Abstract: To reduce the environmental impact of the coal yard dust on the urban PM_2.5, self-adaptive fuzzy control was used to improve the dust removal effect. Firstly, the membership functions of deviation of dust concentration and variation of dust concentration deviation and the spray pressure were established based on the Gaussian distribution function. Secondly, the fuzzy rule base and fuzzy controller were constructed to establish the mathematical model of spray pressure fuzzy control. Finally, the error back propagation learning algorithm was used to adjust the parameters of the fuzzy controller. The simulation results with MATLAB show that error back propagation learning algorithm enhances the adaptability of the fuzzy control. The spray practice in coal yard indicates that self-adaptive fuzzy control improves the fit degree of the spray pressure value with test data and can output the optimized spray pressure to achieve the best dust removal effect.