Final Report Summary - MARE-WINT (new MAterials and REliability in offshore WINd Turbines technology)
MARE-WINT (new MAterials and REliability in offshore WINd Turbines technology)
MARE-WINT project outcomes have been presented in a published open access book that will be used in universities as training material for courses on wind energy technologies.
MARE-WINT Open Access Book 2016
https://link.springer.com/book/10.1007%2F978-3-319-39095-6
MARE-WINT: New Materials and Reliability in Offshore Wind Turbine Technology
Editors: Wiesław Ostachowicz, Malcolm McGugan, Jens-Uwe Schröder-Hinrichs, Marcin Luczak
ISBN: 978-3-319-39094-9 (Print) 978-3-319-39095-6 (Online)
DOI 10.1007/978-3-319-39095-6
MARE-WINT (new MAterials and REliability in offshore WINd Turbines technology) was a high quality training network which trained 15 well qualified researchers in the area of wind turbine technology. 10 Early Stage Researchers developed their PhD Theses.
Implemented training programme of Early Stage Researchers from industry and academia combined expertise from fluid dynamics, composite material, Computational Fluid Dynamics (CFD), experimental techniques and numerical modelling of coupled multiple simultaneous physical phenomena. Major project objective was transfer of knowledge in the highly multidisciplinary filed of innovative approach for coupled multi physics co-simulation, testing, design and optimization of reliable Offshore Wind Turbines (OWTs). Overall objective of proposed project was transfer of knowledge in the field. The objective has been attained by combining the best available techniques in numerical simulation and experimental validation approach. OWT is a highly multi-disciplinary research field encompassing mechanical engineering, material science, fluid mechanics, harsh environment engineering, condition monitoring, maintenance, safety, experimental and numerical techniques. This enforced a multidisciplinary approach in the investigations.
MARE-WINT training network was closely cooperating with the Innwind.eu project which aims at the high performance innovative design of a beyond-state-of-the-art 10-20MW offshore wind turbine and hardware demonstrators of some of the critical components.
Innwind.eu project has shared the data of the 10 MW reference wind turbine which enables a common platform for the research within MareWint project to work on the same research object.
MARE-WINT Initial Training Network has provided the structured, integrated and multidisciplinary training program for the future OWT technology experts. The Consortium composed of public and private organizations and based on a common research programme, successfully increased the skills exchange between public and private sectors. A research and dissemination programme has been within 6 cross-linked WP. For a management of the ITN project and training organisation dedicated WP7 was defined.
In Work Package 1, Innovative Rotor Blades investigations were focused on the techniques for the damage detection in blades, understanding the mechanisms of different damage modes propagation, and developing solutions for designing more reliable structures using new materials, processes, and measurement systems. Different fracture modes were identified and analysed by means of the experimental and finite element method. Numerical models were validated against the experimental Destructive and Non-Destructive Testing (NDT). For this purpose the Double Cantilever Beam test specimen was developed and measured.
Next to the aero-elastic analyses of the blade were performed. This includes the CFD investigation of the near-blade 3D flow for a complete OWT configuration with elastic blades or flaps. For this purpose, the structural model of the reference 10-MW wind turbine has been produced using NASTRAN. Further, the dynamic representation of the entire wind turbine mounted on the semi-submersible floating support was developed. A static aero-elastic computation was performed at a wind speed of 11m/s. For static analysis, the loads per NASTRAN beam length were extracted at the locations of the grid nodes. The pressure was then integrated along the curve resulting from blade surface plane intersection and loads were extracted in form of forces and moment per element length. Next, the loads were applied in the NASTRAN model and new blade shape was obtained. Finally, the CFD mesh was deformed according to the new blade shape and new loads were calculated, transferred to un-deformed blade shape and extracted to be applied again to the NASTRAN model. This procedure was repeated until convergence in loads and blade shape was obtained. For unsteady aero-elasticity, the strongly-coupled modal approach was adopted, where the blade modes were computed using NASTRAN.
