基于鱼群优化BP神经网络的切削温度建模  

Cutting Temperature Modeling Based on Fish Swarm Optimization BP Neural Network

作　　者：王晔 马廉洁[1,2] 左宇辰 刘涛 白威[1] 常昊 WANG Ye;MA Lian-jie;ZUO Yu-chen;LIU Tao;BAI Wei;CHANG Hao(School of Mechanical Engineering and Automation,Northeastern University,Shenyang 110819,China;School of Control Engineering,Northeastern University at Qinhuangdao,Qinhuangdao Hebei 066004,China)

机构地区：[1]东北大学机械工程与自动化学院,沈阳110819 [2]东北大学秦皇岛分校控制工程学院,河北秦皇岛066004

出　　处：《组合机床与自动化加工技术》2019年第11期22-24,共3页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家自然科学基金资助项目(51275083);河北省自然科学基金资助项目(E2018501078)

摘　　要：针对BP神经网络易陷入局部最优值的缺点,采用人工鱼群算法进行优化。通过分析切削温度随工艺参数的变化趋势,结合最小二乘拟合法,建立切削速度-切削温度、进给速度-切削温度、切削深度-切削温度的一元模型。根据一元模型提出了切削温度的多元模型假设,并用遗传算法求解最终获得切削温度的多元模型,试验结果表明:模型具有可靠性。Aiming at the shortcoming of BP neural network that is easy to fall into local optimum value,artificial fish swarm algorithm was used to optimize it.By analyzing the changing trend of cutting temperature with process parameters,Combined with least squares fitting method,the one-dimensional model of cutting speed-cutting temperature,feed speed-cutting temperature,cutting depth-cutting temperature were established.Based on the one-dimensional model,the hypothesis of multi-element model of cutting temperature was put forward,and the multi-element model of cutting temperature was finally obtained by genetic algorithm.The experimental results show that the model is reliable.

关 键 词：切削温度 鱼群优化 BP神经网络 

分 类 号：TH161[机械工程—机械制造及自动化] TG506[金属学及工艺—金属切削加工及机床]