Pipeline Remaining Life Prediction Based on PCA-PIO-FLN
In order to accurately predict the remaining life of a corroded pipeline and improve the prediction accuracy,a fast learning network(FLN)prediction model based on principal component analysis(PCA)and pigeon swarm optimization(PIO)is proposed for predicting the remaining life of the pipeline.Key corrosion factors are extracted by PCA to reduce the dimension of prediction indicators,and PIO is used to optimize the input weights and hidden layer thresholds of FLN to improve the pre-diction accuracy.In order to test the effectiveness of the model,50 sets of data from a water injection pipeline are taken as an example,and are compared with the FLN and BP models.The results show that the MAE,MAPE and RMSE of the PCA-PIO-FLM model are 0.036,0.553,0.0014,which are better than the comparison models.The results prove that the construc-ted model can accurately predict the remaining life of the annotation pipeline.
万方数据
维普
