Periodic Reporting for period 1 - TURBO (Towards tURbine Blade production with zero waste)
Período documentado: 2022-10-01 hasta 2024-03-31
Wind turbines are already part of everyday European life and are an essential part of the strategy to meet the Green Deal targets. Almost 3500 turbines (>10000 blades) were installed in 2019 alone . Wind turbine blade (WTB) size is rapidly increasing with new offshore blades >100 m in length. Yet remarkably, the technologies used to manufacture blades have not changed significantly since the late 1970s. Composite blades are manufactured using resin infusion and coating processes which are prone to defects resulting in high rates of re-work, scrap and repair. Even in state-of-the-art factories, non-destructive testing (NDT) solutions are employed only after the blade casting process is completed. This dictates a reactive approach to rectifying defects which almost always results in glass lamination repairs after identification. As a result, WTB repairs generated 1.5 Mt of unrecyclable scrap in 2020 and is on track to reach 3 Mt p.a. by 2030 .
TURB0 aims to combine several new developments from different disciplines to greatly improve the sustainability of WTB production by reducing defect formation and improving repair strategies in composites and coatings:
• Simulation of production process to minimise defect formation: Combining the latest multi-physics process modelling with reduced order models to improve understanding of the composite manufacturing process.
o Fundamentally improved infusion process design to reduce defect formation.
o Training data and strategies for machine learning (ML) algorithms to optimise processes in real-time.
• In-line NDT to monitor WTB composite infusion: TURB0 will combine three cutting-edge NDT technologies (dielectric, wireless, sensor-less) for the first large scale in-line in situ composite production monitoring.
• Sub-surface WTB coating inspection: Combining ultrasound (US) for deep penetration and mid-infrared (MIR) optical coherence tomography (OCT) to image the critical upper layers will allow the most detailed assessment of coating defects ever performed.
• Digital twin for WTB production and data warehouse: The production equipment, monitoring sensor and in-line NDT data will be combined to establish a digital twin for real-time analysis of production. This approach has never been attempted for a large-scale composite part. This digital twin will:
o Provide input signals to the real-time control system driven by a ML algorithm.
o Populate a “data warehouse” accessed and fed by multiple sites for ongoing algorithm training.
• ML-based in-line system control: Through analysis of the digital twin, the ML-based system will provide closed-loop process control to minimise defect formation and waste and optimise process efficiency.
• Automated repair strategy: The digital twin will enable a ML-based analysis of defect severity in composites, with improved automated strategies to reduce repairs by 90 % and increased recycling of off-cuts.
• Full-scale demo: TURB0 fabrication of an >80 m WTB section will be performed at the SGRE Aalborg factory.
• Quantified sustainability and circular economy improvements: The TURB0 workplan includes full lifecycle analysis (LCA), environmental assessments and production efficiency analysis to quantify the project benefits.
Dissemination and exploitation: An extensive dissemination and exploitation plan is in place including skills development, training, interaction with standards bodies and a business plan
TURB0 aims to combine several new developments from different disciplines to greatly improve the sustainability of WTB production by reducing defect formation and improving repair strategies in composites and coatings:
• Simulation of production process to minimise defect formation: Combining the latest multi-physics process modelling with reduced order models to improve understanding of the composite manufacturing process.
o Fundamentally improved infusion process design to reduce defect formation.
o Training data and strategies for machine learning (ML) algorithms to optimise processes in real-time.
• In-line NDT to monitor WTB composite infusion: TURB0 will combine three cutting-edge NDT technologies (dielectric, wireless, sensor-less) for the first large scale in-line in situ composite production monitoring.
• Sub-surface WTB coating inspection: Combining ultrasound (US) for deep penetration and mid-infrared (MIR) optical coherence tomography (OCT) to image the critical upper layers will allow the most detailed assessment of coating defects ever performed.
• Digital twin for WTB production and data warehouse: The production equipment, monitoring sensor and in-line NDT data will be combined to establish a digital twin for real-time analysis of production. This approach has never been attempted for a large-scale composite part. This digital twin will:
o Provide input signals to the real-time control system driven by a ML algorithm.
o Populate a “data warehouse” accessed and fed by multiple sites for ongoing algorithm training.
• ML-based in-line system control: Through analysis of the digital twin, the ML-based system will provide closed-loop process control to minimise defect formation and waste and optimise process efficiency.
• Automated repair strategy: The digital twin will enable a ML-based analysis of defect severity in composites, with improved automated strategies to reduce repairs by 90 % and increased recycling of off-cuts.
• Full-scale demo: TURB0 fabrication of an >80 m WTB section will be performed at the SGRE Aalborg factory.
• Quantified sustainability and circular economy improvements: The TURB0 workplan includes full lifecycle analysis (LCA), environmental assessments and production efficiency analysis to quantify the project benefits.
Dissemination and exploitation: An extensive dissemination and exploitation plan is in place including skills development, training, interaction with standards bodies and a business plan
The requirements definition was successfully completed and proved very important in setting the technical targets for the various TURBO technologies, but also the technical steps needed to achieve the overall project objectives.
The infusion simulation work is progressing as planned, with infusion models successfully adapted to show defect formation.
Experimental trials of the Synthesites sensors have begun at SGRE, DTU-W and NCC, and the CPI wireless sensor is in production. Trials of both sensor types have been carried out at NCC.
NORBLIS technology development has shown for the first time, that defects in WTB coatings may be imaged with MIR OCT and that the MIR can penetrate the coating layer, top coating and primer coating.
Using publicly available data and simulations of infusion from other TURBO partners, UPV has established the database in preparation for the real data which will come later in the project. This has been used to design the ML algorithms which will be trained on real data and used for process control later in the project.
The demo work is not due to start in Period 1, but SGRE has planned the necessary machine integration and interfaces, and the preparations for the demo at SGRE (Aalborg, Denmark) are already in progress.
Arditec activities on sustainability are on track and it has delivered its preliminary results on the three pillars (LCA, LCC and S-LCA) on time.
The infusion simulation work is progressing as planned, with infusion models successfully adapted to show defect formation.
Experimental trials of the Synthesites sensors have begun at SGRE, DTU-W and NCC, and the CPI wireless sensor is in production. Trials of both sensor types have been carried out at NCC.
NORBLIS technology development has shown for the first time, that defects in WTB coatings may be imaged with MIR OCT and that the MIR can penetrate the coating layer, top coating and primer coating.
Using publicly available data and simulations of infusion from other TURBO partners, UPV has established the database in preparation for the real data which will come later in the project. This has been used to design the ML algorithms which will be trained on real data and used for process control later in the project.
The demo work is not due to start in Period 1, but SGRE has planned the necessary machine integration and interfaces, and the preparations for the demo at SGRE (Aalborg, Denmark) are already in progress.
Arditec activities on sustainability are on track and it has delivered its preliminary results on the three pillars (LCA, LCC and S-LCA) on time.