3D printing intersects with Machine Learning
Author – Swagata Ashwani, Product Marketing Specialist, 3DParadise
Machine learning, simply put Is the application of advanced algorithms against previously collected data to develop predictions, classifications and decisions about future events. Its applications have now started expanding to diverse industries, the hottest one is additive manufacturing. According to a report from Deloitte, the global 3D printing industry is expected to reach $5.2 billion in 2020. With the industry that has the possibility of creating disruptions in the manufacturing industry, The software is invading and playing a vital part in its ever-expanding growth. Additive manufacturing involves numerous and complex variables to be monitored and controlled in the process to achieve an acceptable level of accuracy in printing. Machine learning is currently being used to provide an optimized way to realize this by using generative design and testing in the pre-fabrication stage, with the aim of improving printing efficiency and cost savings. Artificial intelligence is currently finding applications in 3D printing and additive manufacturing for creating intelligent service-oriented production processes for the industry. This article goes through some of the major advancements occurring in this area of intersection.
Customizable heavy Metal printing
Ever since, MX3D announced a plan to 3D print an entire steel bridge designed by Joris Laarman, the anticipation around this intersection is growing immensely. The company and its chief investor Autodesk agreed to share an exclusive update with Co.Design.
The bridge shown above, is being printed, can still bend and twist fantastically, in a way that could only be done with 3D printing.
The challenge is in printing big pieces. When steel melts, its physical properties change. Constant reheating makes it brittle. Hence, you can’t simply build up a 3D-printed steel structure as you can with plastic, applying one layer at a time. As the successive layers of steel are applied, they reheat the layers below A robot that’s 3D printing with steel does that by constantly spilling acid while orienting a layer. Because the printer is no longer waiting for steel to cool in a particular spot, the printer itself can work twice as quickly. That’s not all. Intricate 3D geometries are very different from one another, so it’s hard to know in advance where the machine will have trouble. This is where machine learning can be useful. The industrial robots have sensors that detect how much current is being used to heat up the metal, how hot that metal gets, and where exactly the welds are being applied them all together, coming up with new patterns of movement that allow each layer to build up properly.
Agile Metal Technology Software by Sculpteo
Sculpteo is a french company which specializes provides online 3D printing services for prototyping and manufacturing. The Business Case tool takes inputs from users who want to design and manufacture a particular part in the form of CAD files and desired output parameters and uses machine learning to evaluate if additive manufacturing would be the appropriate process for producing the part. The tool can also assist in optimizing lattices in the CAD files and evaluate the most efficient printing paths.
Netfabb 2018 by Autodesk
Autodesk recently launched a new additive manufacturing software product as a part of their Netfabb 2018 software suite. Autodesk’s Netfabb’s additive manufacturing software claims to use machine learning to generate and evaluate digital models for industrial 3D printing production.
ADAPT Center at the Colorado School of Mines
A graphic interface draws on the findings of machine learning so far. Each coordinate position represents a different combination of build parameters. The dots’ different colors correspond to different predicted maximum defect sizes for that parameter set. The presence of a dot indicates that the model can make a confident prediction in that area. More data resulting from more builds will be necessary to add dots to this grid, not to mention to expand the array of predictive grids like this.
Machine Learning and 3d Printing for Spare Part Manufacturing Industry
In a discrete manufacturing setting, predictive maintenance models could use ML to accurately predict the remaining lifetime of specifics part or pieces of equipment. In an additional layer, machine learning could also be used to proactively identify time for part replacements by using predetermined replacement schedule data.
Computer Vision for Defect Detection by GE
General Electric’s GE labs claim to have previously developed computer vision technology that can find microscopic cracks in machine parts and other microscopic aberrations. The Additive Research Lab at GE Global Research claims to enable the 3D printer to perform an inspection of parts after they are completely built in order to improve cost and time savings in the manufacturing industry. GE claims that a proprietary machine-learning platform then matches recorded powder patterns to defects revealed by CT scanners. In essence, the ML platform is trained through the use of high-resolution camera footage and CT scan data and can eventually ‘learn’ to predict problems and detect defects in the printing process.
Additive Manufacturing Leader at GE Global Research claims that it is possible to take defect detection using computer vision and dynamically control 3D printers to compensate for the defects. The longer-term goal for the team at GE Additive is the idea that the 3D printer can create compensation strategies based on what the computer vision data predicts.