摘要
针对粉末分装系统精度不高、加样策略的相关研究较少以及高精度算法理论较多但实物偏少等问题,文中提出了一种高精度粉末分装系统。文中阐述了系统总体设计,介绍了机械结构设计、电子设计以及软件设计思路与方法。对于系统固有缺陷提出了一种LSTM(Long Short-Term Memory)模型来优化加样策略,并使用土壤样本对系统进行性能测试实验。结果表明,系统分装样品的精度可达±1 mg。所提系统可应用于高精度粉末分装场景,并为分装系统和加样策略设计提供理论参考。
In view of the problems such as low precision of powder packing system,less research on adding sample strategy and more high-precision algorithm theory but less real objects,a high-precision powder packing system is proposed in this study.The overall design of the system is described,and the ideas and methods of mechanical structure design,electronic design and software design are introduced.A LSTM(Long Short Term Memory)model is proposed to optimize the sampling strategy for the inherent defects of the system.Soil samples are used to test the performance of the system.The results show that the accuracy of the system can reach±1 mg.Therefore,the proposed system can be applied to high-precision powder packaging scenario,and provides theoretical reference for the design of packaging system and sampling strategy.
作者
郑阳
袁旭军
张荣福
陈奇
ZHENG Yang;YUAN Xujun;ZHANG Rongfu;CHEN Qi(School of Optical-Electrical and Computer Engineering,University of Shanghai for Science andTechnology,Shanghai 200093,China;Shanghai Cohere Electronics Technology Co.,Ltd.,Shanghai 201612,China)
出处
《电子科技》
2025年第10期73-81,共9页
Electronic Science and Technology
基金
基础科研条件与重大科学仪器设备研发计划(2022YFF0706003)。
关键词
动态称重
机器学习
LSTM
加样
高精度
控制系统
土壤
优化设计
dynamic weighing
machine learning
LSTM
sample addition
high precision
control system
soil
optimal design
