حسين عيوض اوغلي

The increasing global demand for energy, limited fossil fuel resources, and intensified effects of climate change have led to a significant focus on renewable energy sources. Ocean waves are one of the promising sources of renewable energy. Various devices have been developed to harness energy from these sources, with the Oscillating Wave Surge Converter (OWSC) recognized as the most efficient system. Climate change, changes in atmospheric loading, wind patterns, and sea level variations have important effects on wave characteristics. These changes directly impact the performance and efficiency of wave converters, both qualitatively and quantitatively. Therefore, predicting environmental conditions for future years is essential when aiming to establish a sustainable wave energy source. In this study, the prediction of climate parameter conditions for the years 2030 to 2040 was initially addressed. The M5p decision tree was used as a powerful machine learning tool to downscale wind speed results from the Global Climate Model (GCM) and develop a wave prediction model. The developed models included annual, seasonal, and monthly models, with different data separation scenarios examined for training and testing the decision tree model. The results showed that the monthly model performed better for the control period, and the 30/70 data separation scenario resulted in improved model performance. The prediction models based on the RCP2.6, RCP4.5, and RCP8.5 scenarios indicated that the monthly average wind speed in the study area, which is the North Atlantic Middle Region, would be similar for the years 2030 to 2040, with increased fluctuations observed, particularly in the RCP8.5 scenario. Furthermore, the comparison of prediction results indicated a 7% to 8% decrease in average wind speed compared to the historical period. For wave prediction model development, various scenarios and techniques were employed, including modifying the drag coefficient relationship and calibrating the CEM formulation, data limitations, and prediction capabilities. The results showed that the M5p model outperformed the modified CEM model, with a near-zero average prediction error and a 30% improvement in RMSE error compared to CEM. The predicted monthly average wave height for the years 2030 to 2040 indicated a slight increasing trend in March and July, while the rest of the months showed a decreasing trend across the entire study domain, with a predicted approximately 10% reduction in annual average wave height. However, it should be noted that the coastal region and the selected installation location for the converter, which is currently Oyster, will experience larger waves during the spring and winter seasons. To develop a numerical model of the wave converter, OpenFOAM was used as a powerful CFD-based model that, using the moving mesh and wave generation and absorption boundary libraries, reproduced accurate results from laboratory experiments used for validation. The present study showed that changes in the seabed surface near the converter had no significant effect on its performance, while changes in the cross-sectional area near the water level and the distribution of increases and decreases in the surface had a considerable impact on the OWSC performance, resulting from the effects of the horizontal velocity profile of particles in the water depth. The results indicated that an increase in the cross-sectional area near the water level and towards the coast leads to the formation of a trapped air package near the coast when the converter returns to the sea, at its highest level. This has an effect on the rotational response and angular velocity of the converter. By comparing the performance of proposed converters, OWSC-07 was selected as the superior converter due to its highest efficiency coefficient and higher net power, and was eva‎luated more comprehensively under the desired sea conditions. The eva‎luation results of climate change effects showed that the highest performance of OWSC-07 was at a wave period of 2.4 seconds and a height of 0.27 meters, installed at a water depth of 12 meters on a real scale, and that changes resulting from climate change would have a negative effect on its efficiency coefficient. Additionally, despite increased energy absorption with increased wave height, the efficiency coefficient experiences a decreasing trend.

ريزمقياس‌نمايي , سرعت باد , ارتفاع موج , اپن‌فوم , درخت تصميم , يادگيري ماشين , مبدل انرژي موج

downscaling , wind speed , wave height , OpenFOAM , decision tree , Machine Learning , wave energy converter

Hossein Eivaz Oghli

Dr. Abbas Yegane Bakhtiyari