Abstract: In order to predict the corrosion rate of pipelines accurately, a pipeline corrosion rate prediction model based on principal component analysis（PCA）, artificial bee colony（ABC） and extreme learning machine（ELM） was proposed. The PCA algorithm was used to reduce the dimension of corrosion factors to reduce the information correlation between corrosion factors. The IABC algorithm was used to optimize the input layer weight and hidden layer bias of ELM model to improve the generalization ability of the model. The improved artificial bee colony（IABC） was used to optimize the input layer weights and the hidden layer thresholds of the ELM model to improve the generalization ability of the model.Taking the experimental data of 26 groups of A3 steel in seawater as an example, and the PCA-IABC-ELM model was compared with BP model, ELM model and PCA-ELM model to verify the reliability. The results show that PCA algorithm is effective in preprocessing corrosion data, and the IABC algorithm is effective in optimizing the model parameters, the average relative error of the combined model is 1.2274%, the prediction accuracy is better than other models. Therefore, PCA-IABC-ELM mode could be used to predict the corrosion rate of pipelines and guide the protection work of pipeline.