Automatic identification of corrosive factors categories according to the environmental factors

Abstract

Time of wetness, and pollutants are three key factors for the selection of metal materials in engineering applications and the determination of atmospheric corrosivity categories. In the past, when one or more corrosive factors data is missing, corrosive factors categories were often subjectively determined according to expert experience. In order to overcome such difficulty, this paper presents a method to automatically determine corrosive factors categories using detected environmental factors data instead of expert scoring. In this method, Bayesian network was used to build the mathematical model. And the inference was obtained by clique tree algorithm. The validity of the model and algorithm was verified by the simulation results.