The effects of atmospheric inflow profile and atmospheric turbulence using Mann’s model were also investigated, showing increased load fluctuations for the case with atmospheric turbulence. Also, the aerodynamic performance of a 10MW rotor under pitch and yaw motion was studied. The results showed that depending on the amplitude of pitch motion, a wind turbine may enter turbulent or vortex ring states. Finally, a 10MW rotor was equipped with the trailing and leading edge flaps. Computational results showed potential of such devices to locally change aerodynamic forces. For instance, the trailing edge flap may be used to control flap-wise bending or to eliminate the adverse effect of the blade passing in front of the tower. At the same time, the leading edge flap may be used to counter additional pitching moment created by the trailing edge flap.
Work Package 2 work has been focused on the Drive Train with Gearbox. The gearbox is one of the key subsystems in a geared wind turbine providing the task to transfer power from the low speed shaft connected to the rotor to the high speed shaft connected to the generator. As turbines become larger, more power is demanded and gearboxes with higher load capacity need to be designed. A deep knowledge into gearbox dynamics becomes of fundamental importance and noise and vibration measurements are demanded.
Large scale wind turbine gearbox was investigated by means of numerical and experimental approach. Multibody model of the gearbox was oriented towards the research on extension of the capabilities of the current multibody model towards a generic multi-physical OWT driveline. It was confirmed that component flexibilities play a relevant role in the simulation of different driving scenarios. Within experimental activity framework the extensive experimental campaign was performed on the unique large scale test bench available to MareWint Project from ZF Wind Associated Partner.
Vibration measurements are mainly quality estimation methods for gear mesh vibrations and overall sound power levels. They are based on standard techniques for the estimation of dynamic characteristic in general applications and not focusing on the wind turbine gearbox case. Since several components properties depend on the applied torque and on the rotational shaft speed, a validation in operational conditions needs to be performed. Building on existing techniques such as “Order Tracking” and “Operational Modal Analysis”, a dedicated methodology has been developed for the analysis of operational gearbox dynamic behavior.
Operational Modal Analysis (OMA) is used to derive an experimental dynamics model from vibration measurements in operational conditions. It cannot be applied in a straightforward way due to the self-induced vibrations at several rpm-dependent frequencies (gear meshing orders). These frequencies with high vibration levels can be wrongly considered resonance frequencies of the system. In order to face these problems, an extensive measurement campaign has been performed at ZF Wind Power on a 13.2MW test rig facility. Accelerations have been measured at more than 250 locations on the test rig and for different load levels and operating conditions.
Offshore Support Structure of bottom fixed and floating were subjects of study in Work Package 3. As the result of this research a methods for integrated modelling and analysis of floating offshore wind turbine (FOWT) has been developed. The research involved hydro-elastic-aero-servo analysis of FOWT and therefore dealt with the modelling of aerodynamic and hydrodynamic loads, as well as the structure and mooring systems, drive train systems and automatic control systems.
Work Package 4: Reliability and Predictive Maintenance. Task 4.1 Offshore Wind Turbine condition monitoring based on acoustic emission and long range ultrasonic performed many experimental tests using the Long Range Ultrasonic equipment. For the monitoring of wind turbine blades, a damage detection technique using specific transducers has been developed. The Cross-Correlation technique was used to compare signals between a baseline signal and a damage state signal, in order to detect the position of the damage. This effort was put together with Detection of damage in metallic and composite structures for offshore applications where most of the work was focussed on the development of damage detection strategy for the tower structure of the wind turbine. The proposed methodology for the damage detection was Neutral Axis tracking, using strain sensors along diametrically opposite points of the sensors. For the proof of concept of the proposed methodology, a FE model of the 10MW Reference Wind Turbine was made in FA software followed by the validation of the model against the experimental data. The results indicate that damage can be detected, through the use of NA tracking based on Kalman Filter. Reducing fatigue loads due to wake effects for offshore wind farm was another research focus. To estimate fatigue loads due to wake turbine interaction the development of wind turbines wakes was studied. It was found that a factor that greatly influences wake development is its interaction with the Atmospheric Boundary Layer (ABL). Field data possess high standard deviation and is at times erroneous. Instead, Large Eddy Simulation (LES) has been used to generate data that offers ‘higher fidelity and detail’ for tuning engineering models. ECN has developed an Energy-Conserving Navier-Stokes (ECNS) code that has been verified for laminar flows. It offers benefits like unconditional stability and zero numerical dissipation, which are very important for the LES of wind farms (LES). This code was modified and the LES based analysis of wind farms to suggest improvements for engineering models. Another task focussed on a risk assessment for collisions between vessels and the offshore wind farm. This is an area of considerable interest – particularly as a result of the rapidly increasing number of wind farms near high-density vessel-traffic waterways, such as the North and Baltic Seas. The tools and methods most widely used for risk assessment in the maritime industry has been compared. The various parameters that should be taken into account for wind-farm vessel collision risk assessment have also been evaluated. The goal of this research is to optimize the use of limited sea-space by balancing navigational safety and environmental efficiency requirements. It is anticipated that the research will demonstrate that offshore wind does not conflict with safety requirements - paving the way for further development of this technology.
