镁合金交流等离子体表面处理膜层厚度的建模

In this paper, the modeling of a multi-variable complicated process in AZ91D magnesium alloy AC plasma surface treatment was carried out by artificial neural networks method. The results show that: The artificial neural networks technique can establish the relationship model between the coating thickness and the treatment parameters such as concentration of KF, concentration of KOH, voltage, treatment time and temperature of treatment solution perfectly. The maximum error between the desired test value and the output value is only 2.13%. This research lays a firm foundation for the development of AZ91D magnesium alloy eletrochemical treatment.