a job requires a specified configuration which can be reviewed after clicking on the respected simulation and going to the HPC Jobs tab at the top. This tab provides the job submission history such as the node the job was submitted to, the number of CPUs used, etc. The information can be useful as comparisons and guides for submitting new simulations.

Figure 1: HPC Job



4.5. Simulation Files

Simulation files are processed through the server and compute node to be available to you locally.

Figure 1: Client-Server-Compute Node File Location


All the output and log files generated by the solver (bin-outs, d3plot, solver.log etc.) are available in the Files tab (1). Download, share or view individual files by clicking on the 3 dots next to it (2). Add a new file at the top (3).

Figure 2: Simulation Files


Viewing a file opens it in another window that is adaptive to the file type. Here, you can review the file by utilizing the search bar (1) or download it (2).

Figure 3: File Viewer


Download or share multiple files at once by selecting them via the checkmark (1) and using the Selected drop-down menu (2).

Figure 4: Multiple Files Actions


New as of October, 2022, the data viewer supports JFIF images as shown in the following image.

Figure 5: JFIF Image Support


File Watcher

You can watch files in Simulations using File Watcher which has options to refresh the timer and url. They following video shows an example:



Lucy Log

You can view lucy.log files for simulations and visualize the rate of communication between Lucy and d3VIEW.

Figure 6: Lucy Log


You can also change view type to log nester for lucy.log files which groups the aspects such as OUTPUT and ERROR.

Figure 7: Log Nester



4.6. Simulation Responses

Data output extracted from the simulation and its bin-outs, d3plots (nodal displacements, element_history, etc) are available in the Responses section (1). Any top, front and left view animations are available by default. Use the View Type and search bar (1) or the quick filters such as the image button (2) at the top to search through your responses.

Figure 1: Simulation Responses


You can perform actions on a single response (1) such as viewing in full screen, exporting, sharing or duplicating it. You can also perform actions to selected responses from the drop-down menu (2) such as creating a template from the responses or exporting them to PowerPoint.

Figure 2: Response Actions


Add a new Response to your simulation by using the data extraction tool.

Figure 3: Extract Responses


Curve Responses link with d3VIEW’s Curve Viewing Application Newton for enhanced analysis. To learn about Newton, follow this link.

View Images

Use the View Images button to quickly view images in a simulation.

Figure 4: View Images


Filtering

Click on the filter icon at the top to use customized filters for sifting through your simulation responses.

Figure 5: Filter Responses


The datatype filter has an array of filetypes to choose from including MDF/DAT file options which have been adde as of February 16, 2022.

Figure 6: Filter Data Source


Comparing Responses

Simulation responses can be compared in Simlytiks with other simulation or physical test responses. Let’s review

Comparing with a Template

We can compare responses using a template which will have visualizations set up already for us. Watch the following video to see how it’s done.

Comparing with a Template



Comparing without a Template

We can also compare without a template. The following shows how to do this with an example where the simulation is compared with a physical test.

Comparing without a Template to a Physical Test


Compare responses now has an option to set the extraction type to all simulations using a single button at the top.



This process is the same depending if you are comparing from the Physical Tests or Simulations page. To see how this is done via a step-by-step image explanation, please navigate to the Physical Tests section on Comparing Responses.


4.7. Extracting Responses

As mentioned earlier, create a new response by modifying simulation data under “Extract Response” in the Simulation Responses tab.

Figure 1: Choose Extract Response


In the next window, you can choose from a few different data extraction options. The most common ways to extract data will be from a data source (1) or employing a transformation (2). For data sources, available options will be dependent on the database files of the simulation.

Figure 2: Choose Extraction Data Source


Finish all extractions by choosing “Extract” (1). Feel free to see a preview of your extraction first by clicking “Preview” (2).

Figure 3: Finish Extraction


Binout Extractor

The ls-dyna binout (binary output) extractor post-processes a variety of simulation files efficiently. Depending on the type of simulation, we’ll have a list of different file types we can choose. Here are the options available for a Occupant Belted simulation:

Figure 4: Binout Extractor Choose File


Here we’ve chosen nodout as our file type. The boxes show the labels and IDs of all database options available.

Figure 5: Binout Extractor


Then, we’ll select our components and IDs in the multi-select boxes. In this video example, we name our extraction “Displacement” and choose displacement components and node ID 1.

Binout Multi-Select



Find IDs

The Binout Extractor has an advanced feature for find specific IDs. We can use the Find IDs option instead of selecting IDs if we are unsure of which ID to select for a specific extraction. We’ll specify the scope to find a group of responses extracted as one and then decided which ID from the group we would like to extract individually. Let’s review an example using a Punch Specimen simulation.

After choosing binout in the response extraction window choose the file type (1), select the component (2) and name the response (3). Here, we are choosing elout/shell for the file, lower_eps_xy as the component and naming the response as such. We’ll then want to click on Find ID under the Ids section (4).

