【作者】 李琦；

【导师】 张洪潮；

【作者基本信息】 大连理工大学 ， 机械制造及其自动化， 2019， 硕士

【摘要】 激光熔覆作为机械产品表面改性和修复的重要技术,在国民生产和机械制造中有着重要的地位,在航空航天、石油装备、船舶、汽车等领域有广泛应用。而Ni基合金激光熔覆层具备良好的耐热性、抗氧化性和耐蚀性能,是表面改性和修复的重要金属材料之一。在表面改性和修复工艺中,熔覆层的形貌尺寸难以准确预测,并且激光熔覆得到的Ni基合金熔覆层容易出现裂纹,这是影响表面改性和修复技术的主要问题。因此对Ni基合金熔覆层进行形貌和裂纹的研究是十分必要的。本文在分析激光工艺参数对熔覆层高度、宽度和面积影响规律的基础上,建立了以加工工艺参数对熔覆层高度、宽度、和面积进行预测的基于误差反向传播(BP)的神经网络预测模型。依据输入量与输出量特点设计了神经网络结构,通过激光熔覆试验采集熔覆层高度、宽度和面积值样本,利用训练集训练BP神经网络,建立输入量与输出量之间的映射关系,并用测试集对训练好的预测模型进行检验。测试结果表明,基于BP神经网络的预测模型对熔覆层形貌尺寸的预测精度较高。基于实验及微观检测分析,以Ni60A熔覆层为研究对象,从组织和应力方面对激光熔覆过程中熔覆层裂纹的开裂及扩展理论进行了研究。综合熔覆层微观组织、物相以及断口形貌分析发现,熔覆层不同位置组织差异显著,材料硬度与裂纹敏感性随硬质相分布的增加而升高。Ni60A熔覆层的裂纹属于准解离的脆性断裂,裂纹萌生于熔覆层顶部,并沿硬质相向内部规律性扩展。本文根据热残余应力模型与Griffith脆性断裂强度模型建立了熔覆层开裂判据,为裂纹的产生的影响因素提供理论依据。此外,通过实验分别从改变工艺参数、减小熔覆过程中的温度梯度、以及更换基体材料等方面对Ni60A激光熔覆层裂纹产生的影响因素进行探究。实验同时也验证了熔覆层裂纹的开裂及扩展理论的合理性,并在此基础上提出相应的裂纹控制方法。更多还原

【Abstract】 As an important technology for surface modification and repair of mechanical products,laser cladding has an important position in national production and machinery manufacturing,and is widely used in aerospace,petroleum equipment,ships,automobiles and other fields.The Ni-based alloy laser cladding layer has good heat resistance,oxidation resistance and corrosion resistance,and is one of the important metal materials for surface modification and repair.In the surface modification and repair process,the shape of the cladding layer is difficult to accurately predict,and the Ni-based alloy cladding layer obtained by laser cladding is prone to cracking,which is the main problem affecting the surface modification and repair technology.Therefore,it is necessary to study the morphology and crack of the Ni-based alloy cladding layer.Based on the analysis of the influence of laser process parameters on the height,width and area of the cladding layer,a neural network based on error back propagation(BP)based on the processing parameters to predict the height,width and area of the cladding layer is established.The neural network structure is designed according to the input quantity and output quantity characteristics.The height,width and area value samples of the cladding layer are collected by laser cladding experiment.The BP neural network is trained by the training set to establish the mapping relationship between input quantity and output quantity.The test set is used to test the trained prediction model.The test results show that the prediction model based on BP neural network has higher prediction accuracy for the size of the cladding layer.Based on the experimental and microscopic analysis,the Ni60 A cladding layer was taken as the research object,the cracking and expansion theory of the cladding crack during laser cladding was studied from the aspects of crystal structure and stress.The microstructure,phase and fracture morphology of the cladding layer showed that the crystal structure of the cladding layer was significantly different at different locations.The hardness and crack sensitivity of the cladding layer increased with the increase of the hard phase distribution.The crack of the Ni60 A cladding layer belongs to Quasi-dissociated brittle fracture,cracking on the top of the cladding layer and extending regularly along the hard phase.Based on the thermal residual stress model and the Griffith brittle fracture strength model,the cladding layer cracking criterion is established to provide a theoretical basis for crack generation.In addition,the factors affecting the cracking of Ni60 A laser cladding layer were studied by changing the process parameters,reducing the temperature gradient during the cladding process,and replacing the matrix material.The experiment also verified the rationality of the cracking and expansion theory of the cladding crack,and proposed the corresponding crack elimination method.更多还原