The continuous development of society has destroyed the living environment of human beings to a large extent. The problem of air pollution is becoming more and more serious, which affects people’s daily life and threatens human health and the natural environment. It is very important to evaluate and prevent air pollution. The existing atmospheric environmental pollution assessment system has problems such as low accuracy of prediction results and incomplete analysis of problems, resulting in untimely prevention and control of regional atmospheric environmental pollution. This paper analyzes the needs of atmospheric environment data mining, monitoring the prevention and control of atmospheric environment pollution, using data mining technology combined with power data to evaluate and prevent air pollution. Through the real-time detection and evaluation of the regional environmental pollution index, it is possible to carry out timely treatment when the atmospheric environmental pollution does not meet the standard. Through the prediction accuracy test, the environmental pollution index test, the atmospheric environment quality score test and the prevention and control effect test in different regions, it is found that the data mining technology is used for prevention and control. The prediction accuracy increased by 4.18%, and the environmental pollution index decreased. It is more conducive to the control of regional atmospheric pollution, the rate of atmospheric environmental quality score has been increased by 2.38%, and the control effect has been optimized by 3.1%. Data mining technology is more conducive to the assessment and prevention of regional air pollution.