大功率激光立体成形熔道几何参数量化模型

Quantitative model of clad track geometric parameters of high power laser solid forming

大功率激光立体成形过程中,单道熔道的形貌特征与成形件的精度、质量指标等因素密切相关。为建立激光立体成形1Cr13工艺参数与熔道的宽高比、深高比等形状参量的量化分析模型,基于正交法研究了激光功率、扫描速度和送粉量等3个因素与熔道形貌特征的定性关系。实验表明:3个因素对宽高比的影响为扫描速度最大,送粉量次之,激光功率最小;3个因素对深高比的影响为送粉量最大,激光功率次之,扫描速度最小。为进一步优化工艺参数取值范围,以Box-Behnken实验方案补充实验样本,建立工艺参数与熔道形状参量的BP神经网络计算模型。进而以神经网络模型为计算工具,以熔道宽高比、深高比为约束条件建立激光立体成形熔道形貌控制工艺窗口,不同工艺分区的实验结果与计算结果一致。

In the process of high power laser solid forming,the clad parameters are closely related to the accuracy and quality of the forming parts.In order to establish a quantitative analysis model for the laser solid forming process variables of 1Cr13 and the clad parameters,such as the ratio of width to height and the ratio of depth to height,the qualitative relationship between laser power,scanning speed,powder feed rate and the clad parameters is built by the orthogonal experiment.The results show that laser scanning speed has most significant influence on the ratio of width to height,powder feeding rate is the secondly important factor and laser power has least influence.Powder feeding rate has most significant influence on the ratio of depth to height,laser power is the secondly important factor and laser scanning speed is has least influence.In order to optimize the range of process variables,the BP neural network model is trained based on the experimental results of the orthogonal experiment and Box-Behnken experiment.It is developed to express the relationship between the clad process variables and the clad parameters.Furthermore,the BP neural network model is used as the computing tool,and the shape control process window of the laser solid forming is established with the ratio of width to height and the ratio of depth to height.The experimental results of different process window zones are in good agreement with the calculated results.