Figure 6: Choose File, Component and Find ID


This will give us a list of options we will fill-out as the scope for extracting the group of responses (scroll down to see them all). Here, we are choosing Any ID value (1), Any State (2), Sort by Y Max (3), Descending (4), and 5 as the Limit (5).

Figure 7: Find ID - Fill Out Scope


Click Extract to add the response. We’ll see (1) in the Enqueue as the extractor is processing.

Figure 8: Extract Group


Under our simulation responses, we’ll see the newly extracted response group. For this Punch simulation, we have a group of curves and want to look for the ID of the curve with the highest peak: the blue curve with ID 439.

Figure 9: Find ID for Highest Peak


Now, we can go back to our binout extractor and just extract that curve. Here are the settings we’ve chosen for it. The same file and component, but now we are searching for the 439 ID and selecting it. Make sure to give this response a unique name.

Figure 10: Extract Highest Peak


Once we extract, we’ll see the curve with the highest peak as an individual response in our simulation.

Figure 11: Highest Peak Response


d3plot Extractor

The ls-dyna d3plot extractor post-processes simulations in 3D visualizations and animations. d3plot is most commonly used for creating 3D models (with or without plastic strain) of our simulations but has an array of file options.

Figure 12: d3plot Extractor


Let’s review how to create a Peacock model using d3plot extractor. First, we’ll name out response and make sure to end it with the js3d.zip file extension for viewing in Peacock (1). Then, we’ll choose Peacock3D as our output type (2). Next, we’ll choose the Plastic Strain fringe type (3). (There are also options for von-Mises Stress and Thickness). Lastly, we’ll indicate the final state to be 100 (4). All other options can remain as the default (or as shown in the following image). Click extract to finish.

Figure 13: d3plot Peacock 3D Response Set-up


On our simulation responses tab, we’ll see our new response with a Peacock logo. Click on the logo to initiate it in peacock.

Figure 14: Peacock 3D Response in Simulation


We can now explore the model in 3D space. To learn how to navigate Peacock, check out that section here.

Figure 15: Peacock 3D Model


d3hsp Extractor

The ls-dyna d3hsp extractor is most commonly used to check the model information such as to investigate the causes of the simulation’s error termination, similar to a solver log. d3hsp can print all the information related the model and the solver states as the simulation is being solved.

Figure 16: d3hsp Extractor


Here are all the d3hsp options available for extraction. Read on to see examples of each.

Figure 17: d3hsp Options


Mass Summary

For this set-up, we’ve chosen d3hsp for the file name, Mass Summary for the file type and named the response.

Figure 18: d3hsp Mass Summary


This extraction gives us individual summary responses for each aspect of the simulation execution.

Figure 19: d3hsp Mass Summary Responses


Summary

For this set-up, we’ve chosen d3hsp for the file name, Summary for the file type and named the response.

Figure 20: d3hsp Summary


This extraction gives us individual responses for a summary of the simulation execution.

Figure 21: d3hsp Summary Responses


Errors

For this set-up, we’ve chosen d3hsp for the file name, Errors for the file type and named the response.

Figure 22: d3hsp Errors


This extraction gives us an individual response for solver errors.

Figure 23: d3hsp Errors Responses


Warnings

For this set-up, we’ve chosen d3hsp for the file name, Warnings for the file type and named the response.

Figure 24: d3hsp Warnings


This extraction gives us an individual response for solver Warnings.

Figure 25: d3hsp Warnings Responses


Control Cards

For this set-up, we’ve chosen d3hsp for the file name, Control Cards for the file type and named the response.

Figure 26: d3hsp Control Cards


This extraction gives us individuals responses for Control Cards.

Figure 27: d3hsp Control Cards Responses


Bill of Materials

For this set-up, we’ve chosen d3hsp for the file name, Control Cards for the file type and named the response.

Figure 28: d3hsp Bill of Materials


This extraction gives us individuals responses for Bill of Materials.

Figure 29: d3hsp Bill of Materials Responses


Material Details

For this set-up, we’ve chosen d3hsp for the file name, Control Cards for the file type and named the response.

Figure 30: d3hsp Material Details


This extraction gives us an individual response table for Material Details.

Figure 31: d3hsp Material Details Responses


Keyword Count

For this set-up, we’ve chosen d3hsp for the file name, Keyword Count for the file type and named the response.

Figure 32: d3hsp Keyword Count


This extraction gives us an individual response for Keyword Count.

Figure 33: d3hsp Keyword Count Responses


Contacts Summary

For this set-up, we’ve chosen d3hsp for the file name, Contacts Summary for the file type and named the response.

Figure 34: d3hsp Contacts Summary


This extraction gives us an individual table response for Contacts Summary.

Figure 35: d3hsp Contacts Summary Responses


Contacts Parameters

For this set-up, we’ve chosen d3hsp for the file name, Contacts Parameters for the file type and named the response.

