期刊文献+

基于BP神经网络和FPA的高速干切滚齿工艺参数低碳优化决策 被引量:6

Low carbon optimization decision for high-speed dry hobbing process parameters based on BP neural networks and FPA
下载PDF
导出
摘要 为解决高速干切滚齿工艺参数决策中存在的主观依赖性强和用时较长的问题,并实现滚齿加工低碳化,提出一种基于实例推理和优化算法的高速干切滚齿工艺参数低碳优化决策方法。利用反向传播(back propagation,BP)神经网络构建加工效果评价值的预测模型,通过改进K-means聚类算法获取待决策工艺问题的相似实例抽取集,以此构建待优化工艺参数约束,再运用花朵授粉算法(flower pollination algorithm,FPA),以碳耗最小为优化目标,获取待决策工艺问题的最优工艺参数。以某企业高速干切滚齿机为例,验证了该方法的可行性和有效性。使用该方法生成的工艺参数,加工效果更好,碳耗更低,可避免对工艺手册或个人经验的依赖,提高决策效率。研究结果有利于高速干切滚齿机的低碳运行,对机械制造企业实现低碳制造具有一定的参考意义。 Aiming at some problems of high subjective dependence and long time consuming in the process of high speed dry hobbing process parameters decision,a low carbon optimization decision method for high speed dry hobbing process parameters based on case-based reasoning(CBR)and optimization algorithms was proposed.At the same time,it was a method to achieve the low carbon of the hobbing processing.At the beginning,a back propagation(BP)neural networks model was established based on the cases of high speed hobbing process,which could predict the machining effect evaluation of hobbing processing.In addition,an improved K-means algorithm was used to obtain the similarity example extraction set for target process problem,and several process solutions were obtained to construct process parameter constraints.Moreover,the flower pollination algorithm(FPA)was applied to search the optimal process parameters for target process problems,which took the minimum carbon consumption of the hobbing processing as the optimization objective.A high speed dry hobbing machine in an enterprise was used as an instance to verify the feasibility and effectiveness of proposed method.The experimental results indicate that the proposed optimization method is a very useful tool for achieving lower energy consumption and better processing effect.The method can also effectively avoid relying on process manuals,personal experience or cutting experiments so as to improve decision efficiency.Moreo-ver,the results also show that it is conducive to achieve high performance and low carbon operation of high speed dry cutting hobbing machine,which can provide important reference value for machinery manufacturing enterprises to achieve low carbon manufacturing.
出处 《工程设计学报》 CSCD 北大核心 2017年第4期449-458,共10页 Chinese Journal of Engineering Design
基金 国家自然科学基金资助项目(51575071)
关键词 高速干切滚齿 工艺参数 低碳 BP神经网络 花朵授粉算法 high speed dry hobbing process parameters low carbon BP neural networks flower pollination algorithm
  • 相关文献

参考文献10

二级参考文献103

共引文献194

同被引文献37

引证文献6

二级引证文献27

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部