Influence of Heat Treatment on Microstructure and Mechanical Properties of Plasma Sprayed FeCrMoCBY Amorphous Coatings

The changes of the microstructure and the mechanical properties of FeCrMoCBY amorphous coatings prepared by plasma spraying after heat treatment were investigated.300,400,500 and 600℃were selected as the heat treatment temperature,and the crystallization phenomenon occurred after the heat treatment at 600℃.The crystallization products of the coating heat-treated at 600℃were a-Fe and Fe23(C,B)6.Heat treatment was beneficial to the microhardness and the bonding strength of the coatings.The microhardness of the coating heat-treated at 600℃increased obviously,and the strongest bonding strength occurred in the coating heat-treated at 500℃.The improvement of the wear resistance of the coatings could attribute to heat treatment as well,and the wear resistance of the coating heat-treated at 600℃was the optimum,compared with the coating heat-treated at 500℃.

热处理对等离子喷涂铁基非晶合金涂层微观结构和耐腐蚀性能的影响

采用等离子喷涂技术在Q235钢基体上制备Fe48Cr15Mo14C15B6Y2非晶合金涂层,之后对涂层进行200,300,500,600,700℃热处理,研究了热处理对涂层微观结构、耐电化学腐蚀性能和耐均匀腐蚀性能的影响。结果表明:随着热处理温度的升高,涂层的非晶含量降低,孔隙率先减小后增大,经300℃热处理后涂层的孔隙率最低,且低于未热处理涂层的;热处理后涂层中的晶体相主要包括α-Fe,Fe-Cr,Fe63Mo37,Fe3C等;随着热处理温度的升高,涂层的自腐蚀电流密度先减小后增大,经300℃热处理后,自腐蚀电流密度最小,涂层的耐电化学腐蚀性能最好;经过热处理后,涂层在NaCl溶液中浸泡31d后的单位面积质量损失减小,且热处理温度越高,单位面积质量损失越小,涂层的耐均匀腐蚀性能提高。

Non-isothermal Crystallization and Thermal Transformation Behaviors of Plasma Sprayed Fe_(48)Cr_(15)Mo_(14)C_(15)B_(6)Y_(2)Amorphous Coating

The non-isothermal crystallization dynamic behavior and mechanism of plasma sprayed Fe_(48)Cr_(15)Mo_(14)C_(15)B_(6)Y_(2)coating were thoroughly studied.The phase transition and crystallization kinetics of the coating were elaborately investigated by differential scanning calorimetry(DSC),X-ray diffraction(XRD),scanning electron microscopy(SEM),and transmission electron microscopy(TEM).The findings reveal that the characteristic temperatures of the coating shift to an elevated temperature at a higher heating rate and the crystallization processes are thermally activated.The 3-rd step of crystallization processes is more susceptible to the continuously increased heating rate while the onset crystallization reaction is less sensitive to the continuously enhanced heating rate.Fe_(23)(C,B)_6 phase is inclined to precipitate than other crystal phases due to the substantial pre-generation ofα-Fe.The onset nucleation and growth ofα-Fe crystals is tough due to a higher onset apparent activation energy.Meanwhile,the transformation from Fe_(23)(C,B)_(6)to FeB is harder in comparison with the precipitation of other crystals.The most parts of the three crystallization processes are dominated by three-dimensional diffusion model due to the fact that most values of local Avrami exponent are higher than 2.5.

图卷积网络模型识别道路正交网格模式

街道网作为城市的骨架,其模式识别对于城市景观分析、市政规划、交通流量分析等都具有重要作用,是综合了道路几何特征、语义特征及上下文关系的智能识别问题。道路网模式识别目前主要是基于边-节点的图结构和道路网眼的多边形群结构两种模型,运用相关几何度量和统计指标通过模式识别规则探测获得。由于道路模式表达受人的空间认知、视觉心理影响,具有一定的不确定性和复杂性,基于规则推理与统计分析的方法很难获得与人工判断相一致的结果。在人工智能技术的驱动下,引入一种图结构上的深度学习模型,即图卷积网络,用于识别道路网正交网格模式。先以道路交叉点作为节点、道路连接作为边构建图结构,并划分出网格和非网格两类,作为模型的标注和预测目标;同时将道路网线性剖分以获取图节点特征,作为模型的输入信息;然后提取两端点都被预测为网格点的路段,实现网格模式的识别。实验结果表明,该方法中,道路网节点分类的准确率达到89.2%,道路段分类的准确率达到86.1%,能有效用于网格模式识别。