The accurate determination of the annual lightning flash number to a wind turbine is essential for designing proper lightning protection measures. However, the method to assess the lightning risk of wind turbines recommended by the IEC is not associated with the actual lightning attachment process. Additionally, there is little research on positive cloud-to-ground (+CG) lightning. In this study, a lightning risk assessment method correlated with wind turbines on the basis of an improved stochastic lightning model is proposed. Based on the model, the influence of the lightning current amplitude, wind turbine heights and blade rotations on lightning strike risk are quantitatively analyzed. Moreover, the development and distribution characteristics of negative cloud-to-ground (−CG) lightning and +CG lightning are discussed. Finally, a more accurate calculation method of lightning strike risk is proposed considering the polarity effect. The results show that the effect of blade rotations on lightning risk cannot be ignored when the lightning current is large or when the wind turbine is tall. The −CG lightning has more bifurcation, but the dispersity of its development path and the lightning strike point are both small. The +CG lightning has scattered lightning strike points, and the development paths are tortuous, and. According to the verification calculation, the results calculated in this paper are in better agreement with the observation data than the results calculated by the IEC (International Electrotechnical Commission) recommended method.

The damage of wind turbines suffered from lightning strikes has been a key issue for the safe and reliable operation of wind farms. Accurate determination of the annual lightning flash number to a wind turbine is essential for designing proper lightning protection measures. The interaction of downward and multiple upward leaders (MULs) is studied in this paper, which also considers the stochastic nature and branched behaviour of the lightning attachment phenomenon. Firstly, with an improved stochastic lightning model, the relationship among the striking distance, the height of wind turbines and the return stroke current is established. Moreover, a modified method for predicting the annual lightning flash number strikes to a wind turbine is proposed. The simulation results show that the striking distance and the collection area not only depend on the return stroke current, but also on the height and blade angle. Besides, the comparations between the calculation results and field statistics indicate that the conventional electric geometry model is not satisfied with the need of lightning protection for wind turbines. Note that there is only a difference of 4% between the modified method and the observation value. These quantitative researches provide the guidance of calculation results for the optimization of the blade lightning protection system (LPS) design.

No abstract