Corrosion Fatigue Mechanism And Corrosion Fatigue Life Prediction Of Microalloyed 780 MPa High-strength Marine Engineering Steel

This manuscript reveals the corrosion fatigue behavior and mechanism of 780 MPa high-strength marine engineering steel in the marine environment through axial stress corrosion fatigue experiment.The microalloyed 780 MPa marine engineering steels with different content of Cu,Sb,Nb elements are prepared by vacuum smelting technology.The influence of adding microalloyed elements Cu,Sb,Nb on the corrosion fatigue behavior and mechanism of 780 MPa marine engineering steel in the marine environment is observed by the method of microstructure characterization,electrochemistry,mechanics-electrochemistry,axial stress corrosion fatigue test,and etc.And the corrosion fatigue life prediction model is established based on machine learning method and characteristic engineering,and the main controlling factors and key element characteristics to the corrosion fatigue damage process of high-strength marine engineering steel is determined therefore.The main results are as listed follows:Above all,the corrosion fatigue behavior and mechanism of 780 MPa marine engineering steel in marine environment has been obtained systematically.It is found that the fatigue limit of 780 MPa marine engineering steel in air is 686 MPa or so;the logarithm of corrosion fatigue life in seawater has a negative linear correlation with the fatigue load,and there is no obvious corrosion fatigue limit.The corrosion fatigue crack initiation and propagation mechanism of 780 MPa marine engineering steel in marine environment is controlled by the combination of anodic dissolution and hydrogen induced cracking.Corrosion ions can promote corrosion fatigue crack initiation and propagation in areas with heavy strain accumulation such as grain boundaries.At the same time,hydrogen generated by electrochemical hydrogen evolving is prone to gather around special areas such as grain boundaries,promoting hydrogen-caused corrosion fatigue crack initiation and propagation.Secondly,the addition of microalloying elements Cu,Sb and Nb has a positive effect on lengthening the corrosion fatigue life of 780 MPa marine engineering steel in the marine environment.It is observed that the microalloying of Cu,Sb and Nb can refine the grain of 780 MPa marine engineer steel,lower the proportion of large angle grain boundaries,and reduce the strain accumulation inside the structure.The addition of Nb can inhibit the electrochemical cathodic hydrogen evolving process,and the addition of Cu and Sb together can restrain the electrochemical anodic dissolving process.When adding the microalloying elements Cu,Sb and Nb,it can optimize the microstructure and inhibit the electrochemical cathodic and anodic reactions,which is the precondition to enhance the corrosion fatigue resistance of 780 MPa marine engineering steel in the marine environment.Thirdly,the effect of adding Nb on the corrosion fatigue mechanism of 780 MPa marine engineering steel has been obtained systematically.This manuscript found that the addition of Nb can lower the corrosion fatigue sensitivity of highstrength engineering steel in seawater as well as in hydrogen filled environment.In the marine environment,hydrogen has little influence on the corrosion fatigue crack initiation of 780 MPa marine engineering steel;on the contrary,anodic dissolution plays a major role in causing the initiation and propagation of corrosion fatigue cracks.When Nb is added,the content of hydrogen inside the steel is brought down by restraining the electrochemical hydrogen evoluving reaction.Meanwhile,the NbC nano-precipitates generated after adding Nb act as irreversible hydrogen traps,which can reduce the accumulation of hydrogen atoms in special areas such as grain boundaries.Therefore,hydrogen induced cracking of 780 MPa marine engineering steel is inhibited.Then,the effect of adding Cu and Sb together on the corrosion fatigue mechanism of 780 MPa marine engineering steel has been systematically acquired.The addition of Cu and Sb can significantly reduce the corrosion fatigue sensitivity of 780 MPa marine engineering steel in the marine environment;adding 0.9%Cu and 0.1%Sb can achieve the best effect.Sb can cooperate with Cu in the steel to inhibit electrochemical anodic active dissolution and cathodic hydrogen evolution.In the meantime,Sb,in the form of insoluble oxides Sb2O5 and Sb2O3,accumulating in the inner rust layer can improve the protection ability of the rust layer,so as to optimize the distribution of Cl-and inhibit the mechanical-electrochemical effect in the process of corrosion fatigue.This can slow down the local anodic dissolution and hydrogen induced cracking and inhibit the crack initiation and propagation under the rust layer,thus improving the corrosion fatigue resistance.Last but not least,the classification model and prediction model of corrosion fatigue life of high-strength marine engineering steel has been established by support vector classification model and support vector regression model;the main controlling factors influencing the process of corrosion fatigue damage of highstrength marine engineering steel have been analyzed through correlation screening,recursive screening and exhaustive screening;the influence of element characteristics on corrosion fatigue life prediction model has been analyzed on the premise of unifying the mechanical factors and the environmental factors.It is observed that there are 4 main controlling factors to the corrosion fatigue classification model(whether its corrosion fatigue life is≥107),namely,Clconcentration,Cr content,load,and Pot.There are 6 main controlling factors to the corrosion fatigue life prediction model,with the order of importance:load,Cu content,Pot,Cl-concentration,Cr content,Sb content.The influence of 14 elements on the accuracy of the corrosion fatigue life prediction model with the order of importance:Cu,Sb,Mo,Cr,Ni,Nb,Mn,C,S,Ti,P,Al,Fe,Si,which is in good agreement with the test results,proving the accuracy of the corrosion fatigue life prediction model.