Senvol ML™

Machine learning software for additive manufacturing

Machine learning software specifically for additive manufacturing.

Senvol ML™ can be used to analyze data from any AM machine, any AM material, and any AM process in order to:

  • Rapidly optimize parameters & develop materials
  • Support qualification of AM machines & materials
  • Predict material properties
  • Minimize data generation costs

Senvol ML™ Features

k

Design experiments more efficiently

The software combines machine learning with traditional design of experiment approaches to intelligently recommend what additional data points to generate. This improves prediction accuracy (i.e. the user is guided to generate smaller, targeted data sets), thereby saving time and money.

Build predictive models in seconds

Senvol ML™ can be used to analyze all major types of AM data. This includes (but is not limited to) process parameter data, process signature data (i.e. from in-situ monitoring sensors), material property data, and mechanical performance data. The software’s algorithms use this data to generate predictive models that help organizations quickly characterize or qualify AM materials and processes.

+

Conduct forward and inverse predictions

Predict mechanical performance (e.g. yield strength) from a given set of process parameters. Inverse predictions are also possible. For example, when given a target mechanical performance (e.g. a target ultimate tensile strength), the software determines what process parameters to use in order to achieve the target.

Determine which inputs actually matter

Whether you are analyzing 3 inputs or 300 inputs, you may not initially know which inputs influence your outputs, and to what degree. The software’s capabilities enable you to ascertain which inputs actually matter, allowing you to focus your development efforts.

Leverage pre-existing data

The software’s Transfer Learning capability enables data from prior machines and prior materials to be used to help predict performance on a new machine and/or new material. This significantly reduces the time and cost of process development or qualifying a new AM machine or material.

t

Assess how much data to generate

No more guessing how many builds to do or how many specimens to fabricate. Senvol ML™ contains capabilities to guide the user on how much data to generate to achieve their desired confidence level.

Example use cases of Senvol ML™

U.S. Air Force

FlexSpecs program focused on qualifying the EOS M400-4

Read more

U.S. Army

Developed a qualification plan and statistically-based material property predictions

Read more

U.S. Navy

Developed capabilities to analyze data such as process parameters, process signatures, material properties, and mechanical performance

Read more

Materialise / Rosswag

Process optimization collaboration with Materialise, Rosswag, and Zeiss

Read more

BMW

Senvol ML™ optimized process parameters outperformed the conventionally optimized process parameters

Read more

Advanced Materials and Processes Magazine

Cover story detailed the use of Senvol ML™ to predict material allowable values

Read more

Ready to assess the Senvol ML™ software?