Figure 36: d3hsp Contacts Parameters


This extraction gives us an individual table response for Contacts Parameters.

Figure 37: d3hsp Contacts Parameters Responses


Curve By ID

For this set-up, we’ve chosen d3hsp for the file name, Curve By ID for the file type and named the response.

Figure 36: d3hsp Curve By ID


This extraction gives us individual curve responses by ID.

Figure 37: d3hsp Curve By ID Responses


Digitized Curves

For this set-up, we’ve chosen d3hsp for the file name, Digitized Curves for the file type and named the response.

Figure 38: d3hsp Digitized Curves


This extraction gives us individual digitized curve responses.

Figure 39: d3hsp Digitized Curves Responses


nastran_pch Extractor

The nastran_pch extractor uses a data source from a NASTRAN simulation for response creation.

Figure 40: nastran_pch Extractor


Let’s review an example of extracting a response for a NASTRAN simulation. When setting up the extraction, we’ll want to indicate the subcase ID (1), entity ID (2) and entity component (3) as well as give the response a suitable name (4).

Figure 41: nastran_pch Extractor


Upon extraction, we’ll see the new response in the simulation, this one being a raw curve.

Figure 42: nastran_pch Extracted Response


We can extract more responses using the nastran_pch extractor or use dedicated template. Here is an example of responses in a NASTRAN template with the one created above indicated with an arrow. To learn more about Templates, c:ref:check out that section here. <Templates>

Figure 43: NASTRAN Response Template


Transformations

Applying transformations to simulation data involves employing a worker to a current simulation response. There are multiple ways to apply transformations, but the easiest way is to choose the worker from the Simple menu.

Figure 44: Simple Transformation


We can search for our desired worker or use the drop-down menu to sift through categories.

Choosing A Worker



Click on your desired worker to see it’s inputs. Here, we’ve chosen curve_reversepoints.

Figure 45: Choose Worker


Then, we’ll drag-and-drop the curve response to be transformed from the right side menu into the curve input and choose which axis to reverse.

Curve Reverse Points Set-Up




4.8. Simulation Mid-Run Extraction

NEW as of June 7, 2022: You can now extract stale/delete-and-extract/and extract transformations only. This avoids the need to manually remove old responses for a solving simulation.

Figure 1: Mid-Run Extraction


Here is an example:

Figure 2: Mid-Run Extraction Example


You can also use the refresh button if there are more responses to be extracted.

Figure 3: Refresh button


These options refresh data applied for partially completed simulation.

Figure 4: Refresh data



4.9. Simulation Tracker

If we are writing, editing and polishing simulations, we can create a database for tracking the run log. Let’s review.

Run Log Viewer Options

There are four different ways we can view our run log in the viewer. The following image illustrates these with examples:

Figure 1: Run Log Viewer Options


Ways to Track Runs

There are 3 ways to track runs:

Options A – Self-managed Tracker in Excel (current practice but denormalized) Option B – Managed Manually in d3VIEW using Databases Option C - Insert from a simulation or using a Worker

Let’s go over each.

Option A

For this option, we’ll continue to maintain the Excel but in de-normalized data as shown in the example image below. We’ll then drop the file in the Simlytiks data handler to visualizer.

Figure 1: Option A: Self-managed Tracker in Excel


Option B

For this option, we’ll create a database in d3VIEW and manually add records.

Figure 2: Option B: Managed Manually in d3VIEW using Databases


Let’s review the steps.

Step 1. Navigate to the Databases App


Step 2. Click on New Database


Step 3a. Input a name and some meta-data information


Step 3b. Define Fields by Dragging the inputs from the right menu


Step 3c. Add Users


Step 3d. Save


Step 4. The New Database is now created. Click on the Name to open


Step 5. Click ‘Add Record’ to add entry to the database


Last Step. Visualize records


Option C

For this option, we’ll create a database in d3VIEW and insert from a simulation or use a worker to populate the database as shown in the following image. (This requires simulations to be run and responses be available in d3VIEW).

Figure 3: Option C: Insert from a simulation or using a Worker



4.10. Simulation Sharing

NEW as of March 21, 2022: There is added support to share records with selected teammates.

Figure 1: Share Simulation Records


There is also support to share a public link to anyone even those who do not have a d3VIEW account.

Figure 2: Share Public Simulation Link



4.11. Simulation Comments

Under the Comments tab, add any important notes about the simulation. This is especially useful for team communication. As of March 19, 2022, you can now directly tag other d3VIEW users on your team as well as tag simulation IDs in a comment.

Figure 1: Simulation Comments



4.12. Simulation filters

Physical tests/Simulations can now be ordered randomly using filters.

Random


4.13. Join Split Files

New option available under each simulation in the menu list to Join split files.


For additional questions about how to navigate the d3VIEW platform, please feel free to email our team at: support@d3view.com.