Oops! It appears that you have disabled your Javascript. In order for you to see this page as it is meant to appear, we ask that you please re-enable your Javascript!

Senvol ML

Data-driven machine learning software for additive manufacturing

Data-driven machine learning software to analyze the relationships between additive manufacturing process parameters and material performance.

The Senvol ML software suite helps companies quickly characterize or qualify additive manufacturing materials and processes. Senvol ML assists in developing statistically substantiated material properties in order to reduce conventional material characterization and testing that is needed to develop design allowables.

Senvol ML’s capabilities will allow users to select the appropriate process parameters on a particular additive manufacturing machine given a target mechanical performance. This presents a unique opportunity to reduce the high level of trial and error that is currently required, which saves a tremendous amount of time and money.

Data-Driven Machine Learning Algorithm

A modularized ICME (integrated computational materials engineering) probabilistic framework for AM data serves as the foundation for the Senvol ML software. In this framework, AM data is categorized into four modules: Process parameters, process signatures, material properties, and mechanical performance. The software being developed is powered by an algorithm that quantifies the relationships between the four modules. The algorithm is AM material, machine, and process agnostic.

Capabilities of the Senvol ML data-driven machine learning algorithm will include:

  • Forward Prediction: Predict mechanical performance (e.g. fatigue life) from a given set of process parameters
  • Inversion: When given a target mechanical performance (e.g. a target tensile strength), the algorithm determines what process parameters to use in order to achieve the target
  • Machine Learning: The algorithm “learns” from previous data sets and applies those “learnings” to new data sets, thereby reducing the amount of data needed in the future and improving prediction accuracy
  • Recommended Data Collection: The algorithm recommends to the user what additional data points are needed to improve prediction accuracy (i.e. the user is guided to generate smaller, targeted data sets), thereby saving time and money

Computer Vision Algorithm
In addition to Senvol ML’s machine learning capabilities, Senvol ML also includes a computer vision algorithm that analyzes, in real-time, in-situ monitoring data. This enables a user to detect irregularities in real-time and begin to quantify the relationships between irregularities in the build and the resulting mechanical performance.

Senvol ML Development
The Senvol ML software is currently under development and will be made commercially available to any company looking to qualify AM parts.

To read about the U.S. Navy’s use of Senvol ML, click here.

If your company is interested in potentially gaining beta access to the software, you can contact Senvol at info@senvol.com for more details.