王春,马立峰,贾伟涛,等.高温镁合金板材超声振动辅助剪切工艺参数优化与质量预测[J].精密成形工程,2024,16(8):28-40.
WANG Chun,MA Lifeng,JIA Weitao,et al.Process Parameter Optimization and Quality Prediction for Ultrasonic Vibration-assisted Shearing of High-temperature Magnesium Alloy Plates[J].Journal of Netshape Forming Engineering,2024,16(8):28-40. |
| 高温镁合金板材超声振动辅助剪切工艺参数优化与质量预测 |
| Process Parameter Optimization and Quality Prediction for Ultrasonic Vibration-assisted Shearing of High-temperature Magnesium Alloy Plates |
| 投稿时间：2024-03-26 |
| DOI：10.3969/j.issn.1674-6457.2024.08.004 |
| 中文关键词: 超声振动辅助剪切 响应面法 多目标优化 改进灰狼算法 最小二乘支持向量回归 |
| 英文关键词: ultrasonic vibratory-assisted shear response surface methodology multi-objective optimization improved grey wolf algorithm least squares support vector regression |
| 基金项目:国家自然科学基金(U1910213，52105388，52075357，52375362);山西省科技创新人才团队专项计划(202204051002002);山西省重点研发计划(202102050201005);山西省专利推广实施资助研究项目(202303004) |
|
| 摘要点击次数: 258 |
| 全文下载次数: 44 |
| 中文摘要: |
| 目的 为实现镁合金板材超声振动辅助剪切过程的工艺参数优化和质量预测，提出了一种多目标优化及决策方法，建立了多目标质量预测模型。方法 综合考虑经济成本及断面质量等因素，确定了最大剪切力、塌角宽度和毛刺高度为剪切质量的评价指标。依据正交实验结果建立了剪切质量关于各工艺参数的数学响应模型并进行分析，采用改进的多目标灰狼优化(MOGWO)算法进行寻优，利用基于熵权-TOPSIS法对优化后的非支配解进行综合评价，获得了最优工艺参数解。最后基于多输入多输出最小二乘支持向量回归机(MLSSVR)，以3个工艺参数为输入、3个评价指标为输出建立了剪切质量预测模型。结果 在此范围内，最大剪切力和塌角宽度主要受剪刃间隙和超声振幅的影响，随剪刃间隙的增大而增大，随超声振幅的增大而减小;而毛刺高度则主要受板材温度的影响。在此基础上，根据剪切质量的数学响应模型确定最佳工艺参数为:板材温度为250 ℃、剪刃间隙为8%板厚、超声振幅为10 μm。MLSSVR质量模型精度达95%以上，并且能综合考虑各工艺参数与各评价指标之间的影响，具有较高的预测精度和泛化能力。结论 超声振动辅助剪切(UVS)可有效提高板材剪切断面质量，所提方法对于寻求最佳板材剪切工艺参数具有一定的参考价值。 |
| 英文摘要: |
| To realize the optimization of process parameters and quality prediction of magnesium alloy plate in ultrasonic vibration-assisted shear process, the work aims to propose a multi-objective optimization and decision-making method and establish a multi-objective quality prediction model. Considering the economic costs and section quality, the maximum shear force, the rollover width and the burr height were determined as the evaluation indexes of shear quality. Based on the results of orthogonal experiments, the mathematical response model of shear quality on each process parameter was established for analysis, the improved multi-objective grey wolf optimization (MOGWO) algorithm was used for optimization, the TOPSIS method based on entropy weights was used for the comprehensive evaluation of optimized non-dominated solutions, and the optimal process parameter solutions were obtained. Finally, based on the multiple-input and multiple-output least squares support vector regression machine (MLSSVR), a shear quality prediction model was established with three process parameters as inputs and three evaluation indexes as outputs. In this range, the maximum shear force and the rollover width were mainly affected by the shear edge clearance and the ultrasonic amplitude, which increased with the increase of the shear edge clearance and decreased with the increase of the ultrasonic amplitude, while the burr height was mainly affected by the plate temperature. On this basis, according to the mathematical response model of shear quality, the optimal process parameters were determined to be a plate temperature of 250 ℃, a shear edge clearance of 8% plate thickness, and an ultrasonic amplitude of 10 μm. The accuracy of the MLSSVR quality model reached more than 95%, and it could comprehensively take into account the effect of each process parameter and the evaluation indexes, with high prediction accuracy and generalization ability. Ultrasonic vibration-assisted shearing (UVS) can effectively improve the shear section quality of the plate, and the proposed method has certain reference value for seeking the best shear process parameters of the plate. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
渝公网安备 50010702501719号