An artificial neural network method is proposed to correlate the relationship between the corrosion rate of 3C steels with seawater environment factors. The predictions with the unseen test data are in good agreement with experimental values. Further, the developed model used to simulate the combined effect of environmental factors (temperature, dissolved oxygen, salinity, pH values, and oxidation-reduction potential) on the corrosion rate. 3D mappings remarkably reveal the complex interrelationship between the input environmental parameters on the output corrosion rate. The quantitative estimation of corrosion by virtual addition/subtraction of environmental factors individually to a hypothetical system helps to understand the impact of each parameter.

提出了一种人工神经网络方法，将3C钢的腐蚀速率与海水环境因素之间的关系进行关联。带有看不见的测试数据的预测与实验值非常吻合。此外，开发的模型用于模拟环境因素（温度，溶解氧，盐度，pH值和氧化还原电位）对腐蚀速率的综合影响。3D映射显着揭示了输入环境参数与输出腐蚀速率之间的复杂相互关系。通过将环境因素虚拟添加/减去到假设的系统中，对腐蚀进行定量估计有助于了解每个参数的影响。