Work Package 5: Fluid-Structure Interaction
At deep water common fixed-bottom OWTs are found to be too expensive, motivating the so-called floating offshore wind turbines (FOWTs). Due to the violent loading scenario of their location (combining both wind and wave excitations), this kind of machines are subjected to strong dynamical effects. FSI prediction methods were adapted to FOWTs context for rotor analysis. Hydrodynamic and mooring line forces coming from sea-state excitation induce considerable motions in the 6 DOF of the wind turbine tower top. These motions are translated into the rotor blades, amplifying the unsteady aerodynamics and FSI effects. Commercial computational fluid dynamics (CFD) codes from NUMECA International were used in order to model this challenging multi-physics problem. Results were also obtained with in-house tools at the University of Glasgow, showing substantial deformations of the blade with the displacement at the tip up to 10% of the radius. Steady and unsteady FSI simulations of the rotor were performed, where re-adaptation of the fluid domain to the deformations/displacements experienced by the blades was necessary. Next to it within twist-coupled aero-elastic design for passive loads reduction on a full scale blade aimed at optimization of the passive load reduction behaviour through bend-twist coupling of the DTU 10 MW reference wind turbine. A reduction in extreme and fatigue blade root design bending moment, blade mass and foundation loads was aimed for while maintaining power output and stability characteristics of the initial design. Apart from the beam model implementation a parameter study on swept blades (geometrical-coupling) were conducted. Flow control for improving aerodynamic performance and noise reduction can be introduced through active methods. The main objective of the task was to develop open and closed-loop active flow control strategies to: delay flow separation and stall in real time; and optimize the overall efficiency and minimize associated noise.
MARE-WINT project outcomes have been presented in a published open access book that will be used in universities as training material for courses on wind energy technologies.
MARE-WINT Open Access Book 2016
https://link.springer.com/book/10.1007%2F978-3-319-39095-6
MARE-WINT: New Materials and Reliability in Offshore Wind Turbine Technology
Editors: Wiesław Ostachowicz, Malcolm McGugan, Jens-Uwe Schröder-Hinrichs, Marcin Luczak
ISBN: 978-3-319-39094-9 (Print) 978-3-319-39095-6 (Online)
DOI 10.1007/978-3-319-39095-6
MARE-WINT (new MAterials and REliability in offshore WINd Turbines technology) was a high quality training network which trained 15 well qualified researchers in the area of wind turbine technology. 10 Early Stage Researchers developed their PhD Theses.
Implemented training programme of Early Stage Researchers from industry and academia combined expertise from fluid dynamics, composite material, Computational Fluid Dynamics (CFD), experimental techniques and numerical modelling of coupled multiple simultaneous physical phenomena. Major project objective was transfer of knowledge in the highly multidisciplinary filed of innovative approach for coupled multi physics co-simulation, testing, design and optimization of reliable Offshore Wind Turbines (OWTs). Overall objective of proposed project was transfer of knowledge in the field. The objective has been attained by combining the best available techniques in numerical simulation and experimental validation approach. OWT is a highly multi-disciplinary research field encompassing mechanical engineering, material science, fluid mechanics, harsh environment engineering, condition monitoring, maintenance, safety, experimental and numerical techniques. This enforced a multidisciplinary approach in the investigations.
MARE-WINT training network was closely cooperating with the Innwind.eu project which aims at the high performance innovative design of a beyond-state-of-the-art 10-20MW offshore wind turbine and hardware demonstrators of some of the critical components.
Innwind.eu project has shared the data of the 10 MW reference wind turbine which enables a common platform for the research within MareWint project to work on the same research object.
