摘要
为了提高智能车间生产流程优化等级,采用物联网监测智能车间生产情况,根据监测数据建立生产流程优化目标函数。建立基于最小完工时间和最小生产成本的双目标函数,获得物联网监测实时数据样本,采用卷积神经网络对数据样本进行训练,获得目标函数的最优参数,最后求解最小完工时间和最小生产成本。通过差异化设置卷积核尺寸,验证不同卷积核尺寸的完工时间和生产成本,选择适合生产流程优化的卷积核尺寸。通过最小完工时间和最小生产成本二维可视化,可获得生产流程最优值,将卷积神经网络算法和其他常用优化算法分别进行实例仿真。试验结果表明,该文算法优势明显,在生产流程优化方面适用性强。
In order to improve the production process optimization level of intelligent workshop,the production situation of intelligent workshop is monitored by the Internet of things,and the production process optimization objective function is established according to the monitoring data.Firstly,a double objective function based on the minimum completion time and the minimum production cost is established,and the real-time monitoring data samples of the Internet of things are obtained.The convolution neural network is used to train the data samples to obtain the optimal parameters of the objective function.Finally,the minimum completion time and minimum production cost are solved.By setting the size of convolution kernel differently,the completion time and production cost of different convolution kernel size are verified,and the convolution kernel size suitable for production process optimization is selected.Through the two-dimensional visualization of the minimum completion time and the minimum production cost,the optimal value of the production process can be clearly obtained.The convolution neural network algorithm and other commonly used optimization algorithms are simulated respectively.The results show that the algorithm here has obvious advantages and strong applicability in production process optimization.
作者
吴昌钱
杨旺功
罗志伟
Wu Changqian;Yang Wanggong;Luo Zhiwei(School of Computer Information,Minnan Science and Technology University,Quanzhou 362000,China;School of Information,Beijing Forestry University,Beijing 100083,China;School of Mechanical and Automotive Engineering,Xiamen University of Technology,Xiamen 361000,China)
出处
《南京理工大学学报》
CAS
CSCD
北大核心
2021年第5期589-595,共7页
Journal of Nanjing University of Science and Technology
基金
福建省教育科学“十三五”规划(FJJKCG20-014)
福建省中青年科技类科研项目(JAT190931)
福建省本科高校教育教学改革研究项目(FBJG20200327)
福建省自然科学基金(2019J01863)。
关键词
生产流程
物联网
卷积神经网络
完工时间
生产成本
production process
internet of things
convolutional neural network
completion time
production cost
