基于RBF神经网络的Ni-TiN镀层耐蚀性能预测研究

Prediction of Corrosion Resistance of Ni-TiN Coating Based on RBF Neural Network

采用脉冲电沉积方法在40Cr钢表面制备Ni-TiN复合镀层,并以TiN粒子浓度、电流密度以及占空比为输入层,以Ni-TiN复合镀层腐蚀量为输出层,建立RBF神经网络模型,对镀层腐蚀量进行预测研究,最后利用扫描电镜观察不同工艺参数下镀层表面形貌。结果表明,RBF神经网络对镀层腐蚀量有较强的预测能力,其预测值与实验值相对误差最小仅为0.73%;SEM分析表明,当TiN粒子浓度10 g/L,电流密度5 A/dm2,占空比60%时,Ni-TiN复合镀层经腐蚀后表面较为平整,腐蚀坑较少,耐腐蚀性能较好。

The Ni-TiN composite coatings were deposited on the surface of 40 Cr steel by pulse electrodeposition method. The RBF neural network model was established by using the TiN particle concentration,current density and duty cycle as the input layer,and the corrosion rate of Ni-TiN composite coating as the output layer. The corrosion rate of the coating was predicted,and the surface morphology of the coating was observed under the scanning electron microscope. The results show that the corrosion loss weight of the coating was predicted really by the RBF neural network,and the min relative error was only 0. 73% between the predicted value and the experimental value; SEM analysis show that when the concentration of TiN particles 10 g / L,current density of 5 A / dm^2,duty cycle 60%,Ni-TiN composite coating on the surface after the corrosion was flat,less corrosion pits,and the corrosion resistance was better.