Research On The Layout Optimization Of Offshore Wind Farms Based On Wake Model

China’s offshore wind power is developing towards larger-scale,more massive and deeper waters.Several offshore wind farm clusters with tens of millions of kilowatts have been planned and are currently under construction.As the wind farm cluster’s size and wind turbine unit capacity continue to increase,the wake effect of large wind farms on their annual power generation,planning layout,and fatigue load becomes increasingly significant,becoming a hot and difficult issue in wind energy research.The construction of offshore wind farm clusters is different from that of individual projects,involving a wide range of factors,and whether the expected goals can be achieved after the construction of the wind farm clusters largely depends on the rationality and scientific nature of the planning layout.However,traditional wind turbine wake models severely underestimate the wake length of wind farms and require the development of engineering wake models and cluster layout optimization algorithms applicable to offshore wind farm scales.Based on this,this study developed a high-precision wind turbine position recognition technology based on "offshore wind power base machine position recognition → wind farm cluster wake effect evaluation → offshore wind farm cluster layout optimization → load and fatigue characteristic analysis." It revealed the wake effect of offshore wind farm clusters,constructed a wake model for offshore wind farms,and optimized the cluster layout method to further reveal the load and fatigue characteristics of wind turbines in offshore wind farms.The main content and conclusions are as follows:Firstly,a wind turbine position recognition technology based on the Google Earth Engine platform and Sentinel-1 satellite remote sensing data for offshore wind farms was developed.Nonwind turbine objects,such as moving floating objects or objects with a volume that is too large or too small,were removed by image recognition technologies such as time-series image pixel overlay averaging and image convolution kernel algorithms to improve the recognition accuracy to 95%.Based on this technology,coordinate information for all wind turbines in China’s coastal areas was obtained.Secondly,the impact of the wake effect of offshore wind farm clusters was revealed.Based on the mesoscale WRF model,the wake effect of Jiangsu Province’s offshore wind farm cluster base was simulated and studied to quantitatively evaluate the wake effect length and its impact.The wake length of the offshore wind farm cluster base currently reaches 40 km to 100 km,and the local wind farm is too dense,resulting in a maximum wind speed loss of 2.97 m/s and a local power loss of up to 50%.Thirdly,a Jensen wake model for wind farms at the farm scale and an offshore wind farm cluster layout optimization algorithm were innovatively established.The appropriate wake recovery coefficient was determined based on satellite observation data combined with mesoscale WRF simulation results to describe the wake effect evolution characteristics of wind farms.By constructing the relationship between wind farm capacity density,feature size,wind farm power curve,and blockage coefficient curve,a wind farm cluster and wind turbine coordinate optimization algorithm and program based on a genetic algorithm were developed.Under the same boundary conditions,the annual power generation of a 3100 MW offshore wind farm cluster base can be increased by 1% to 2.5%.Finally,the load and fatigue characteristics of wind turbines at different locations within offshore wind farms were quantitatively evaluated.Based on a certain offshore wind farm,the load and fatigue characteristics of wind turbines at different positions were analyzed,providing a reference for offshore wind farm operation and maintenance.