Artificial intelligence in production, Fraunhofer ILT uses AI to significantly improve the results of metal 3D printing

Today, we are accustomed to seeing AI take over more and more tasks-not only in our daily lives, but also in medical applications or industrial production. The development of artificial intelligence has made great progress. It is now possible to use artificial intelligence to predict component failures in production or extract information from images to perform interference tasks in a fraction of a second.

Additive Manufacturing-3D printingAs a typical digital manufacturing technology, massive amounts of data run through the entire workflow from design to production to quality assurance (QA). The complexity of additive manufacturing design is interdependent with many factors such as materials, production parameters, quality requirements, etc. In the face of huge complexity, it is difficult for human experience to stimulate the potential of additive manufacturing technology, making it a kind of production Manufacturing Technology. These characteristics just provide the application soil for artificial intelligence.

This issue, combined with3D printingIn the field, the Fraunhofer ILT in the Fraunhofer Institute for Laser Technology has made progress in the application of artificial intelligence in laser processing, providing insights into how the development of complex artificial intelligence is driven by deep digital twins3D printingEntering production, let’s appreciate the new atmosphere of artificial intelligence empowered additive manufacturing in progress.

Machine learning allows to understand complex data from different sensors and use it for process control
© Aachen Fraunhofer ILT

AI empowerment process adaptive

The Fraunhofer Institute for Laser Technology ILT is the “heart and brain” of laser manufacturing technology worldwide. The Fraunhofer Institute for Laser Technology ILT will hold the second “Laser Technology Artificial Intelligence Conference” from September 28 to 29, 2021. Participants will discuss how artificial intelligence is currently used in laser material processing. Researchers and factory engineers, software developers and machine manufacturers will meet here to exchange ideas. In addition to the technical demonstration, the Fraunhofer Institute for Laser Technology ILT laboratory will also be open for virtual tours.

At the first “AI for Laser Technology” conference two years ago, the question still being discussed was where artificial intelligence (AI) can be used in manufacturing. At the same time, the development of artificial intelligence is very rapid, and many ideas have reached the factory floor.

Various trends converge: For example, manufacturers require 100% quality control. If in the past it was only possible to inspect welds in automobile production through random samples, artificial intelligence can now monitor the quality of each individual part with high precision and quickly identify possible production failures. This requires online process diagnosis developed in recent years. Real-time processing of the resulting large amounts of data is the field of artificial intelligence, which can only be possible through modern computing technology.

Capture complex data and use it for process control

There are many ways to use AI in manufacturing, usually starting with analyzing images or other data. With artificial intelligence algorithms given by human “teachers”, artificial intelligence can even identify structures in complex data. Therefore, it can detect deviations from the pre-defined optimal value at an early stage, so that the process can be adjusted. When data recording and processing are combined with process control to form an autonomous process, it reaches the highest level: intelligently adjust processing strategies.

exist3D printingOn the one hand, Fraunhofer ILT is currently able to significantly improve the results of metal 3D printing through AI. In the Laser Powder Bed Selective Metal Melting (LPBF) process system, a high-resolution HDR camera is used to take pictures of the surface of the components in each layer. Image data can capture two effects: on the one hand, it can measure the possible warpage of the component during the process; on the other hand, it can carefully check the surface roughness. Therefore, defects can be classified during the production process.

Artificial intelligence can control manufacturing processes such as laser welding in real time
© Aachen Fraunhofer ILT

With the help of artificial intelligence, laser parameters can also be specifically changed during the process in order to dynamically react to changes in the process state. This improves the quality of parts and prevents defects before they occur.

Correspondingly, this year’s “AI for Laser Technology Conference”-laser technology artificial intelligence conference held by the Fraunhofer ILT will focus on the acquisition and processing of laser welding and other production process data. A key goal here is end-to-end process and quality control.

The second focus is the control process based on artificial intelligence. In addition, the meeting will also discuss the development of artificial intelligence software in various applications. Here, artificial intelligence not only enables users to optimize production processes and achieve zero-defect production. In processes with large amounts of complex data, such as the development of modern optics, artificial intelligence has also reduced the complexity. The development process becomes clearer, more certain, and less dependent on the intuition of individual experts.

Unlock3D printingPotential

A typical additive manufacturing process is usually composed of thousands of layers, and each layer generates several MB of rasterization or time series monitoring and sensor data to generate technical data packages for common production scenarios. In terms of time saving, repeatability and data efficiency, the benefits of using data to guide processing operations will be huge.

So far, classical data analysis using image and signal processing techniques can also work, in fact they are widely used for this task. However, because different part geometries and process conditions in AM additive manufacturing equipment can cause so many process interdependencies, it turns out that it is almost impossible to find the correct parameterization of traditional analysis algorithms to provide “correct” instructions to the equipment of.

In the book “Design for Additive Manufacturing (DfAM) Guide”, the Ishikawa diagram of the factors affecting the quality of AM parts is cited. In the Ishikawa diagram, more than 160 factors that affect the quality of the processing are listed in detail. It is only the laser scanning process. , Including scanning line length, scanning line type, outer contour, inner contour, scanning mode, scanning speed, beam correction, shrinkage compensation, scanning line sequence, filling pitch, filling direction, laser power, (off) focus, surface filling Parameters, offsets, etc. It can be seen that it is very difficult to control and balance more than 160 variables that affect processing quality through human experience.

Offline CT testing will not only increase the overall cost, but also limit the geometry, because the part must have the appropriate shape to be scanned and tested. If offline monitoring is replaced by intelligent in-process monitoring and testing, this opens up new space and may reduce overall costs.

In terms of commercialization, a process monitoring system with integrated AI will support this transition and realize a direct method that evolves from full-detail testing to intelligent testing. According to market observations from 3D Science Valley, international commercial companies that use AI for quality control of additive manufacturing currently include printsyst in Israel, addiguru in the United States, nebumind in Germany, and Nnaisense in Switzerland.

On the research side, the ACAM Aachen Additive Manufacturing Center in Aachen, Germany, through the close cooperation of its R&D consortium, the ACAM Aachen Additive Manufacturing Center puts the research dimension on the promotion3D printingIn terms of the goal of mass production, research focuses include functional integration, multi-material printing, next-generation equipment and material development, automation, sustainable development, mass execution, artificial intelligence in additive manufacturing, and 5G empowerment adaptive self-evolution Manufacturing, collaborative production, digital management of additive manufacturing processes, etc.

(Editor in charge: admin)