Plasma spray coating properties frequently depend -to a great extent- on the spray parameters. However, it is difficult to analyze and obtain a comprehensive model of the entire plasma spray process due to the complex chemical and thermodynamic reactions that take place during the process. In this study, Ni60CuMo coatings were prepared on ZL109 substrates. A Back Propagation (BP) Neural Network model in the artificial neural network was used to predict the change in bonding strength, microhardness, and porosity of the coatings under different spraying distances, spraying powers, and powder feeding rates. The results show that the R-value of the trained network training is 0.8828. Comparison of experimental and predicted results reveals that both show similar trends, which verifies that the BP model can effectively predict the properties of Ni-based coatings.

等离子喷涂性能经常在很大程度上取决于喷涂参数。然而，由于过程中发生复杂的化学和热力学反应，很难分析和获得整个等离子喷涂过程的综合模型。在本研究中，在 ZL109 基材上制备了 Ni60CuMo 涂层。使用人工神经网络中的反向传播 (BP) 神经网络模型来预测涂层在不同喷涂距离、喷涂功率和送粉速率下的结合强度、显微硬度和孔隙率的变化。结果表明，R- 训练后的网络训练值为 0.8828。实验结果和预测结果的比较表明，两者表现出相似的趋势，这验证了 BP 模型可以有效地预测镍基涂层的性能。