Parameter optimization of microjoule high pulse repetition frequency laser induced breakdown spectroscopy for aluminum alloy identification

In the last decade, the analysis and recycling of aluminum alloy continue to raise additional concerns. A large number of aluminum alloy grades and their similar appearance makes it difficult to distinguish these visually manually. A fast and effective real-time in-situ classification method is urgently needed to guide the recycling and reuse of aluminum alloy. In this paper, we propose a laser-induced breakdown spectroscopy (LIBS) based on a microjoule high repetition frequency laser, which significantly reduces the volume and cost of the instrumentation required to realize LIBS compared to conventional LIBS. We used polished flat aluminum alloys for seven wrought alloy classes as samples and designed two spectral acquisition modes based on the existing experience, and further studied the influence of experimental parameters on the spectrum (the fixed mode and the motion mode). When the integration time is 80 ms and the integration windows is 80 to 160 ms, the intensity and stability of the spectral peak can be effectively improved in the fixed mode, when the integration time is 50 ms and the movement speed is 7 mm/s, the spectral peak intensity and stability of the spectrum can be significantly improved in the motion mode. Considering the practical application, we combined with back propagation artificial neural network (BP-ANN) to classify samples from two manufacturers, and the highest classification accuracy after optimization is 97.71%.