基于角度散射法的气溶胶粒子粒径分布反演研究

Retrieval of Aerosol Particle Size Distribution Based on the Angular Light-Scattering Measurement Method

本文基于人工蜂群-差分近似优化算法(Artificial Bee Colony and Differential Evolution algorithm, ABC-DE)和角度光散射法对气溶胶粒径分布进行了反演研究。此外,本文还提出了一种基于敏感度分析和主成分分析的角度散射测量信号筛选方法,用以提高粒径分布反演结果精度。研究表明:采用角度散射法反演颗粒粒径分布时,选择前向散射角度,如[0°,30°],更有利于提高反演结果精度;粒径分布反演精度与所选择测量角度信号的贡献率相关,高贡献率的测量信号更有利于确保反演结果精度和鲁棒性。

Compared with the artificial bee colony and differential evolution algorithms, the angular light-scattering measurement method was proposed to study the aerosol particle size distribution. Moreover, an optimized selection principle of angular light-scattering measurement signals on the basis of sensitivity analysis and principle component analysis was proposed to obtain more useful measurement signals to improve the retrieval accuracy. Results revealed that when the angular light-scattering measuremen method was used to retrieve the aerosol particle size distribution, the measurement signals obtained from forward scattering angles, e.g. [0°, 30°], were benefit to improving the retrieval accuracy. Meanwhile, the retrieval accuracy was relevant to the contribution rates of angular light-scattering measuremen signals. The measurement signals with high contribution rates were more advantageous to ensure the retrieval accuracy and robustness.