MARE-WINT Initial Training Network has provided the structured, integrated and multidisciplinary training program for the future OWT technology experts. The Consortium composed of public and private organizations and based on a common research programme, successfully increased the skills exchange between public and private sectors. A research and dissemination programme has been within 6 cross-linked WP. For a management of the ITN project and training organisation dedicated WP7 was defined.
In Work Package 1, Innovative Rotor Blades investigations were focused on the techniques for the damage detection in blades, understanding the mechanisms of different damage modes propagation, and developing solutions for designing more reliable structures using new materials, processes, and measurement systems. Different fracture modes were identified and analysed by means of the experimental and finite element method. Numerical models were validated against the experimental Destructive and Non-Destructive Testing (NDT). For this purpose the Double Cantilever Beam test specimen was developed and measured.
Next to the aero-elastic analyses of the blade were performed. This includes the CFD investigation of the near-blade 3D flow for a complete OWT configuration with elastic blades or flaps. For this purpose, the structural model of the reference 10-MW wind turbine has been produced using NASTRAN. Further, the dynamic representation of the entire wind turbine mounted on the semi-submersible floating support was developed. A static aero-elastic computation was performed at a wind speed of 11m/s. For static analysis, the loads per NASTRAN beam length were extracted at the locations of the grid nodes. The pressure was then integrated along the curve resulting from blade surface plane intersection and loads were extracted in form of forces and moment per element length. Next, the loads were applied in the NASTRAN model and new blade shape was obtained. Finally, the CFD mesh was deformed according to the new blade shape and new loads were calculated, transferred to un-deformed blade shape and extracted to be applied again to the NASTRAN model. This procedure was repeated until convergence in loads and blade shape was obtained. For unsteady aero-elasticity, the strongly-coupled modal approach was adopted, where the blade modes were computed using NASTRAN.
The effects of atmospheric inflow profile and atmospheric turbulence using Mann’s model were also investigated, showing increased load fluctuations for the case with atmospheric turbulence. Also, the aerodynamic performance of a 10MW rotor under pitch and yaw motion was studied. The results showed that depending on the amplitude of pitch motion, a wind turbine may enter turbulent or vortex ring states. Finally, a 10MW rotor was equipped with the trailing and leading edge flaps. Computational results showed potential of such devices to locally change aerodynamic forces. For instance, the trailing edge flap may be used to control flap-wise bending or to eliminate the adverse effect of the blade passing in front of the tower. At the same time, the leading edge flap may be used to counter additional pitching moment created by the trailing edge flap.
Work Package 2 work has been focused on the Drive Train with Gearbox. The gearbox is one of the key subsystems in a geared wind turbine providing the task to transfer power from the low speed shaft connected to the rotor to the high speed shaft connected to the generator. As turbines become larger, more power is demanded and gearboxes with higher load capacity need to be designed. A deep knowledge into gearbox dynamics becomes of fundamental importance and noise and vibration measurements are demanded.
Large scale wind turbine gearbox was investigated by means of numerical and experimental approach. Multibody model of the gearbox was oriented towards the research on extension of the capabilities of the current multibody model towards a generic multi-physical OWT driveline. It was confirmed that component flexibilities play a relevant role in the simulation of different driving scenarios. Within experimental activity framework the extensive experimental campaign was performed on the unique large scale test bench available to MareWint Project from ZF Wind Associated Partner.
Vibration measurements are mainly quality estimation methods for gear mesh vibrations and overall sound power levels. They are based on standard techniques for the estimation of dynamic characteristic in general applications and not focusing on the wind turbine gearbox case. Since several components properties depend on the applied torque and on the rotational shaft speed, a validation in operational conditions needs to be performed. Building on existing techniques such as “Order Tracking” and “Operational Modal Analysis”, a dedicated methodology has been developed for the analysis of operational gearbox dynamic behavior.
Operational Modal Analysis (OMA) is used to derive an experimental dynamics model from vibration measurements in operational conditions. It cannot be applied in a straightforward way due to the self-induced vibrations at several rpm-dependent frequencies (gear meshing orders). These frequencies with high vibration levels can be wrongly considered resonance frequencies of the system. In order to face these problems, an extensive measurement campaign has been performed at ZF Wind Power on a 13.2MW test rig facility. Accelerations have been measured at more than 250 locations on the test rig and for different load levels and operating conditions.
