Prediction of hydrocarbon composition of naphtha is researched.Aiming at the features of complex and redundant of the composition of naphtha

and the problem of complex and long time of traditional method

a predicting method of principal component analysis（PCA） and particle swarm optimization（PSO） is proposed based on online Raman spectroscopy to improve the accuracy of the prediction.First

the dimension of data is reduced using the PCA to eliminate the redundant information after obtaining the Raman spectrum of naphtha sample.Then using the particle swarm algorithm to predict data after analyzed by PCA

and then get the mass fraction of various components and analyze the correlation of the predictive value and the true value of the sample.Experimental results show that compared to other predicting methods

the proposed method is more accurate

more efficient and more adaptable to the site.It is an efficient prediction method with high real-time performance.