Measurement and classification platform "MK4" offers process and data analytics for high-performance photovoltaic production
Automated and measurement-based process control as a new path to a photovoltaic industry 4.0: The Fraunhofer Center for Silicon Photovoltaics CSP aims to drive technological developments in international photovoltaics for competitive production processes from Germany with an industry-oriented platform that offers the possibility of interplay between statistical quality control and new approaches to data-based process analysis.
With an annual production of more than 100 GWp, which corresponds to more than 25 trillion solar cells, the photovoltaic industry has developed into a high-throughput technology with a high degree of automation. However, such state-of-the-art technologies and trends in the photovoltaic industry pose special challenges along the entire manufacturing chain. Cycle rates of less than one second in the production of highly efficient and energy yield-stable solar cells require new technological solutions both in inline measurement technology and in the automated evaluation of the huge amounts of data generated in the process. Thus, the need for metrological monitoring of key parameters in the production process is growing in order to be able to identify weak points and quality deviations as early as possible. Inline measurement systems and statistical analyses of measurement parameters and process data already offer manufacturers, suppliers and machine builders opportunities to improve product quality. What is currently missing, however, is the consistent implementation of automated, digitally networked and measurement data-based process control and monitoring within the production chain.
This is where Fraunhofer CSP, together with its project partners WAVELABS Solar Metrology Systems GmbH, HEGLA boraident GmbH & Co. KG, ASYS Automatisierungssysteme GmbH and CE Cell Engineering Gmbh, starts the development and expansion of the "MK4" platform. As part of a two-year project, the photovoltaics 4.0-capable measurement and classification platform was developed for working on current research and development topics in the areas of correlative data analysis, machine learning and adaptive control processes in order to achieve process and method development under near-production conditions.
Three other currently ongoing research projects, which are being worked on in connection with the platform project, focus on a faster transfer of new process control methods from the laboratory to production. This includes the integration of flexible data interfaces and server structures for the development of an Industry 4.0 as well as data analysis and control algorithms. "In particular, we bring our excellent competencies in metrology, correlative data analysis for adaptive process control using machine learning and evaluation of new inline metrology applications under industrial conditions to the joint projects with our industrial customers," says Dr. Marko Turek, team leader in the "Diagnostics and Metrology" group at Fraunhofer CSP.
"As an industry-oriented platform, the "MK4" measurement and classification platform is open for cooperation and joint research projects. I am thinking here specifically of collaboration with metrology manufacturers and software providers as well as operators of manufacturing facilities. The platform also offers attractive opportunities for integrating new applications for existing inline tools, such as new measurement recipes or correlative analyses," adds Turek.