Corrosion rate prediction of 3C steel under different seawater environment by using support vector regression

The support vector regression (SVR) approach combined with particle swarm optimization (PSO) for its parameter optimization is proposed to establish a model for prediction of the corrosion rate of 3C steel under five different seawater environment factors, including temperature, dissolved oxygen, salinity, pH value and oxidation-reduction potential. The prediction results strongly support that the generalization ability of SVR model consistently surpasses that of back-propagation neural network (BPNN) by applying identical training and test samples. The absolute percentage error (APE) of 80.43% test samples out of 46 samples does not exceed 1% such that the best prediction result was provided by leave-one-out cross validation (LOOCV) test of SVR. These suggest that SVR may be a promising and practical methodology to conduct a real-time corrosion tracking of steel surrounded by complicated and changeable seawater.