Damaged coins can be identified effectively via spectral analysis based on LIBS, which is of great significance for coin recycling. This paper takes the Renminbi (RMB), the Chinese currency, as the example, including the denominations of YI FEN, ER FEN, WU FEN, YI JIAO, WU JIAO and YI YUAN. Some characteristic lines of Mg, Al, Fe, Cr, Cu, Sn, Ni, Na and Ca were observed in the spectra, as well as the molecular bands of AlO. Principal component analysis (PCA) was used to reduce the dimension of the spectra of the different RMB coins. The samples after dimension reduction are classified by k-Nearest Neighbors (KNN), and 4 categories were obtained with a classification accuracy of 100%. Further, new spectra of different denomination RMB coins were added to the original data for the same analysis. The results are in good agreement which shows the potential of the combination of LIBS, PCA and KNN for the analysis and identification of different coins.