Artificial intelligence in surface technology

Pretreatment with plasma technology and decoating with laser techniques are precision processes for preparing component surfaces for further processing in the process chain, e.g. for bonding or painting. The slightest fluctuations in process parameters or starting materials can have serious consequences, even leading to component failure. Surface pretreatment processes must therefore be extremely controlled. To counteract process fluctuations, IFAM is researching into making these processes self-adjusting using optical and acoustic monitoring methods in combination with artificial intelligence (AI) and thus making them safer.

 

Controlling pretreatment processes with AI

Fraunhofer IFAM has many years of experience in the optimization of production processes with automation and digitization. In particular, controlling pretreatment processes with self-learning algorithms is an approach currently being pursued by researchers in the "Plasma Technology and Surfaces" department and successfully transferred to various application areas.

In the first step, characteristic emissions of these processes, actual values of the process parameters and the resulting product properties are recorded. In a second step, these data are linked with existing artificial intelligence algorithms and correlations between these data are learned. Based on the knowledge gained, deviations in the process can be reliably detected in real time and countermeasures can be initiated in order to finally obtain a reproducible surface condition even in the case of variable input variables (e.g. in the case of process instabilities or variable initial conditions).

 

Optical monitoring in laser material processing

Laser decoating is a process in which layers are removed from surfaces using laser technology. This process can be monitored using optical - spectroscopic or imaging - methods. When these are combined with AI methods, high-precision, automated, self-controlled control becomes possible.

When a laser is set according to predetermined parameters (e.g., laser power or distance from the surface), the interaction of the laser beam with the component being treated or the treatment result is recorded. This is supplemented by a precise understanding of the treatment effects. At Fraunhofer IFAM, different methods for in-line characterization (e.g. optical emission) are available.

Subsequently, AI methods are used to learn correlations between the sensor signals and the desired state of the surface. These data index the target state for subsequent treatment processes. Deviations in the treatment process are detected by continuous monitoring of the laser process and adjusted by suitable control technology in the sense of closed-loop control ("closed loop").

 

Optical monitoring of plasma coating

Similarly, plasma coating processes can be monitored in an AI-controlled manner. After the parameters for the ideal plasma are found, its light emission is characterized optically. Characterization of the desired coating thickness is also determined optically, e.g. by interferometry. Using AI methods, correlations between light emission and surface properties can also be established here. After the learning process, a camera is sufficient to monitor and control the plasma process.

 

Acoustic monitoring of tribological parameters

The process of wear or abrasion of a surface can be characterized acoustically. In this case, the noise detection is correlated with tribological parameters such as contact pressure and friction coefficient. AI methods are then used to establish the correlations. The tribological process can then be monitored with a microphone. If acoustic deviations are measured, appropriate corrections are initiated automatically.

 

Research, development, and consulting around the use of AI in pretreatment processes

Fraunhofer IFAM has distinctive expertise in pretreatment processes using laser technology and plasma technology and combines this know-how with knowledge of artificial intelligence to make processes safer and more efficient. Focal points in contract development and consulting are:

• Design of pretreatment processes with plasma and laser technology including integrated, AI-controlled monitoring methods.
• Selection of suitable monitoring methods, plasma sensors and AI methods.
• AI-controlled monitoring of blasting and tribological processes

Dr. Ralph Wilken is head of the department "Plasma Technology and Surfaces" at Fraunhofer IFAM. His team of researchers has extensive knowledge of all aspects of the use of plasma, laser, and VUV technologies in surface engineering. They offer particular expertise in the control and monitoring of pretreatment processes using digital solutions and artificial intelligence.