Abstract: With the continuous deepening of air pollution control, the demand for refined, precise and intelligent environmental management is increasing. It is urgent to carry out high-density monitoring in specific areas to make up for the shortcomings of the existing traditional monitoring.. In this context, PM_2.5 sensor monitoring methods are developing rapidly in China. In order to meet the needs of fine management of air pollution prevention and control, and study the performance of atmospheric PM_2.5 small sensor monitoring methods based on light scattering principle, a comparison test platform of atmospheric PM_2.5 sensor equipment was established in Beijing from July 2018 to July 2019. The PM_2.5 sensor monitoring equipment with different principles and brands was tested and analyzed for one year.The results showed that the performance of the industrial-grade sensors was better than that of the civil-grade sensors, the performance of PM_2.5 sensors with laser particle counting method is better than that with infrared light-scattering method, and the correlation between cross-brand PM_2.5 sensors is significantly lower than that within the same brand and type. During the comparison, the industrial-grade sensors had an data efficiency rate greater than 95%, and most of the industrial-grade sensors with laser particle counting method had a R² above 0.75, SD below 10 μg/m³ and CV below 15%, which were in good consistency with the standard equipment; the influence of relative humidity on PM_2.5 sensors generally showed a positive error and tended to increase with the elevation of ambient relative humidity, but different types of sensors were affected by humidity in different patterns. The monitoring results of sensor devices under different pollution scenarios showed various deviations compared to standard devices due to a combination of factors such as the chemical composition of the particles and the surrounding environment; the sensor equipment showed a period of stability, but the performance of all aspects of the instrument declined as the operating time continued to increase. For the operational application of PM_2.5 sensors, it is necessary to establish a suitable quality control calibration system and carry out regular quality control efforts based on the local performance assessment results in order to ensure the quality of the data. Studies show that the laser particle counting method PM_2.5 sensor is more suitable for large-scale high-density grid monitoring. Through continuous optimization of hardware and quality control algorithm, the data quality can meet certain environmental management requirements..