基于BP神经网络的Fe基合金粉末喷涂工艺参数优化

发布时间：2018-04-09 23:21

本文选题：超音速等离子喷涂　切入点：Fe基合金粉　出处：《表面技术》2015年09期

【摘要】：目的基于BP神经网络具有自学习、自训练和输出预测的功能,将其应用于热喷涂过程中的参数优化问题。方法依托高效能超音速等离子喷涂系统实验平台,以Fe基合金粉末为喷涂材料,将等离子喷涂中的主气流量、电功率和喷涂距离作为模型输入,涂层沉积速率和硬度作为模型输出,不断调整隐含层节点个数,最终建立3-7-2网络结构的BP神经网络以优化工艺参数。利用优化出的工艺参数制备Fe基合金涂层,测试其性能,并计算误差。结果神经网络优化出的最优喷涂工艺参数为:主气流量96L/min,电功率56 k W,喷涂距离95 mm。采用该工艺参数制备涂层,涂层增厚实测平均值为360μm,硬度为672HV0.3,而模型的预测值分别为332μm和611HV0.3,与预测值的相对误差分别为7.8%和9.1%。结论 BP神经网络对等离子喷涂参数优化问题的拟合精度比较高,误差在可以接受的范围之内。将BP神经网络运用于热喷涂工艺参数的优化具有科学性和可操作性。
[Abstract]:Aim based on the self-learning, self-training and output prediction functions of BP neural network, the BP neural network is applied to the optimization of parameters in thermal spraying process.Methods based on the experimental platform of high performance supersonic plasma spraying system, Fe-based alloy powder was used as the spraying material. The main gas flow, electric power and spraying distance in plasma spraying were input as the model.The coating deposition rate and hardness are taken as the output of the model, and the number of hidden layer nodes is constantly adjusted. Finally, the BP neural network of 3-7-2 network structure is established to optimize the process parameters.The Fe-base alloy coating was prepared by using the optimized process parameters. The properties of the coating were tested and the error was calculated.Results the optimum spraying parameters were obtained as follows: main gas flow rate 96L / min, electric power 56kW, spraying distance 95 mm.The average thickness of coating is 360 渭 m, the hardness is 672HV0.3, and the predicted values of the model are 332 渭 m and 611HV0.3, respectively. The relative error between the model and the predicted value is 7.8% and 9.1%, respectively.Conclusion the fitting accuracy of BP neural network for plasma spraying parameter optimization is high and the error is within acceptable range.It is scientific and operable to apply BP neural network to the optimization of thermal spraying process parameters.
【作者单位】：装甲兵工程学院;河北工业大学;
【基金】：国家自然科学基金面上项目(51175513)~~
【分类号】：TG174.4;TP183

【共引文献】