[Objectives] With the rapid development of the wind power industry, the number of wind turbines has increased rapidly across the country, and the impacts of wind farms on birds have attracted widespread attention. However, there is still a lack of research on assessing the collision risk of birds on wind farms. [Methods] We took Guangxi Xing’an wind farm as an example and investigated avian community and bird collisions within 5 km of the wind farm as well as its surroundings in April-May and September-October 2023 by using the line transect method and the carcass search method. We constructed a collision risk assessment model for birds in wind farms and analyzed its validity by combining bird mortality data in Xing’an wind farm in the past three years and literature review. [Results] The results showed that among the 85 bird species investigated, fourteen species (16.5%) were ranked as extremely high collision-risk, twenty-three (27.1%) ranked as high collision-risk, twenty-two (25.9%) ranked as medium collision-risk, and twenty-six (30.6%) ranked as low collision-risk (Appendix 1). A total of 29 bird strikes were recorded through a combination of field surveys and literature review, including 18 field investigation records and 11 literature collection records, and the birds that had bird strikes belonging to 13 species in 5 orders, 7 families. Among these species, 6 were ranked as extremely high risk, with 19 bird strike fatalities observed, accounting for 65.5% of the total bird strike fatalities; 5 species were ranked as high risk, with 8 bird strike fatalities observed, accounting for 27.6% of the total bird strike fatalities; and 2 species were ranked as medium to low risk, with a total of 2 fatal bird strikes observed, accounting for 6.9% of the total bird strike fatalities (Table 1). [Conclusion] The results indicated that birds of extremely high risk and high risk classes have a higher probability of collision, and the results of the bird collision risk model are consistent with field survey results. This model provides a practical value to assess the risk of bird strikes in wind farms. Its utilization can support bird strike risk assessment for prevention in advance.