Please wait a minute...
工程设计学报
工程设计学报  2019, Vol. 26 Issue (4): 395-402    DOI: 10.3785/j.issn.1006-754X.2019.04.004
优化设计     
基于人工神经网络-遗传算法的展成法球面精密磨削参数优化
刘春青, 王文汉
中国航空工业集团公司 西安飞行自动控制研究所， 陕西 西安 710065
Parameter optimization of generating method spherical precision grinding based on ANN-GA
LIU Chun-qing, WANG Wen-han
Xi'an Flight Automatic Control Research Institute, Aviation Industry Corporation of China， Xi'an 71006, China
 全文: PDF(1511 KB)   HTML
摘要： 航空制造业的迅速发展对球面磨削加工精度提出了更高的要求，国内目前广泛依靠加工者经验选取磨削参数，工件球面轮廓度最高仅为2.2μm。为实现球面高精度加工，针对GCr15材料柱塞零件的展成法球面磨削加工，提出一种可靠的工艺参数优化方法。以砂轮速度、工件转速、进给速度为影响因素，以磨削后球面轮廓度和磨削加工时间为目标函数，设计了正交试验，分析了各磨削参数对球面轮廓度的影响程度。考虑到各磨削参数和球面轮廓度之间关系的复杂性，在正交试验的基础上利用遗传算法优化的BP（back propagation，反向传播）人工神经网络建立了球面轮廓度和各磨削参数之间的非线性映射关系，并进一步利用遗传算法进行极值寻优，得到最优的磨削参数组合。最后，根据最优磨削参数组合进行了多组验证实验。结果表明，使用优化后的磨削参数能够在保证加工效率的前提下将工件球面轮廓度提高到1.4μm，较传统经验法提高了36%。所提出的磨削参数优化方法可靠、有效，可直接应用于实际工程，对实现球面精密加工意义显著。
Abstract: With the rapid development of aviation manufacturing industry, higher requirements have been put forward for spherical grinding process. At present, grinding parameters are widely determined based on the processors’ experience, and the spherical contour can only be up to 2.2 μm. In order to achieve higher precision spherical machining, aiming at the generating method spherical grinding of GCr15 plunger parts, a feasible optimizing method for grinding parameters is provided. The orthogonal tests were designed and implemented by taking the grinding wheel speed, workpiece speed and feed speed as the influencing factors, and the spherical contour and consumed time as the objective function. The effects of different grinding parameters on the spherical contours were analyzed. Considering the complex relation between the grinding parameters and spherical contour, the back propagation (BP) artificial neural network (ANN) optimized by genetic algorithm (GA) was established to map the nonlinear relation between the spherical contour and grinding parameters based on the orthogonal experimental. Furthermore, the GA was used to optimize the extremum and locate the best grinding parameters combination. Finally, according to the optimal combination of grinding parameters, a number of validation experiments were carried out. According to the results, the spherical contour could be improved to 1.4 μm by using the optimized grinding parameters while remaining the machining efficient, which was 36% higher than that of traditional empirical method. The proposed optimization method of grinding parameters is reliable and effective, and can be directly applied to engineering, which is of great significance for realizing spherical precision machining.
收稿日期: 2019-03-04 出版日期: 2019-08-28
CLC:  V 261.25  
基金资助: 军委装备发展部预研先进制造技术专项（41423020501）
作者简介: 刘春青（1989—），男，陕西渭南人，工程师，硕士，从事机械工艺、液压偶件超精密加工研究，E-mail:happycqliu@163.com，https://orcid.org/0000-0003-3128-6905
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章　　

引用本文:

刘春青, 王文汉. 基于人工神经网络-遗传算法的展成法球面精密磨削参数优化[J]. 工程设计学报, 2019, 26(4): 395-402.

LIU Chun-qing, WANG Wen-han. Parameter optimization of generating method spherical precision grinding based on ANN-GA. Chinese Journal of Engineering Design, 2019, 26(4): 395-402.

链接本文:

https://www.zjujournals.com/gcsjxb/CN/10.3785/j.issn.1006-754X.2019.04.004    或    https://www.zjujournals.com/gcsjxb/CN/Y2019/V26/I4/395

No related articles found!