针对Projection迭代反演算法对噪声极其敏感，及在实际应用中造成粒度测量失真的问题，引入Vondrak数据平滑算法对Projection反演进行平滑处理。借助Visual C++6.0开发平台，编写测试软件对算法应用进行了仿真研究。通过分析比较平滑处理前后的反演结果，得出Vondrak数据平滑处理算法的引入有效提高了Projection反演的抗干扰性能。实验结果表明：基于Vondrak数据平滑处理的Projection反演结果能够反映粒度的真实分布，满足现代粉体工业粒度测量的稳定可靠、精度高、抗干扰能力强的要求。

Inversion of particle size distribution from scattered light intensity distribution with laser particle size analyzer based on Mie theory is a typical process for solving inverse problems, in which particle size inverse algorithm is critical. Projection iterative inverse algorithm is sensitive to noise and it causes particle size measurement distortion in actual applications. Vondraks smoothing algorithm was adopted to conduct smoothing processing to Projection inversion, and simulation study on algorithm applications was carried out by writing testing software on VC++60. By analyzing and comparing the inversion results obtained before and after smoothing processing, it is concluded that the introduction of Vondrak data smoothing processing algorithm effectively improves the anti-interference ability of Projection inversion. The experiment result shows that the inversion result obtained after Vondrak data smoothing processing reflects the real particle size distribution and meets the measurement requirements on stability, reliability, high precision and strong anti-interference capability for modern powder production industry.