Objective With the continuous advancement of the "14th Five-Year Plan" for energy, China's wind power industry has entered a new stage of large-scale and high-proportion development. For the operational efficiency of wind farms, improving the accuracy of wind speed forecasts for wind turbines is of vital importance.

Method This paper took the coastal wind farm as a representative to conduct an applicability study on the application effect of the three-dimensional wind speed forecast product of the STMAS-WRF forecast model.

Result The research shows that: (1) The trends of the predicted wind speed and the measured wind speed at the height of the wind turbine are generally consistent, and the predicted wind speed is generally larger than the measured wind speed. There is a good correlation between the forecast and the measured wind speed, but as the forecast period increases, the accuracy gradually decreases. (2) The forecast trends of the daily and monthly average wind speeds of the wind turbines are consistent with the measured wind speeds. The deviation of the monthly average wind speed forecast is the smallest in August and the second smallest in June. The wind speed is relatively high in November and October, indicating that it is more difficult to predict wind speed in autumn. (3) The forecast effect of the variable wind speed range and the low wind speed range is better than that of the rated wind speed range, indicating that the model's prediction ability for extreme values is insufficient. (4) The prediction effect of wind turbines in areas A and B is better than that in areas C and D, which may be related to the higher terrain in Areas C and D, increasing the difficulty and complexity of prediction.

Conclusion The prediction performance of the STMAS-WRF model is relatively good, and the error distribution has certain regularity. In the next step, this product can be corrected to improve the accuracy of wind speed prediction.