The SEGIWA (series production of electrolyzers in the gigawatt range) joint projects objective is to develop fundamentals in order to transfer the Silyzer® 300 technology platform, with a plant consisting of individual modules of 0.75 MW each with a total output of 17.5 MW and a hydrogen production of 340 kg/h (4000Nm³/h), from manual assembly to automated series production with the highest quality standard, utilizing the possibilities offered by virtualisation and digitalisation.
The aim is to span the entire production chain, starting with materials and semi-finished products, the manufacturing processes required to produce the central membrane electrode assembly (MEA), up to module assembly, including the necessary quality assurance measures, and to transfer this to fully automated series production in the gigawatt range. The research project therefore focuses on different automation solutions along the entire production process as well as the development of a stepwise scalable production and factory concept to enable economic and efficient electrolyser production. In addition, the necessary material flows for the series production of PEM electrolysers in the GW range are warranted. For this purpose, process steps are adapted to the new waste streams and bottlenecks in precious metal refinement as well as in the subsequent catalyst production will be eliminated.
Together with the partners, OFFIS is developing a Digital Twin of the series-produced water electrolysis system fleet. With its help, it should be possible to continuously analyse the operating data of the numerous customer plants created as part of series production, and in this way establish a feedback loop from the field into production as well as into development and engineering. The functions of the Twin include process analysis/diagnosis, ageing modelling, production improvement, operational planning, optimised maintenance concepts (predictive maintenance), an interface between field and development, and quotation management. The focus of OFFIS is on the following topics: first, on the conception and evaluation of designs for the different functionalities. Second, on the linking of individual plant data from manufacturing and operational use. Third, on the ongoing individual calibration of combined physico-chemical and data-driven plant models, taking into account ageing effects and the abstraction of general fleet characteristics from direct data analyses individual plant characterisations. Fourth, on developing and evaluating methodological approaches for deriving recommendations for manufacturing improvements and fifth, on testing analogue approaches as a basis for maintenance recommendations and for generating relevant information for supply management.
Nils Huxoll, Sharaf Alsharif, Jan C Diettrich, Tobias Brandt and Michael Brand; Abstracts of the 11th DACH+ Conference on Energy Informatics - Volume 5, Supplement 2; 0September / 2022