The OWA GloBE (Global Blockage Effect) project was introduced in the OWA GloBE webinar released in August 2024. Rera Tech's technical notes explain this project, which can also serve as a reference for offshore wind power generation in Japan, in two parts.
Last timehas summarized the overview of this project and the results of the observed blockage effect. This time, we will focus on the results of the numerical model reproduction of the observed wake and blockage effect, and the guidelines that were formulated based on the results. For details, please seeThis videoIt is explained in the article. Please take a look at it.
first
What is the OWA GloBE Project?
This project investigated modeling methods for blockage and wake effects with the aim of improving the accuracy of energy forecasts for large-scale offshore wind farms. Through detailed wind data collection and analysis using numerical models, the project quantified the impact of these effects on energy losses and redistribution throughout the wind farm. In addition, the importance of the Planetary Boundary Layer (PBL) height was demonstrated, and model improvements were made based on this.
Overview of the blockage effect
The blockage effect is a phenomenon in which wind speeds slow down upstream of a wind farm (up to 20% slower) and speed up downstream (up to 10% faster). This reorganizes the wind speed distribution in the wind farm and affects the energy distribution between the turbines. Traditional models do not take this effect into account sufficiently, and in particular underestimate the energy losses across the power plant cluster. The main goal of this project is to build a numerical model that accurately reproduces this effect and to validate it using measurements.
For details, Previous article: OWA GloBE Webinar Technical Explanation ①: Results and Significance of Blockage Effect Measurement
Results and issues of reproducing blockage effect using numerical model
The results of reproducing blockage effects and other issues using the numerical model implemented in this project are summarized in the table below.
Comparison table of reproductions by numerical models
| Model name | the purpose | Key results | Computational cost | Task |
|---|---|---|---|---|
| Viscous Vortex (VV) Model | -The blockage effect is decomposed into inviscid (deceleration/acceleration) and viscous (wake formation) effects. – The effects of boundary layer height and thermal stability can be reproduced. | – Errors within ±5% of measured data. – Deceleration width is reduced by up to 15% when the boundary layer height is high. – At low boundary layer heights, deceleration increases by up to 20%. | Moderate: Per case About 1 time | – Additional computational load on large clusters. – Boundary layer height and stability settings have a significant impact on the results. |
| RANS CFD Model | – Advanced fluid dynamics analysis is used to reproduce the wind speed distribution and pressure field between the turbines. | – Wind speed reduction (up to 20% reduction) is reproduced with high accuracy by comparing with measured data. – Detailed analysis of wake effects is possible. | expensive: Per case About 10 time | – Long computation time, not suitable for large wind farms or clusters. |
| PALM Model | – LES (large eddy simulation) is used to analyze the detailed interactions between blockage and wake effects. | – Achieves the highest accuracy with measurement data error within ±3%. – Faithfully reproduces the effect of wind speed distribution due to boundary layer height changes. | Very high: Per case Approximately 100 hours or more | – The computational cost is very high, making it of low practical use in industry. |
| Industry standard model | – Simplified wake model for rapid energy assessment. | – The blockage effect can be partially reproduced, but with an error of about ±15%. – Accuracy decreases for complex wind farm clusters. | low: A few minutes per case | – Boundary layer height and thermal stability cannot be taken into account, resulting in reduced accuracy, especially in complex conditions. |
The role of the meso-scale meteorological model WRF
In this analysis, the mesoscale meteorological model WRF played two important roles in analyzing regional meteorological conditions and boundary layer height:
- The distribution of boundary layer height (500-1500m) is tabulated by direction and speed, and used as input data for VV models and CFD models.
- When compared to measured data, WRF's boundary layer height evaluation values tend to be calculated about 10% higher on average, suggesting that this difference may affect the results.
On the other hand, it has been identified as a technical challenge that the calculation of boundary layer height contains uncertainties due to the characteristics of the WRF PBL scheme.
About the Planetary Boundary Layer (PBL)
As mentioned frequently in this article, the planetary boundary layer (PBL) has a significant impact on the blockage effect. The boundary layer is the lowest layer of the atmosphere that is directly affected by friction and heat exchange with the Earth's surface. The thickness (altitude) of the boundary layer varies depending on the characteristics of the Earth's surface and meteorological conditions, and can range from several hundred meters to several kilometers over the ocean and coasts. Changes in the thickness of the boundary layer cause the following effects related to the blockage effect:
1. Changes in wind speed
- At low boundary layer heights, the blockage effect becomes significant, reducing wind speed by up to 20%.
- At higher altitudes, the blockage effect is mitigated and the reduction in wind speed is limited to about 10-15%.
2. Changes in energy distribution
- The wind speed profile in the boundary layer affects the interactions of the turbine array and varies the overall power generation efficiency.
