Abstract: In recent years, with the continuous enhancement of marine oil and gas exploration and development in China,there are more and more submarine oil and gas pipelines in service and under construction. The prediction of the internal corrosion rate of submarine oil and gas pipelines is extremely important for the daily operation, maintenance and overhaul of submarine oil and gas pipelines. In order to improve the prediction accuracy and stability of internal corrosion rate of submarine oil and gas pipelines, a combined model PCA-TSO-BPNN for predicting the internal corrosion rate of submarine pipelines based on principal component analysis(PCA) and tuna swarm optimization(TSO) optimized BP neural network(BPNN) was established. PCA was used to reduce the dimension of the data, and the main factors affecting the internal corrosion rate of submarine pipelines were screened out; the BPNN prediction model of internal corrosion rate of submarine pipeline was established, and the weight and threshold parameters of the BPNN prediction model were optimized by TSO algorithm;the PCA-TSO-BPNN combined prediction model was used to predict internal corrosion rate of submarine pipelines,which was compared with comparative prediction models to verify the reliability and feasibility of the PCA-TSO-BPNN combined prediction model. The results show that the mean absolute percentage error(MAPE) and root mean square error(RMSE) of PCA-TSO-BPNN combined prediction model are 1.8441% and 0.06757 respectively, which are far lower than comparative prediction models, and the combined models have higher prediction accuracy and stability, and can provide decision support for internal corrosion protection and flow assurance of submarine pipelines.