Development Of High-precision Light-Scattering Aerosol Mass Concentration Measurement System

With the rapid development of modern industry, the environmental pollution is more and more serious. Aerosol mass concentration is an important parameter to evaluation the changes in the atmospheric environment, and the light scattering method is an effective method to measure the parameter on-line. It includes a single particle counting and an ensemble-measuring technique. In this paper, using the former method, the collection and processing of the scattered light signals of particles and the influence of relative humidity on the measurement of aerosol mass concentration are studied in detail. The main work is as follows:Based on the non-uniform segmentation method for the voltage signal amplitude of particles, the multi-channel signal collection circuit is designed, which is used to collect the signals outputted by the optical sensor. Then combing the FPGA and STM32, aerosol mass concentration is inversed by using the fractal model. The experimental results indicate that, for the dust and air samples, the mass concentrations inversed by the voltage signal amplitude distribution obtained by the multi-channel signal acquisition circuit are in agreement with the actual measured values. The absolute errors between the slope of the fitted line and 1 are 0.034 and 0.016, the correlation coefficients are above 0.999, and the average relative errors are less than 7.7%. But the average relative errors of mass concentrations inversed through the integrated signal are less than 18.7%.Physical properties, such as density, refractive index and morphology of the aerosol particles will be changed when the relative humidity of the environment is relatively high, which leads to a significant variation of the scattered light signal collected by the single particle light scattering measurement system. In this case, if the calibration parameters in the condition of low humidity are applied to inverse aerosol mass concentration, the measurement accuracy will certainly be degraded. Therefore, the calibration parameters of the system must be corrected. That is, the relationship of the calibration parameters and the relative humidity need to be determined. Hence, firstly, combining the Mie scattering theory with the humidity growth model of the particle size and the refractive index, the influence of relative humidity on the scattered light signals of spherical particles is calculated; Secondly, the aerosol mass concentrations in the high relative humidity are revised using the PSO-BP neural network algorithm. The experimental results indicate that the effect of the relative humidity is effectively modified by the neural network algorithm.In this work, the study of the measurement of aerosol mass concentration is just satisfied the actual demands for the atmospheric environmental monitoring of the China. Meanwhile the single particle light scattering for inversion of the mass concentration is more improved.