Reproduction of boundary layer by numerical model
The WRF model is used as a primary tool to analyze regional boundary layer heights. In this study, it was confirmed that the boundary layer heights estimated using the model were approximately 10% higher than the measured data, which may affect the reproduction of wind speed distributions.
Achievements and challenges of the OWA GloBE project
Main achievements
1. Confirmation of the existence of the blockage effect
- Redistribution effects in the energy distribution within wind farms have been observed, which can lead to energy losses of up to 6%
2. Elucidation of the effect of boundary layer height
- When the boundary layer height is low, the wind speed reduction reaches nearly 20%, and the blockage effect becomes significant.
- It was confirmed that the blockage effect is mitigated when the boundary layer height is high.
3. Contribution to the industry
- Industry-wide unified modeling guidelines established through the "Joint Statement"
- Presents criteria for model selection and points to note when using the model
Technical Issues
1. Reducing computational costs
- In order to improve the practicality of LES and CFD models, a method to reduce the computational load is required.
2. Unexplained physical phenomena
- Elucidation of the influence of gravitational waves and Coriolis force on the blockage effect
- Detailed analysis of interactions between wind farms (power plant clusters)
Commentary on the "Guidelines (Joint Statement)"
About the Joint Statement
The Joint Statement, an industry-standard guideline for accurately assessing the blockage effect in offshore wind power generation, was developed based on the contents of the OWA GloBE project and is shown in the table below.
| Conclusion | Details |
|---|---|
| 1. Existence of blockage effect | – Based on the measurement data, a slowdown in wind speed upstream (up to 20% decrease) and an acceleration in wind speed inside (up to 10% increase) were confirmed. – Redistribution effects occur in the distribution of energy between wind turbines and power plants. |
| 2. Identifying the physical effects | – The blockage effect causes streamwise and lateral energy redistribution. – Positive and negative energy losses occur in turbines and power plants. |
| 3. The Importance of Boundary Layer Height | – It was confirmed that the change in boundary layer height significantly affects the magnitude of the blockage effect. – At lower altitudes, the effect is amplified (wind speed reduction up to 20%). At higher altitudes, the effect is mitigated (up to 10-15%). |
| 4. Model Differences and Recommendations | – Significant differences in accuracy between models (±10-20%). – Three model categories are proposed: ① Decoupled (simple wake model, low accuracy and low cost) ② Tightly Coupled (medium accuracy, medium cost) ③ Fully Coupled (high precision, high cost). |
Recommended numerical model
Details of the three numerical models recommended in the guidelines are as follows:
- Decoupled(Independent model): A method that performs wake model and blockage effect analysis separately. After calculating the wind speed reduction between the turbines using the wake model, the blockage effect is added separately as a correction value.
- Tightly Coupled(Partially bonded model): A method that partially integrates the wake model and blockage effect. The wind speed deceleration is incorporated into the wake model, while a simple blockage effect is modeled.
- Fully Coupled(Fully bonded model): A method that fully integrates wake and blockage effects, using LES (large eddy simulation) and CFD (computational fluid dynamics) models to perform detailed analysis of wind speed distribution and energy redistribution.
The choice of these three models, together with the methodology for model validation, is described in the Modeling Guidelines as follows:
Recommended Modeling Guidelines
| Item | Details |
|---|---|
| 1. Criteria for model selection | – Select the appropriate model depending on the project scale and purpose. ① Small-scale projects and initial evaluation: Decoupled model (simple wake model). ② Detailed design and large projects: Tightly Coupled or Fully Coupled models. |
| 2. Consideration of boundary layer height | – It is important to explicitly model the boundary layer height. – Data on boundary layer height and thermal stability obtained from WRF and lidar measurements are used as input conditions for the model. |
| 3. Balancing accuracy and computational cost | – Apply models according to project phases, taking into account model accuracy and computational costs. – Low-cost models are recommended for initial evaluation, and high-precision models for detailed design. |
| 4. Verification Method | – Always validate model outputs against measured data (e.g. lidar measurements). – Based on the measurement data, adjust the model parameters to improve accuracy. |
Summary
The OWA GloBE project has deepened our understanding of the blockage effect in offshore wind farms and significantly improved the accuracy of energy forecasts. In particular, the analysis of boundary layer height using the WRF model and the reproduction of measurement results using each numerical model are important results that overcome the limitations of conventional models. On the other hand, research into the application of this model to large-scale wind farms (power plant clusters) and the influence of unexplained physical phenomena such as the Coriolis force remain as future challenges. The knowledge gained from this project is expected to be useful in the introduction of large-scale offshore wind farms in the future. It is believed that the influence of the blockage effect will become more pronounced in Japan as well, and this result will be a useful reference from a technical perspective.
(Written by Mizuki Konagaya)
References
The Carbon Trust RWE, (2024, August 8th). OWA GloBE: Modelling and Accounting for Blockage and Wake Effects [Video]. YouTube. https://www.youtube.com/watch?v=thWK8KTJJAc