Coordinators

• TU Delft, University of Michigan, NREL

Topics and objectives

This work package will develop tools and reference cases that support the research and technology development of AWE systems. The following specific tasks will be targeted:

– Joint reference models

• Already delivered in 1st term (with varying degree of fidelity) but continued development and refinement planned
  o AWE techno-economic reference model (Joshi, et al.) – refinement in scaling and costs, accompanying data partially available
  o TU Delft MegAWES kite (Eijkelhof et al.) – make accompanying data available
  o TU Delft V3 reference kite (Poland, et al.) – data-based (experimental) validation performed, accompanying data available (<20 kW)
• Any other reference models that are work in process:
  o Add CAD models (some already available)
  o Also interest for smaller fixed-wing kite models: using Roland’s AP2 Model
  o Polimi working on fly-gen reference model (100kW)
• Considerations:
  o Define somewhere the terms, e.g. “reference model”
  o Check IEC 61400-21 on reference models, develop a template and collect needs for data,
  o Check nomenclature hierarchy / update glossary
• Deliverables: Hold an interactive workshop for at least one combination of data-equipped (including environmental data and corresponding measured outputs) and ideally controller-equipped reference model, and simulation framework. The goal is to demonstrate the use of the simulation framework to drive the model with the available reference model wind data, and produce output results (e.g., position, attitude, and power traces, along with averaged values over the simulation) that can be compared against actual reference model measurements. For those reference models that are not accompanied with corresponding data and/or controllers, the demonstration can still be performed, but an expectation of one-to-one correlation between the model and observed outputs is not realistically expected. (example contributor: Oriol Cayon)

– Integrated design and analysis workflow / framework / methods

Develop a workflow for tools / MDAO process:

• Understand what is needed by industrial partners as basis for work at universities
• Allows optimisation of the products developed by the companies
• Use previous work from meetings on ontology
• Define well the interfaces, connect physical model to control -> bring models closer to developing software
• Manage expectations, MDAO not as deliverable but showing the pathway
• Create infrastructure to make it easy to plug&play with different components
• Optimal scaling of the aircraft and the plants by applying numerical optimization to determine the best configuration according to a range of engineering, economic, and environmental impacts. -> AWETRAIN TUD
• Incorporate AWES into the WindIO, a data format for inputs and outputs to wind energy system computational models. If an inclusion is technically not possible an own data format shall be developed

– Design/analysis tools

Available tools that will be further enhanced:

• CORE Lab (Vermillion) closed-loop kite simulation framework (Simulink-based)
• Julia Kite Power tools (Uwe Fechner)
• BORNE model
• AWERA Tool (couples wind statistics for a year with AEP of a specific system)
• Reference model vs. controller
• Include reference models for grid integration in the scope: Need to comply with grid requirements, understand them very beginning to avoid retrofits; depends also buffer – topic of topology of electrical connections
• Comparisons between model predictions and experimental results: CORE Lab has validated its model against tow testing data.
• Deliverables:
  • Integration of dynamic power curve characterisation with AWERA Tool (Kitepower is working on it) – Meriodional deliverable
  • Develop a controller that can be linked to the reference model (Uwe Fechner). Note that this links with the reference model deliverable, which should ideally be achieved using a controller-equipped reference model.
  • Define a framework for an airborne wind energy multidisciplinary design, analysis, and optimization (MDAO) tool. This should not be confused with the development of the tool itself, which is out of the scope of this project. However, this deliverable will ensure that the community has a grasp on a candidate eXtended Design Structure Matrix (XDSM) for optimizing an airborne wind energy system, along with the available tools for performing each associated multidisciplinary analysis (MDA) and the overarching optimization

– Control models and tools

• Extending the operational range of AWE by utilising turbulent models and exploring control techniques. – is being worked on by TUD / AWETRAIN Kitemill Phd
• Investigating component faults and building fault tolerant control algorithms using historical test data. – is being worked on by TUD / AWETRAIN Polimi
• Creating supervisory control methods that enable long-term fully autonomous control with negligible human involvement. – is being worked on by Polimi / UniMichigan
• AWE Farms: Exploring how many devices can be combined and operated within a confined area by applying new control strategies to avoid collisions, mitigate wake interactions, and maximize energy production (AEP). -> AWETRAIN Polimi (Alessandro Croce)

– Grid integration of AWE

• Exploring how AWE control and operation can be used to both improve the power and provide additional services to the grid. (AWETRAIN HM)
• Investigating the electrical subsystems within the AWE plant by exploring with system engineering models different topologies, condition monitoring, and fault tolerant design. (AWETRAIN HM)
• Investigates the interaction of grid stability and AWE based plants, including recommended requirements for AWE based plants and AWE based grid services to improve the grid. (Christoph Hackl, AWETRAIN HM / UC3M)

– Wind resources / meteorology

There are still many open issues to gather data of high altitude winds and to understand fully the wind
resource potential that can be captured by AWE systems.

• DTU is looking into various aspects of wind resource assessment (Mark Kelly)
• Resource assessment and link most suitable sites for AWE through GIS analysis to be carried out for Spain (Lead: Miguel Ángel Gaerner, Uni Castilla-La Mancha).

– Deliverables

D2.1 End-to-end workshop demonstrating application of a dynamic model to a reference model (with accompanying data and ideally an accompanying controller)
D2.2 Defined MDAO framework
D2.3 Power curve from dynamic simulation -> AWERA tool for AEP based on real wind statistics
D2.4 Control reference model
D2.5 Farm-level deliverable (Alessandro Croce)
D2.6 Grid integration modeling deliverable (David Santos)

– Milestones

M2.1 WP kick-off meeting
M2.2 Completion of a database for design tools, analysis tools and reference models
M2.3 Workshop to provide critical discussion and training on AWE design tools (e.g. through the endto-end demonstration of the tool)