|
| 单道激光熔覆Inconel 625熔覆层尺寸预测 |
| Prediction of Size for Single Pass Inconel 625 Clads Deposited by Laser Cladding |
| |
| DOI：10.3969/j.issn.1674-6457.2022.09.015 |
| 中文关键词: 激光熔覆 神经网络 中心复合设计 Inconel 625 熔覆层尺寸 |
| 英文关键词: laser cladding neural network central composite design Inconel 625 cladding layer size |
| 基金项目:重庆市轻合金材料与加工工程技术研究中心开放基金(GCZX202001);陕西省自然科学基础研究计划(2020JQ– 780);国家自然科学基金青年项目(51901180) |
| Author Name | Affiliation | | WEI Wen-lan | School of Mechanical Engineering, Xi'an Shiyou University, Xi'an 710065, China | | WEI Ze-bing | School of Mechanical Engineering, Xi'an Shiyou University, Xi'an 710065, China | | GUO Long-long | School of Mechanical Engineering, Xi'an Shiyou University, Xi'an 710065, China
Chongqing Engineering Technology Research Center for Light Alloy and Processing, Chongqing Three Gorges University, Chongqing 404100, China | | LIU Hong-liang | Baoji Oilfield Machinery Co., Ltd., Shaanxi Baoji 721002, China | | CAO Jia-chen | School of Mechanical Engineering, Xi'an Shiyou University, Xi'an 710065, China | | ZHANG Yi-wei | Chongqing Engineering Technology Research Center for Light Alloy and Processing, Chongqing Three Gorges University, Chongqing 404100, China | | QU Hao | School of Mechanical Engineering, Xi'an Shiyou University, Xi'an 710065, China |
|
| Hits: 2018 |
| Download times: 747 |
| 中文摘要: |
| 目的 准确预测激光熔覆Inconel 625熔覆层尺寸。方法 以送粉速率、扫描速度和激光功率为试验变量，以熔覆层的宽度和高度为评价指标，结合中心复合试验设计方法进行试验设计，开展单道激光熔覆试验，探究工艺参数对单道熔覆层尺寸的影响规律，并建立以工艺参数为输入、熔覆层尺寸为输出的BP神经网络模型，利用粒子群算法对BP神经网络模型进行优化，对比分析优化前后模型的预测效果。结果 激光功率对熔覆层宽度的影响最显著，其次是扫描速度，最后是送粉速率;扫描速度对熔覆层高度的影响最显著，其次是激光功率，最后是送粉速率;粒子群算法优化BP神经网络预测模型对熔覆层尺寸的预测精度较高，熔覆层宽度和高度的测量值和预测值之间的平均相对误差分别为4.238%和2.910%。结论 研究成果可以为激光熔覆Inconel 625熔覆层尺寸的调控和预测提供参考。 |
| 英文摘要: |
| The work aims to accurately predict the size of cladding layer of Inconel 625 laser cladding. With powder feeding rate, scanning speed and laser power as the test variables, and width and height of the cladding layer as the evaluation indicators, combined with the center composite experimental design method, the experimental design was carried out, and the single-channel laser cladding test was carried out, the effect law of process parameters on the size of single cladding layer was explored. The BP neural network model was established with the process parameters as the input and the size of the cladding layer as the output. The particle swarm algorithm was used to optimize the BP neural network, and the prediction effect of the model before and after the optimization was compared and analyzed. The laser power had the most significant effect on the width of the cladding layer, followed by the scanning speed, and finally the powder feeding rate. The scanning speed has the most significant effect on the height of the cladding layer, followed by the laser power and finally the powder feeding rate. The BP neural network prediction model optimized by particle swarm optimization had high prediction accuracy for size of cladding layer. The average relative errors between the measured and predicted widths and heights of cladding layer were 4.238% and 2.910% respectively. The research results can provide reference for size regulation and prediction of cladding layer of Inconel 625 laser cladding. |
|
View Full Text
View/Add Comment Download reader |
| Close |
|
|
|
渝公网安备 50010702501719号