Intellegens and Ansys Partner to Empower 3D Printing with Deep Learning

Machine learning solutions company Intellegens announced a collaboration with engineering simulation leader Ansys to accelerate the development of reliable and repeatable additive manufacturing (AM) processes. The integration of machine learning methods is expected to accelerate AM workflows. Combining the two companies’ technologies will make it quick and easy for AM project teams to analyze data from experiments, simulations, or production generating models that capture vital insights. These models optimize process parameters and powders, improving AM parts’ quality while cutting time to market.

A University of Cambridge spin-out, Intellegens provides a unique machine learning toolset that can train deep neural networks from fragmented, corrupt, or noisy datasets. Its first commercially available product, the trade-marked Alchemite, is a deep learning platform that offers accurate models for predicting missing values, finding errors, and optimizing target properties. Created at Cambridge’s world-famous Cavendish Laboratory, Alchemite has demonstrated to be very helpful for AM, offering scientists and engineers a tool to get more from their AM project data. It also enables users to break through data analysis bottlenecks, reducing the amount of time and money spent on research and supporting better, faster decision-making.

A key focus for Alchemite has been engaging problems in discovery and development, particularly where there is a need to focus or apply results from experimental programs to innovate faster and at a lower cost. Intellegens’ deep learning technology can be embedded in third-party platforms to help with problems related to data that is sparse (with many empty values) and noisy (with a large amount of additional meaningless information).

Cavendish Laboratory, University of Cambridge. Image courtesy of the University of Cambridge.

After successfully using its algorithms on aerospace applications, formulation design, chemicals, drug discovery, materials, batteries, and optimizing manufacturing processes, the company began working with Ansys to integrate Alchemite tools within Granta Material Intelligence (MI), Ansys’ AM data management solution. This new collaboration will empower additive technologies with material intelligence through machine learning.

According to the Cambridge-based development team, Alchemite deep learning algorithms very rapidly find relationships within complex datasets, even when it is sparse. The technology makes Alchemite ideal for AM teams seeking to exploit data brought together from multiple sources. It is expected to extract all possible knowledge from the data to identify the critical combinations of factors that ultimately control AM parts’ performance.

Improving additive manufacturing with deep learning. Image courtesy of Intellegens.

Excited about the new collaboration with Ansys, Intellegens Co-Founder, and CEO Ben Pellegrini said in a LinkedIn post that he was looking forward to embedding Alchemite into Ansys AM workflows to deliver better quality parts and processes quicker. Similarly, Ansys’ Product Manager Sakthivel Arumugam anticipates that the new partnership’s outcome will lead to increased adoption and use of AM within companies.

Alchemite needs no prior knowledge of the parameters that are likely to be necessary, which the company claims is a significant advantage in this emerging technology area. Applications throughout the AM workflow include process parameter optimization, computational design of AM materials, failure analysis and quality control, data validation and gap-filling, and assisted Design of Experiments (DoE) for AM.

“Intellegens’ machine learning technology offers a ready-made solution to key data analysis challenges faced by our Additive Manufacturing customers,” commented Rob Davis, Director of Product Management at Ansys. “Integration with Ansys Granta MI creates a unified workflow for capturing and applying results from AM testing, simulation, and production.”

Overview of the Granta MI Additive Manufacturing Package. Image courtesy of Granta Design.

Like Intellegens, Granta MI originated from a University of Cambridge spin-out company founded in 1994 called Granta Design. Now a subsidiary of Ansys, Granta has remained focused on offering support in materials information management for engineering enterprises. The company’s Granta MI software provides a comprehensive materials data management system that enables engineers to swiftly select materials, reduce errors and drive team collaboration.

“Granta MI is the de facto standard for materials data management in engineering enterprises and is applied in AM applications to capture, in a single place, all of a company’s AM data,” suggests the company. This includes data on the properties of powders and raw materials, machine build parameters, post-build processing data, test results for AM parts, and simulation data from the Ansys AM simulation suite.

Moreover, integrating Alchemite into this holistic system “will make it straightforward to analyze the full range of this data in the search for key process and property relationships and to continuously improve models as the data is updated,” the company indicated. Intellegens Chief Technology Officer (CTO), Gareth Conduit, described that “merging the data management capabilities of Ansys’ Granta MI with the machine learning prowess of Alchemite is a perfect fit,” promising to deliver deep insights to AM workflows.

In recent years, data-driven machine learning software initiatives for AM have garnered a lot of attention. Targeting easier processes and more accurate results, machine learning technologies help accelerate application developments. Alchemite’s deep learning algorithms can see correlations between all available parameters resulting in accurate models that can unravel data problems that are not accessible to traditional deep learning approaches. These unique deep learning solutions that extract valuable information from existing processes and data can become game-changers for the 3D printing community.

Leave a Reply

Your email address will not be published. Required fields are marked *