Study of the corrosion characteristics of the metal materials of an aero‐engine under a marine atmosphere

Abstract

The corrosion tests are carried out on stainless steel, titanium alloy, copper alloy and high temperature alloy samples commonly used in an aviation engine. Scanning electron microscopy (SEM) and optical microscopy (OM) were used to observe the corrosion morphology of the corroded materials and 3D images after the corrosion products were removed. The corrosion rate was calculated by a weight loss method. A gradient boosting regression tree algorithm has good robustness for high‐dimensional data and is suitable for small‐sample data. The model can accurately predict the corrosion rate of metals and can analyze the factors that influence the corrosion rate.

Machine learning‐based analysis and prediction of the interfacial corrosion processes of copper cathode plates during the electrolytic production of copper powders

“…[32] Moreover, machine learning has been applied in the analysis of electrochemical corrosion processes for various materials. [33][34][35][36][37][38][39][40][41][42] For example, Liu et al [43] evaluated the impacts of different antifreezing solutions, corrosion times, flow velocities, and temperatures on the corrosion rates of Cu, Al, and steel alloys applied in heating tower heat pumps (HTHPs) using a support vector machine (SVM) and artificial neural network (ANN) algorithms, and the high reliability of the trained SVM and ANN models was validated based on experimentally measured corrosion rates. The benefits of the trained models are quite significant because the corrosion rates of these materials can be predicted accurately over a wide range of conditions without conducting any further experiments beyond those required for generating the data used for training the models, and these accurate predictions assist in improving the service life of HTHPs.…”