摘要
研究了高速条件下的滚齿工艺参数设置与优化方面的工作,采用新的非支配遗传算法NSGA-Ⅱ设计了相应的优化数学模型,优化达到最低能耗以及最长的刀具使用期限,再以遗传反向传播算法(GABP)神经网络为目标设置预测模型并建立适应度函数,完成迭代优化后获得匹配滚齿工艺的Pareto最优条件。研究结果表明:这里预测模型经过5次循环计算后,均方差为10-5,得到0.000425的最优值,推断上述网络满足良好的稳定性。刀具寿命误差相对后者降低16%,降低了36%的能量损耗,发现GABP算法具备更优收敛能力。Pareto解集获得了比相近加工样本集更优的性能,因此采用多目标优化模型可以确保加工能耗和刀具使用寿命同时达到最佳状态。该研究对提高的滚齿加工工艺参数以及提高机加工效率具有很好的实际应用价值。
Studied under the condition of high speed gear hobbing process parameter setting and optimization,the new non dominant genetic algorithm NSGA-Ⅱdesign,the optimization mathematical model was optimized to achieve the lowest energy consumption as well as the longest tool life,again with GABP neural networks as the goal set up the forecasting model and establish the fitness function,Pareto optimal conditions matching gear hobbing process were obtained after iterative optimization.The results show that the mean square error of the prediction model in this paper is equal to 10-5 after five cycles of calculation,and the optimal value of 0.000425 is obtained.It is concluded that the above network meets good stability.Compared with the latter,the tool life error is reduced by 16%,and the energy loss is reduced by 36%.It is found that the GABP algorithm has better convergence ability.The performance of Pareto solution set is better than that of similar machining sample set,so the multi-objective optimization model can ensure the optimal state of machining energy consumption and tool life simultaneously.The research has a good practical application value to improve the machining parameters and machining efficiency.
作者
班希翼
李强
贺小龙
余建勇
BAN Xi-yi;LI Qiang;HE Xiao-long;YU Jian-yong(College of Locomotive and Rolling Stock,Zhengzhou Vocational and Technical College of Railway,He’nan Zhengzhou 450016,China;School of Mechanical Engineering,He’nan University of Science and Technology,He’nan Zhengzhou 450016,China;Department of Mechanical Engineering,Chongqing University of Arts and Sciences,Chongqing 402160,China;Teaching Affairs Office,Zhengzhou Vocational College of Railway,He’nan Zhengzhou 450016,China)
出处
《机械设计与制造》
北大核心
2024年第10期145-148,156,共5页
Machinery Design & Manufacture
基金
河南省高等学校重点科研项目(22A520050)
河南省教育科学规划一般课题(2021YB0618)。
关键词
滚齿
工艺参数
BP神经网络
遗传算法
多目标优化
Gear Hobbing
Process Parameters
BP Neural Network
Genetic Algorithm
Multi-Objective Optimization
