Today, the MALMO project (Tecniospring Industry) has officially finished. During the last two years, MALMO, led by Dr Ignacio Becerril-Romero and supervised by Dr Victor Izquierdo-Roca from the Solar Energy Materials and Systems group at IREC, has worked intensely on developing advanced methodologies using Artificial Intelligence (AI) and spectroscopic techniques for the monitoring of the industrial production of thin film photovoltaic (PV) technologies.
Thin film PV devices represent a step forward for solar energy as they expand the niches of application of PV and can be mass-produced in a cost-effective way. However, their industrial production is highly complex and small manufacturing deviations result in low quality defective products and, in turn, in a waste of high value materials, energy and time for photovoltaic manufacturers. In this context, MALMO has developed new methodologies that can minimise this issue enabling an early detection of deviations during manufacturing using fast, non-invasive and non-destructive spectroscopic inspection techniques in combination with advance data analysis and AI. In particular, the methodologies have been developed taking as application case a spectroscopic inspection platform in an existing industrial manufacturing pilot line from the Austrian company SUNPLUGGED GmbH, which produces an innovative flexible and customizable solar foil.
During the first 12 months, MALMO, through the secondment of Dr Becerril-Romero at SUNPLUGGED, focused on analysing the real industrial needs of the company for process monitoring and on the fabrication of reference samples through working side-by-side with the plant operators, acquiring knowledge in critical aspects of SUNPLUGGED´s industrial pilot line and learning about the technologic and commercial strategies of the company.
The last 12 months of the project were performed at IREC. Firstly, a large database that included spectroscopic data from the PV materials and related them to their final photovoltaic performance was created using a combinatorial characterization approach. This database was then employed for developing different analytical-statistical and machine learning (AI) methodologies that allow evaluating the quality of the production at an intermediate production stage. In addition, a laboratory inspection system was designed, built and employed to successfully validate the methodologies in simulated industrial conditions.
The methodologies developed in MALMO have proven to have an accuracy ~90% for predicting the production quality. If implemented in the production line of SUNPLUGGED, this could increase production yield by more than 10% and increase the associated sells revenue by, roughly, 1M€/year. Furthermore, the project made thorough interactions with other industrial and research actors to evaluate the applicability and commerciality of the technology in other application cases. Overall, the methods developed in MALMO were found interesting by the thin film PV industry which sees a high potential for the improvement of their production.
Thanks to high innovation degree of the project and the work carried out, Dr Victor Izquierdo-Roca from the Solar Energy Materials and Systems group of IREC is currently coordinating a 10M€ project called Platform-ZERO (www.platform-zero-project.eu), in which the process monitoring methodologies developed in MALMO will be further development and implemented inside a holistic AI-driven process monitoring platform that will be demonstrated in 4 pilots. As such, MALMO can be considered a success whose legacy will contribute to the improvement of EU’s photovoltaic industry.
This project has received funding from the European Union’s Horizon 2020 research and innovation programme under Marie Skłodowska-Curie grant agreement No. 801342 (Tecniospring INDUSTRY) and the Government of Catalonia’s Agency for Business Competitiveness (ACCIÓ).
