Research On The Wind Farm Parameterization Based On The Mesoscale Numerical Weather Prediction Model

In order to achieve the national strategic goal of "Carbon Neutrality ",China’s wind power industry will develop in the direction of scaling up and large-scale.Among them,the weather research and forecasting model(WRF)coupled with the wind farm parameterization(WFP)is playing an important role.In response to the problems of poor applicability,low accuracy and insufficient robustness of traditional models,this paper innovatively develops an efficient and reliable numerical parametric modeling method for wind farms,gradually improving and refining three aspects,such as dynamic air density/rotor equivalent wind speed,sub-grid interference effects and consideration of human factors such as wind farm power regulation,and evaluating the improvement effect of each model separately,finally,based on the above improvements,a hybrid model is proposed to build a WFP model with wide applicability,high accuracy and practicality,and to realize the engineering application of wind farm power prediction.First,a high-precision mesoscale numerical background field applicable to complex atmospheric boundary conditions is constructed.An accurate atmospheric physical background field is the key to accurately characterize the mesoscale simulation of large wind farms.On the one hand,the accuracy and robustness of the background field simulation of the WRF model are improved by introducing high-resolution topographic databases.On the other hand,a coupled sea surface-atmosphere-wind farm model framework is constructed to consider the sea-air-farm interaction within the marine atmosphere boundary layer.Second,an innovative WFP with wide applicability,high accuracy and practicality is constructed.In this paper,three refined WFP models are proposed: 1)Improved WFP: By introducing dynamic air density and rotor equivalent wind speed to consider the spatial and temporal variation of air density and wind shear,the wind power assessment capability of WFP is improved in different regions,seasons,and complex atmospheric stability environments.2)Coupled WFP: A mesoscale wind turbine sub-grid interference model is proposed.The model is coupled to the WRF model to compensate for the model error caused by the interference of multiple wind turbines located in the same grid cell,which has a better performance in wind simulation;3)Control WFP: A Bins Monte Carlo simulation method based on the probabilistic statistics of wind farm historical operation data is proposed.The new power curve for characterizing the statistical relationship between the wind farm power output and wind turbines’ power output is coupled to the WRF model,which has a great advantage in wind power evaluation with an accuracy improvement of nearly 20% compared to the original model.Finally,the hybrid Combined WFP was established by combining the above three refined WFP models and applied to the short-term power prediction of an offshore wind farm in a region of Hangzhou Bay,Zhejiang Province.Compared with the traditional power prediction method,the short-term power prediction accuracy of this hybrid model is improved by 10%.Therefore,the hybrid wind farm parameterized model proposed in this study can provide a model basis for wind farm power prediction and provide key technical support for the sustainable development of China’s wind power industry.