For many companies, digitization and automation are the key to further developing additive manufacturing. More and more manufacturers are relying on cloud-based solutions and are integrating various algorithms into their 3D printing solutions in order to exploit the full potential of the technology. As a digital process itself, 3D printing is part of Industry 4.0 and thus an important part of an era in which artificial intelligence, such as machine learning, is increasingly being used to optimize the value chain. Artificial intelligence (AI) is able to process a large amount of complex data in a very short time, which is why it is becoming increasingly important as a decision maker. We explain what machine learning is and why this form of AI is helping to shape the future of additive manufacturing.
Machine learning is a sub-category of AI and is defined as a system or software that uses algorithms to examine data and then recognize patterns or determine solutions. Contrary to popular belief that machine learning is a newfangled phenomenon, its beginnings can be said to date back to the 1940s when the first researchers began to recreate the brain’s neurons using electrical circuits. In 1957, the Mark I Perceptron was the first great success in this field: the machine was able to classify input data independently. In doing so, the device learned from the mistakes of previous attempts, which improved the classification over time. Since then, the foundation stone has been laid and the researchers have been fascinated by the possibilities and potential of the technology. We now encounter artificial intelligence every day in all areas of life. From speech recognition to intelligent chatbots to personalized treatment plans, machine learning is used in a variety of applications.
Supervised vs. unsupervised machine learning
Within the machine learning spectrum, it is important to differentiate between different methods and models. Machine learning is not always machine learning. For example, a distinction must be made between monitored and unsupervised machine learning. Supervised machine learning requires that categorized data (input data) and the target variable (output data) are available. The model is derived from this, which then examines (new) uncategorized data and determines the target value for this itself. This form of machine learning is used, for example, for predictions, e.g. for predicting maintenance intervals.
The opposite is the starting point for unsupervised machine learning. The software does not have a target variable (output data), but has to recognize patterns based on the input data or suggest solutions. This type of machine learning is used, among other things, in marketing to identify customer segments, so-called “clustering”. But there are other differences. For example, there is also semi-supervised learning, in which only a small amount of predefined data in a large amount of raw data is used to train the model, and reinforcement learning, in which the system learns itself according to predefined rules. The user must therefore select the appropriate method based on the raw data and the target variables.
How is machine learning used in additive manufacturing?
As a digital production process, additive manufacturing benefits from the possibilities of machine learning. Since innumerable data are collected and processed (in real time) along the additive value chain, they can be used to analyze the actual state and then redefine the target state. It is important for companies to first define which data is relevant at all. This decision depends on the method used. The next step is to find and integrate the appropriate measurement tool to capture the values ââbefore defining a suitable model or algorithm for data capture and processing. In this context, it is also important to understand that all steps along the additive value chain influence each other, which is why an isolated consideration is not useful in most cases. For example, the design already influences the later component quality and the desired component quality influences the design. For this reason, more and more companies are trying to offer a comprehensive software solution with which the advantages of artificial intelligence can be used in the best possible way for the additive manufacturing process.
At the beginning of every 3D-printed component there is a file, usually a CAD file. Here companies can already benefit from artificial intelligence. Most software solutions on the market today already use AI to suggest intelligent design variants to users based on predefined variables. This process is known as generative design, among other things. Machine learning is also used for topology optimization. Many software solutions also make suggestions on manufacturing processes, materials and optimal use of installation space. As a result, costs can be saved and parts can not only be produced more efficiently, but also more sustainably.
If the 3D printable file is already optimized, the focus could instead be on the 3D printing process used, the material quality and the component quality. Today, many manufacturers have already integrated cameras and sensors into their machines that track the pressure and, if necessary, can sound an alarm or stop the pressure. In this step, it is important to know how to define the quality of the part as it is printed in order to be able to define the required measurements. It is also important to define which action the machine should perform at which threshold value. Some algorithms are already able to define these parameters independently and to further develop the model on the basis of data that has already been collected. What this can look like can best be explained with a practical example.
EOS has teamed up with NNAISENSE, a Swiss software provider, to develop a digital twin for the DMLS process. During the printing process, thermal images of each printed layer are recorded using optical tomography (OT) and compared with the image predicted by the AI. Anomalies can be recognized immediately and the printing process can be stopped if necessary, which leads to material and cost savings. The model developed by NNAISENSE is a self-monitored deep learning strategy. Siemens emphasizes that quality assurance in additive manufacturing (AM) with artificial intelligence and machine learning can reduce the time from prototype to finished part and accelerate the efficiency of large-scale production. The company appreciates the camera integrated by EOS for monitoring the individual print layers, as it can detect missing powder on the parts to be printed (left) or powder drops when overcoating (right) in real time.
The quality of each coating is recorded as a numerical value and automatically evaluated. When this so-called severity level reaches a certain threshold, it could indicate a serious problem with the coating (as in the example above). According to the company, this simplifies the optical inspection, as only critical layers need to be assessed by an expert.
AUTOMAT3D, the postprocessing software from PostProcess, monitors important process factors in real time and reacts autonomously in order to achieve the best possible finish of 3D printed parts. To do this, the company uses data from hundreds of thousands of benchmark parts. In addition, AI is increasingly used to automate and optimize work processes. Smart sensors can be found in critical components that are the measuring instrument for intelligent and preventive maintenance or âpredictive maintenanceâ. It is foreseeable that the use of machine learning for manufacturers’ production processes will continue to increase in the coming years. The global artificial intelligence and advanced machine learning market is projected to reach $ 471.39 billion by 2028, with a growth rate (CAGR) of 35.2%.
In your opinion, what potential does machine learning have for use in additive manufacturing? Let us know in a comment below or on our Linkedin, Facebook and. to know Twitter Pages! Don’t forget to sign up for our free weekly newsletter here, the latest 3D printing news delivered to your inbox! You can also find all of our videos on our YouTube channel.
* Cover picture credits: Siemens