Abstract: To address the gap in intelligent analytical detection within undergraduate experimental curricula, this study designed an integrative experiment of chemistry on the intelligent detection of microplastics via Raman spectroscopy, incorporating galvanometer scanning technology and deep learning. By implementing galvanometer scanning in a conventional Raman spectrometer, efficient localization and spectral acquisition of microplastic particles in complex systems were achieved. A one-dimensional convolutional neural network (1D-CNN) was employed for intelligent component detection of microplastics in mixed samples. Principal component analysis (PCA) was utilized to reduce the dimensionality of spectral data from identical microplastic types, enabling quantitative analysis of microplastic concentrations. This experiment focuses on the forefront of interdisciplinary advancements in chemistry, encompassing artificial intelligence, optical design, and instrumental analysis, thereby establishing a novel pedagogical paradigm that integrates traditional spectral analysis with artificial intelligence.

Key words: Microplastics, Raman spectroscopy, Galvanometer scanning, Deep learning, Intelligent detection