Corrosion in seawater is simultaneously influenced by multiple environmental factors including dissolved oxygen (DO), temperature, salinity, pH, and so on. These factors vary along with time and are different in different locations. The spatial-temporal variation of the actual marine environment cannot be ignored in corrosion prediction models. This paper proposes a new method for corrosion prediction in the actual time-varying marine environment which includes the design of experiments, calibration of acceleration models, and the modeling of marine environment. Acceleration models capture the effects of environmental factors and acts as the link between the environment and the corrosion process. The marine environment is described with the Kriging spatial-temporal model. Then the proposed method is used to give corrosion predictions for metals in different locations and vessels travel in different waters. This method could be helpful for corrosion resistance evaluation and environment corrosivity assessment.

海水中的腐蚀同时受到多种环境因素的影响，包括溶解氧（DO），温度，盐度，pH等。这些因素随时间变化，并且在不同的位置也不同。腐蚀预测模型中不能忽略实际海洋环境的时空变化。本文提出了一种在实际时变海洋环境中进行腐蚀预测的新方法，包括实验设计，加速度模型的标定和海洋环境建模。加速模型捕获环境因素的影响，并充当环境与腐蚀过程之间的链接。用克里格时空模型描述海洋环境。然后，将所提出的方法用于对不同位置的金属以及船只在不同水域中的航行提供腐蚀预测。该方法可能有助于抗腐蚀性能评估和环境腐蚀性评估。