Wind turbine gust estimation using remote sensing data

Towers, P.; Jones, Bernard J. T.

doi:10.1109/control.2014.6915165

Cited by 2 publications

(3 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Though training RL agents to achieve full control of free-flight systems can be challenging experimentally due to the trial-anderror nature of the learning process, flow-informed agents even have the potential to learn to take advantage of natural flow structures through energy-efficient soaring behaviors 20,47,49 . This technology could also allow wind turbines to safely operate at an increased range of conditions by reducing loads from potentially damaging gusts through actuation of blade pitch 3,5,7 . In the case of static systems such as wind turbines, off-board remote sensing upwind could further enhance performance.…”

Section: Discussionmentioning

confidence: 99%

“…For example, unmanned aerial vehicles (UAVs) and wind turbines both rely on fluid interaction for normal operation, but can be damaged or destroyed when operating in turbulent wind conditions [2][3][4] . Mitigating the effects of these turbulent forces through active control strategies is an ongoing challenge for both UAV and wind turbine applications [3][4][5][6][7][8][9] . However, turbulent conditions in the atmosphere are highly nonstationary and nonlinear making them difficult to model or control in real-time 1,10 .…”

mentioning

confidence: 99%

See 1 more Smart Citation

Machine learning for flow-informed aerodynamic control in turbulent wind conditions

Renn

Gharib

2022

Commun Eng

View full text Add to dashboard Cite

Control of aerodynamic forces in gusty, turbulent conditions is critical for the safety and performance of technologies such as unmanned aerial vehicles and wind turbines. The presence and severity of extreme flow conditions are difficult to predict, and explicit modeling of fluid dynamics for control is not feasible in real time. Model-free reinforcement learning methods present an end-to-end control solution for nonlinear systems as they require no prior knowledge, can easily integrate different types of measurements, and can adapt to varying conditions through interaction. Here, we show that reinforcement learning methods can achieve effective aerodynamic control in a highly turbulent environment. Algorithms are trained with different neural network structures, and we find that reinforcement learning agents with recurrent neural networks can effectively learn the nonlinear dynamics involved in turbulent flows and strongly outperform conventional linear control techniques. We also find that augmenting state observations with measurements from a set of bioinspired flow sensors can improve learning stability and control performance in aerodynamic systems. These results can serve to inform future gust mitigation systems for unmanned aerial vehicles and wind turbines, enabling operation in previously prohibitively dangerous conditions.

show abstract

Section: Discussionmentioning

confidence: 99%

mentioning

confidence: 99%

Machine learning for flow-informed aerodynamic control in turbulent wind conditions

Renn

Gharib

2022

Commun Eng

View full text Add to dashboard Cite

show abstract

“…In previous works, a wind farm is usually presented by a one-machine aggregated model with the assumption that all wind turbines are working under a same working condition. However, in a wind farm, wind turbines are spatially located, and wind speed is spatially distributed in a wide area [16]. The performance of the one-machine model is unsatisfactory when wind turbines have different parameters or work under different conditions [17][18][19].…”

Section: Introductionmentioning

confidence: 99%

Multi‐DFIG aggregated model based SSR analysis considering wind spatial distribution

Liu

Jiang

et al. 2019

IET Renewable Power Generation

View full text Add to dashboard Cite

This study investigates the impact of wind spatial distribution on sub-synchronous resonance (SSR), considering wind speed difference and wind turbine division. For a doubly-fed induction generator (DFIG) wind farm, two equivalent aggregated models, i.e. a one-DFIG model and a multi-DFIG model, are established and both simplified to impedance model for investigation. In one-DFIG model, the SSR mechanism in DFIG wind farm, and the impact of wind speed on SSR are discussed. The concept of damping is introduced to explain how DFIG SSR current changes under the negative resistance of wind farm. Then, in multi-DFIG model, DFIGs are divided into different groups based on their wind speed, and the interaction between DFIG groups is investigated. Simulations are used to validate the proposed analysis, and the results show that, multi-DFIG model performances a higher accuracy when dealing with DFIG wind spatial distribution problems compared with one-DFIG model.

show abstract

Wind turbine gust estimation using remote sensing data

Cited by 2 publications

References 9 publications

Machine learning for flow-informed aerodynamic control in turbulent wind conditions

Machine learning for flow-informed aerodynamic control in turbulent wind conditions

Multi‐DFIG aggregated model based SSR analysis considering wind spatial distribution

Contact Info

Product

Resources

About