标题：ACCURATE PREDICTION OF EXTERNAL CORROSION RATE OF BURIED PIPELINE BASED ON KPCA-MABC-SVR MODEL USING IN OIL AND GAS GATHERING AND TRANSPORTING SYSTEM
摘要：Buried pipelines are an indispensable part in the process of oil and gas gathering and transportation. Due to the increase in service time and harsh working environment, the pipelines are prone to various corrosion effects and damage, and then oil and gas leaks cause serious environmental pollution. Aiming at the problem of external corrosion rate of buried pipelines, this study first analyzes the influencing factors of external corrosion of pipelines, and analyzes the KPCA (Kernel Principal Component Analysis) algorithm, MABC (Improved Artificial Bee Colony) algorithm, and SVR (Support Vector Machine Regression) algorithm separately. After that, the construction method of the KPCA-MABC-SVR model is proposed. Then the KPCA algorithm is used to reduce the dimension of the factors affecting the external corrosion rate of the pipeline. The MABC algorithm is used to optimize the parameters of the SVR algorithm. The 64 sets of data are used to the SVR model. Training was performed to predict the external corrosion rate data of the remaining 1 2 groups of pipelines, and the prediction results were compared with KPCA- ABC-SVR model, KPCA-BP neural network model and KPCA-BAS-GRNN model in the literature to prove the new model (KPCA-MABC-SVR model) that we developed in this study. The research shows that the integrity of the anticorrosion layer outside the pipeline, the effectiveness of cathodic protection, and the soil pH are the main factors affecting the external corrosion of the pipeline. The average absolute error of the prediction of the external corrosion rate of the pipeline using the KPCA-MABC- SVR combined model is 3.0385%, and the training time is 4.76s, and both data are shorter than other models. The research results show that using the KPCA-MABC-SVR model to predict the external corrosion rate of the pipeline has a relatively small error and the model training time is relatively small. Therefore, the KPCA-MABC-SVR model has a strong ability to predict the external corrosion rate of the pipeline.
关键词：Oil and gas gathering and transporting system;environmental pollution;KPCA-MABC-SVR model;buried pipeline;external corrosion rate;influencing factors
