功能材料
功能材料
공능재료
JOURNAL OF FUNCTIONAL MATERIALS
2014年
13期
13079-13081
,共3页
BP神经网络模型%预测%Ni-TiN镀层
BP神經網絡模型%預測%Ni-TiN鍍層
BP신경망락모형%예측%Ni-TiN도층
BP artificial neural network%prediction%Ni-TiN composite coatings
采用脉冲电沉积方法,在45钢表面制备Ni-TiN镀层.采用X 射线衍射仪和扫描电镜对 Ni-TiN镀层进行组织结构和表面形貌分析,利用 BP 神经网络模型对Ni-TiN 镀层的耐腐蚀性能进行预测.结果表明,在Ni-TiN镀层中存在Ni和TiN相.衍射角2θ=44.80,52.23和76.75°分别对应于镍晶面的(111)、(200)和(220);而衍射角2θ=36.63,42.62和61.79°则分别对应于 TiN 晶面的(111)、(200)和(220).在TiN粒子浓度一定时,大电流密度和小占空比脉冲电沉积可获得表面致密、光滑,腐蚀坑较小的镀层.3×8×1型神经网络模型的预测结果与实测结果相差不大,最大误差为4.1%.
採用脈遲電沉積方法,在45鋼錶麵製備Ni-TiN鍍層.採用X 射線衍射儀和掃描電鏡對 Ni-TiN鍍層進行組織結構和錶麵形貌分析,利用 BP 神經網絡模型對Ni-TiN 鍍層的耐腐蝕性能進行預測.結果錶明,在Ni-TiN鍍層中存在Ni和TiN相.衍射角2θ=44.80,52.23和76.75°分彆對應于鎳晶麵的(111)、(200)和(220);而衍射角2θ=36.63,42.62和61.79°則分彆對應于 TiN 晶麵的(111)、(200)和(220).在TiN粒子濃度一定時,大電流密度和小佔空比脈遲電沉積可穫得錶麵緻密、光滑,腐蝕坑較小的鍍層.3×8×1型神經網絡模型的預測結果與實測結果相差不大,最大誤差為4.1%.
채용맥충전침적방법,재45강표면제비Ni-TiN도층.채용X 사선연사의화소묘전경대 Ni-TiN도층진행조직결구화표면형모분석,이용 BP 신경망락모형대Ni-TiN 도층적내부식성능진행예측.결과표명,재Ni-TiN도층중존재Ni화TiN상.연사각2θ=44.80,52.23화76.75°분별대응우얼정면적(111)、(200)화(220);이연사각2θ=36.63,42.62화61.79°칙분별대응우 TiN 정면적(111)、(200)화(220).재TiN입자농도일정시,대전류밀도화소점공비맥충전침적가획득표면치밀、광활,부식갱교소적도층.3×8×1형신경망락모형적예측결과여실측결과상차불대,최대오차위4.1%.
Ni-TiN composite coatings were prepared on 45 steel by using pulse electrodeposition.The microstruc-tures of Ni-TiN coatings were investigated using X-ray diffraction (XRD)and scanning electron microscopy (SEM),and a BP artificial neural network (ANN)model with 3×8×1 layers was used to predict the corrosion resistance of the coatings.The results illuminate that there was Ni and TiN phase in Ni-TiN coatings.For Ni, the diffraction peaks at 44.80,52.23 and 76.75°correspond to (111),(200)and (220).For TiN,the diffraction peaks at 36.63,42.62 and 61.79°correspond to (111),(200)and (220),respectively.At a given TiN concentra-tion,the coatings prepared with high current densities and shorter duty cycles have smooth,compact surface, and the pits are smaller.BP model shows that there was little difference between the predicted and measured re-sults,and the maximum relative error was 4.1%.