The transportation of oil and gas through pipelines is crucial for sustaining energy supply in industrial and civil sectors.However,the issue of pitting corrosion during pipeline operation poses an important threat to...The transportation of oil and gas through pipelines is crucial for sustaining energy supply in industrial and civil sectors.However,the issue of pitting corrosion during pipeline operation poses an important threat to the structural integrity and safety of pipelines.This problem not only affects the longevity of pipelines but also has the potential to cause secondary disasters,such as oil and gas leaks,leading to environmental pollution and endangering public safety.Therefore,the development of a highly stable,accurate,and reliable model for predicting pipeline pitting corrosion is of paramount importance.In this study,a novel prediction model for pipeline pitting corrosion depth that integrates the sparrow search algorithm(SSA),regularized extreme learning machine(RELM),principal component analysis(PCA),and residual correction is proposed.Initially,RELM is utilized to forecast pipeline pitting corrosion depth,and SSA is employed for optimizing RELM’s hyperparameters to enhance the model’s predictive capabilities.Subsequently,the residuals of the SSA-RELM model are obtained by subtracting the prediction results of the model from actual measurements.Moreover,PCA is applied to reduce the dimensionality of the original 10 features,yielding 7 new features with enhanced information content.Finally,residuals are predicted by using the seven features obtained by PCA,and the prediction result is combined with the output of the SSA-RELM model to derive the predicted pipeline pitting corrosion depth by incorporating multiple feature selection and residual correction.Case study demonstrates that the proposed model reduces mean squared error,mean absolute percentage error,and mean absolute error by 66.80%,42.71%,and 42.64%,respectively,compared with the SSA-RELM model.Research findings underscore the exceptional performance of the proposed integrated approach in predicting the depth of pipeline pitting corrosion.展开更多
The increasingly complex and interconnected train control information network is vulnerable to a variety of malicious traffic attacks,and the existing malicious traffic detection methods mainly rely on machine learnin...The increasingly complex and interconnected train control information network is vulnerable to a variety of malicious traffic attacks,and the existing malicious traffic detection methods mainly rely on machine learning,such as poor robustness,weak generalization,and a lack of ability to learn common features.Therefore,this paper proposes a malicious traffic identification method based on stacked sparse denoising autoencoders combined with a regularized extreme learning machine through particle swarm optimization.Firstly,the simulation environment of the Chinese train control system-3,was constructed for data acquisition.Then Pearson coefficient and other methods are used for pre-processing,then a stacked sparse denoising autoencoder is used to achieve nonlinear dimensionality reduction of features,and finally regularization extreme learning machine optimized by particle swarm optimization is used to achieve classification.Experimental data show that the proposed method has good training performance,with an average accuracy of 97.57%and a false negative rate of 2.43%,which is better than other alternative methods.In addition,ablation experiments were performed to evaluate the contribution of each component,and the results showed that the combination of methods was superior to individual methods.To further evaluate the generalization ability of the model in different scenarios,publicly available data sets of industrial control system networks were used.The results show that the model has robust detection capability in various types of network attacks.展开更多
基金Supported by the Natural Science Foundation of Shandong Province of China(ZR2022QE091)the Special fund for Taishan Industry Leading Talent Project(tsls20230605)Key R&D Program of Shandong Province,China(2023CXGC010407).
文摘The transportation of oil and gas through pipelines is crucial for sustaining energy supply in industrial and civil sectors.However,the issue of pitting corrosion during pipeline operation poses an important threat to the structural integrity and safety of pipelines.This problem not only affects the longevity of pipelines but also has the potential to cause secondary disasters,such as oil and gas leaks,leading to environmental pollution and endangering public safety.Therefore,the development of a highly stable,accurate,and reliable model for predicting pipeline pitting corrosion is of paramount importance.In this study,a novel prediction model for pipeline pitting corrosion depth that integrates the sparrow search algorithm(SSA),regularized extreme learning machine(RELM),principal component analysis(PCA),and residual correction is proposed.Initially,RELM is utilized to forecast pipeline pitting corrosion depth,and SSA is employed for optimizing RELM’s hyperparameters to enhance the model’s predictive capabilities.Subsequently,the residuals of the SSA-RELM model are obtained by subtracting the prediction results of the model from actual measurements.Moreover,PCA is applied to reduce the dimensionality of the original 10 features,yielding 7 new features with enhanced information content.Finally,residuals are predicted by using the seven features obtained by PCA,and the prediction result is combined with the output of the SSA-RELM model to derive the predicted pipeline pitting corrosion depth by incorporating multiple feature selection and residual correction.Case study demonstrates that the proposed model reduces mean squared error,mean absolute percentage error,and mean absolute error by 66.80%,42.71%,and 42.64%,respectively,compared with the SSA-RELM model.Research findings underscore the exceptional performance of the proposed integrated approach in predicting the depth of pipeline pitting corrosion.
文摘The increasingly complex and interconnected train control information network is vulnerable to a variety of malicious traffic attacks,and the existing malicious traffic detection methods mainly rely on machine learning,such as poor robustness,weak generalization,and a lack of ability to learn common features.Therefore,this paper proposes a malicious traffic identification method based on stacked sparse denoising autoencoders combined with a regularized extreme learning machine through particle swarm optimization.Firstly,the simulation environment of the Chinese train control system-3,was constructed for data acquisition.Then Pearson coefficient and other methods are used for pre-processing,then a stacked sparse denoising autoencoder is used to achieve nonlinear dimensionality reduction of features,and finally regularization extreme learning machine optimized by particle swarm optimization is used to achieve classification.Experimental data show that the proposed method has good training performance,with an average accuracy of 97.57%and a false negative rate of 2.43%,which is better than other alternative methods.In addition,ablation experiments were performed to evaluate the contribution of each component,and the results showed that the combination of methods was superior to individual methods.To further evaluate the generalization ability of the model in different scenarios,publicly available data sets of industrial control system networks were used.The results show that the model has robust detection capability in various types of network attacks.
