The influencing factors of pipeline corrosion are many and complex, so establishing an accurate prediction model of pipeline corrosion rate is the key to evaluate pipeline safety. Aiming at the defects of the traditional Elman Neural Network model, which is easy to fall into the minimum value and has weak generalization ability, a new ENN model based on improved reptile search algorithm (introducing Circle chaos map and combined hunting strategy with whale optimization algorithm) is proposed, and the effectiveness of the new model is verified by the measured results of pipeline corrosion rate. The prediction results of two examples show that the mean absolute percentage error of the improved new model is 0.5476% and 0.7831% respectively, and its prediction accuracy is obviously higher than that of the traditional ENN model. The new model can optimize the weights and thresholds in the prediction process, so it is helpful to improve the prediction accuracy of the traditional model.