Abstract: Abstract：With the acceleration of urbanization, air pollution has become one of the major problems faced by every city.We used the concentration data, meteorological data and land use type data of 2017 PM2.5 monitoring station in Hefei，and we combined the random forest algorithm (RF) and land use regression model (LUR), to simulated the spatial distribution characteristics of PM2.5 concentration in Hefei.And the influencing factors of PM2.5 were analyzed by principal component analysis.The results showed that: (1) The daily variation characteristics of PM2.5 concentration in Hefei City showed two peaks, most of which appeared at 8∶00-9∶00 in spring, summer and autumn, while appeared at 10∶00-11∶00 in winter.The low values are generally between 15∶00 and 17∶00. The change trend of PM2.5 concentration in the whole year is similar to that in spring. The change of PM2.5 concentration is the most stable in summer.(2) In 2017, the concentration distribution of PM2.5 in Hefei weakened from the city center to the outside, forming a spatial distribution pattern of high in the north and low in the south, high in the West and low in the East.(3) In terms of influencing factors, the change of PM2.5 concentration is negatively correlated with precipitation, wind speed and relative humidity, sunlight has a greater influence on PM2.5 concentration, air pressure and other pollutants have a positive correlation with the change of PM2.5 concentration, and NO2 among atmospheric pollutants has a greater influence on the change of PM2.5 concentration.