摘要
采用磁场电沉积方法在40Cr钢表面制备了Ni-TiN镀层,并在正交实验的基础上建立了BP神经网络模型对镀层腐蚀速率进行预测,最后利用扫描电镜、X射线衍射仪以及显微电子天平对镀层的表面形貌、组分以及腐蚀速率进行分析和研究。结果表明,当工艺组合为A2B2C3D1,即TiN粒子浓度6g/L,磁场强度0.4T,占空比50%,电流密度0.5A/dm2时,Ni-TiN镀层经腐蚀后表面较为平整,凸起状物质较少。BP神经网络模型能够较好的模拟Ni-TiN镀层腐蚀速率,腐蚀速率最小值仅为2.134mg/m2·h,因此也证明了BP神经网络的可靠性。经XRD分析,Ni-TiN镀层存在Ni、TiN两相。
The Ni-TiN coatings on the surface of40Cr steel were prepared by magnetic electrodeposition, and on the basis of orthogonal experiment to establish the BP neural network model to predict the corrosion rate of the coating. Finally, the surface morphology, composition and corrosion rate of the coating were studied by scanning electron microscopy, X-ray diffraction and micro-electron balance. The results indicated that when the process combination for the ,TiN particle concentration of 6 g/L,magnetic field intensity of 0？4 T, Duty ratio of 50% , current density 0. 5 A/dm^2, the surface of Ni-TiN coatings was smooth and the convex shape material was less after corrosion. BP neural network model can simulate the corrosion rate of Ni- TiN coating better, and the corrosion rate is only 2. 134 mg/m^2· h, so it was proved that the BP neural network was reliable. Ni-TiN coatings contained Ni and TiN phases by XRD analysis.
出处
《人工晶体学报》
EI
CAS
CSCD
北大核心
2016年第10期2556-2560,共5页
Journal of Synthetic Crystals
基金
宁波市科技富民项目(2016C10056)
浙江省自然科学基金项目(LQ13F010004)
浙江省公益科技项目(2014C31162)
宁波大红鹰学院校级科研项目(1320133017)
作者简介
彭绪山（1966-）,男,湖北省人,副教授.
