The SMART-Growth project aims to apply Artificial Intelligence and Cyber Physical System-based observation and control for laser crystal growing processes and equipment that provide multi-parametric process data sourcing from a sensor network around the process
Challenges and Opportunities in AI-Driven Crystal Growth
Crystal growing, a high temperature (>1000° C) process, creates high-value components mainly for the photonics industry. Many high-performant laser setups rely on the specific properties of nonlinear crystal materials such as Neodym-YAG or Alexandrite.
Although crystal growing is known for centuries, the crystal growing process remains to be a sophisticated, long term (days to weeks), energy-consuming (kW to MW) and sensitive process that relies heavily on the long-term experience of skilled workers and scientists.
This poses specific challenges for the use of Artificial Intelligence to improv the crystal growing process, as infrastructure, equipment and processes are highly individualized to meet specific demands of the crystalline material to be grown.
However, the potentials for process improvement regarding energy saving, wasting less rare materials, and increasing the yield for the crystal growing process, are high.
The EU co-funded project SMART-Growth, established within the I3 framework of the funding agency EISMEA, aims for establishing a multi-parameter sensor network around a Czochralski-based crystal growing equipment, along with data connectivity to a Cyber-Physical System (CPS) that observes the crystal growing process, learns about if over the course of many long-term processes, and enables a decision support for the manual control of the growing process in the beginning.
The future vision for the sensor-network based CPS is of course to take over crucial process control steps, based on its generated long-term knowledge about process specific parameters and multi-parametric information.
Consortium Expertise Bridging Crystal Growth, AI, and Sensor Technologies
The SMART-Growth project resembles core crystal growing know-how, AI- and CPS-based process control, and multi-parametric sensor networks from research institutions (FVB-IKZ and Fraunhofer IOF from Germany, Western University of Timisoara from Romania, and the POLIMI from Italy), SME that cover the crystal growing process infrastructure and know-how (FILAR Optomaterials from Italy) and the hardware and software-based data connectivity (CiboLabs and Holonix from Italy), while the Italian industrial organization AFIL supports the dissemination and exploitation of the project’s results.