Offshore Support Structure of bottom fixed and floating were subjects of study in Work Package 3. As the result of this research a methods for integrated modelling and analysis of floating offshore wind turbine (FOWT) has been developed. The research involved hydro-elastic-aero-servo analysis of FOWT and therefore dealt with the modelling of aerodynamic and hydrodynamic loads, as well as the structure and mooring systems, drive train systems and automatic control systems.
Work Package 4: Reliability and Predictive Maintenance. Task 4.1 Offshore Wind Turbine condition monitoring based on acoustic emission and long range ultrasonic performed many experimental tests using the Long Range Ultrasonic equipment. For the monitoring of wind turbine blades, a damage detection technique using specific transducers has been developed. The Cross-Correlation technique was used to compare signals between a baseline signal and a damage state signal, in order to detect the position of the damage. This effort was put together with Detection of damage in metallic and composite structures for offshore applications where most of the work was focussed on the development of damage detection strategy for the tower structure of the wind turbine. The proposed methodology for the damage detection was Neutral Axis tracking, using strain sensors along diametrically opposite points of the sensors. For the proof of concept of the proposed methodology, a FE model of the 10MW Reference Wind Turbine was made in FA software followed by the validation of the model against the experimental data. The results indicate that damage can be detected, through the use of NA tracking based on Kalman Filter. Reducing fatigue loads due to wake effects for offshore wind farm was another research focus. To estimate fatigue loads due to wake turbine interaction the development of wind turbines wakes was studied. It was found that a factor that greatly influences wake development is its interaction with the Atmospheric Boundary Layer (ABL). Field data possess high standard deviation and is at times erroneous. Instead, Large Eddy Simulation (LES) has been used to generate data that offers ‘higher fidelity and detail’ for tuning engineering models. ECN has developed an Energy-Conserving Navier-Stokes (ECNS) code that has been verified for laminar flows. It offers benefits like unconditional stability and zero numerical dissipation, which are very important for the LES of wind farms (LES). This code was modified and the LES based analysis of wind farms to suggest improvements for engineering models. Another task focussed on a risk assessment for collisions between vessels and the offshore wind farm. This is an area of considerable interest – particularly as a result of the rapidly increasing number of wind farms near high-density vessel-traffic waterways, such as the North and Baltic Seas. The tools and methods most widely used for risk assessment in the maritime industry has been compared. The various parameters that should be taken into account for wind-farm vessel collision risk assessment have also been evaluated. The goal of this research is to optimize the use of limited sea-space by balancing navigational safety and environmental efficiency requirements. It is anticipated that the research will demonstrate that offshore wind does not conflict with safety requirements - paving the way for further development of this technology.
Work Package 5: Fluid-Structure Interaction
At deep water common fixed-bottom OWTs are found to be too expensive, motivating the so-called floating offshore wind turbines (FOWTs). Due to the violent loading scenario of their location (combining both wind and wave excitations), this kind of machines are subjected to strong dynamical effects. FSI prediction methods were adapted to FOWTs context for rotor analysis. Hydrodynamic and mooring line forces coming from sea-state excitation induce considerable motions in the 6 DOF of the wind turbine tower top. These motions are translated into the rotor blades, amplifying the unsteady aerodynamics and FSI effects. Commercial computational fluid dynamics (CFD) codes from NUMECA International were used in order to model this challenging multi-physics problem. Results were also obtained with in-house tools at the University of Glasgow, showing substantial deformations of the blade with the displacement at the tip up to 10% of the radius. Steady and unsteady FSI simulations of the rotor were performed, where re-adaptation of the fluid domain to the deformations/displacements experienced by the blades was necessary. Next to it within twist-coupled aero-elastic design for passive loads reduction on a full scale blade aimed at optimization of the passive load reduction behaviour through bend-twist coupling of the DTU 10 MW reference wind turbine. A reduction in extreme and fatigue blade root design bending moment, blade mass and foundation loads was aimed for while maintaining power output and stability characteristics of the initial design. Apart from the beam model implementation a parameter study on swept blades (geometrical-coupling) were conducted. Flow control for improving aerodynamic performance and noise reduction can be introduced through active methods. The main objective of the task was to develop open and closed-loop active flow control strategies to: delay flow separation and stall in real time; and optimize the overall efficiency and minimize associated noise.