websights

Fill out the form to download

Required field
Required field
Not a valid email address
Required field
Required field
  • Set up your own cloud-native simulation in minutes.

  • Documentation

    Artificial Intelligence (AI) Model Training in SimScale

    SimScale leverages Physics AI in a cloud-native simulation environment to accelerate computational engineering. This documentation explains how to train simulation models using Artificial Intelligence – AI and how it could benefit users and engineering organizations with faster result prediction in seconds instead of actually simulating.

    Consider the following AI trained example model of a Centrifugal pump. Running over hundreds of parallel simulations with geometry variations in the number of blades, blade thickness, etc. and variable inlet flow rates predicted CFD results within seconds for new user inputs with accuracy up to 99% and above.

    ai model training schematic
    Figure 1: AI model training can help reduce result output to seconds.

    There are four major steps involved:

    If your organization has a license for AI training then as an organization admin you can activate AI training permission for your users. This can be done by first accessing the ‘Manage users’ item on your Dashboard and then enabling the feature ‘AI training permission’ for each user individually.

    Once the feature is activated, a new item AI Model Training can be seen in the left panel in the Dashboard.

    ai training permission dashboard manage users
    Figure 2: Steps for the organization admin to activate the AI training permission for the users

    As an organization admin, you can access all the AI models created within your organization. This includes seeing them on the AI Models Training overview page as well as opening and editing each model individually.

    In order to train a new AI model, go to the AI model training page (see left menu in Figure 2) and click on the top right button ‘+ New model’.

    new ai training model
    Figure 3: Creating a new model for AI training

    Set a model name and description.

    new model ai dialog box
    Figure 4: Dialog box for new AI model
    • Model name: Model name is what will be seen later by users of the AI model when they are applying it within the Workbench for fast result prediction.
    • Short description: Short description will later be shown on the model details page and should give a brief overview of the purpose of the model.

    Proceed using the ‘Select Training Data’ button.

    Select projects whose simulation data shall be used for AI training.

    You need at least 20 simulation runs performed in a single project or across multiple projects, however, more data usually leads to higher AI model prediction accuracy.

    Currently, the following analysis types are available for the AI model training:

    The AI training is restricted to using simulation runs from only a single analysis type. If simulation runs from multiple analysis types have been completed in the selected projects, one has to be chosen.

    ai model training data selection
    Figure 5: AI model training data selection. At least 20 runs are required.

    Select ‘Next’ to choose an analysis type.

    choose analysis type dialog box ai model
    Figure 6: Select the analysis type of interest to display the corresponding projects and confirm.

    Select the analysis type of interest to display the corresponding projects. Click ‘Confirm’ to proceed.

    Best Practices on How to Choose a Model for AI Training

    There are mainly two parameters that can be varied:

  • Geometry: This can include dimensions such as inlet length, outlet diameter, number of holes, thickness and number of turbine blades, etc.
  • Boundary conditions: This can include parameterization of simulation quantities such as a range of flowrate velocity, pressure, temperature, angular velocity, force, etc.
  • In this step, some information related to the AI model training needs to be provided before it can be started.

    ai mode training data settings
    Figure 7: Provide AI model training data before beginning the training.

    The required information is as follows:

    • Version Name: Provide a version name to distinguish between different AI model trainings.
    • Confidence Score: A Confidence score from 0-100 shows the trust level users can have with the result prediction. It is recommended to turn it on.
    • Max runtime: Select the time limit beyond which the AI training time will not be exceeded.
    • Start Training: Click here to start the AI model training.

    The training status is displayed as follows:

    ai model training status display
    Figure 8: Users can see the status of their AI model training until it is completed.

    Note

    Currently, SimScale AI models can provide result predictions for all surfaces of the CAD, for e.g. pressure and velocity at inlet outlet and also on walls. If you want to train an AI model to predict also scalar values (like averaged pressures across a surface, drag coefficients, reaction forces, etc.) or restrict the training only to specific regions of the CAD please contact our support for assistance.

    Before the AI model can be released for use by the organization, the following parameters should be thoroughly investigated.

    Here, quality metrics for the quantities involved in the simulation, like velocity, pressure, etc., can be explored.

    model accuracy ai training loss validation loss
    Figure 9: Model accuracy and loss parameters should be thoroughly reviewed to check AI model accuracy.

    MAPEMax

    Mean absolute percentage error maximum. This should be close to zero.

    MAE

    Mean absolute error. This should ideally be zero. These are absolute values with units of the physical quantity involved.

    R2

    R-squared coefficient. A value close to 1 indicates perfect accuracy.

    Let’s understand the metrics using the figure above. The MAEPMax for pressure is 93.63 %; however, for velocity it is under 24.33 %. This means the AI model can be used for predicting velocity values with a lower degree of error.

    The graph shows how relatable the data is. The loss should tend to zero and be significantly reduced on the last iterations compared to the initial value of the residual at iteration 0. One must be cautious not to run the AI training across too many iterations as this could bear the risk of overfitting the model to the training data. This means it would be difficult for the model to generalize beyond the given training data points, even for input data that is not too far from the training data.

    In order to check that the model is not over-fitting to the training data, one must ensure that the training loss and the validation loss are in the same ranges – so the value of the validation loss over the last iterations should be close to the training loss. If the validation loss is much higher than the training loss, this is a strong sign of over-fitting, and a new training should be started with either more training data or training for fewer iterations.

    If you are not satisfied with the training or if the current version did not finish successfully a new version can be trained. Additionally, a document can be created to provide key information for the users of this AI model. These two options are depicted in Figure 10.

    ai demo new version create documentation
    Figure 10: Begin a new version or create a log of steps involved in the current AI model version.

    Once you are satisfied with the AI model training parameters, you have the option to release the trained model to users within your organization.

    release ai trained model
    Figure 11: Release the AI trained model.

    Click on ‘Release’. This opens a dialog box as shown below:

    Dialog box for the release version settings
    Figure 12: Dialog box for the release version settings
    • Description of changes: Mention the changes performed for this version of AI training.
    • Internal Distribution: Decide who should have access to the trained model. There are two options: yourself or your organization. Access to your organization means every user of your organization will have access to the model from the Workbench and can use it to predict simulation results.

    Click ‘Release’ again and a message that the version has been successfully released will be displayed.

    Once the trained model version is released to Workbench the AI Model Training page shows the release status and the type of internal distribution.

    release history ai
    Figure 13: AI model training page after the release. All the related history is available.
    • Released to Workbench: This shows versions released to Workbench.
    • All versions: This shows all unreleased versions.
    • Release history: It displays past information related to version releases such as time of release, name, and description of changes.

    Last updated: March 28th, 2025