首页|基于BA-PNN算法与数字孪生的车间扰动判定方法

基于BA-PNN算法与数字孪生的车间扰动判定方法

扫码查看
随着科学技术的发展,生产安全和车间管理问题越来越受到重视。传统车间在管理上多依靠人工,使得车间扰动事件发现不及时,扰动认定不清楚,不利于迅速解决扰动事件和保障人员设备安全。为提高管理效率和保障安全,提出一种基于蝙蝠算法优化的概率神经网络(Bat Algorithm-Probabilistic Neural Network,BA-PNN)算法和数字孪生的车间扰动判定方法。首先通过传感器采集数据并对其进行分析和预处理;随后搭建传统概率神经网络(Probabilistic Neural Net-work,PNN)模型和以算法识别率为优化目标的BA-PNN扰动判定模型,并结合数字孪生技术将BA-PNN模型融入孪生平台;最后通过仿真与结果分析,对比优化前模型效果及孪生平台特点,该模型识别效果较之前显著提高,证明了方法的有效性。
Workshop disturbance detection method based on BA-PNN algorithm and digital twin technology
With the progress of society,production safety and workshop management issues are receiving increasing attention.The management of traditional workshops mostly relies on manual work,and managers often fail to timely detect and identify disturb-ances in the workshop.This is not conducive to quickly resolving disturbance events and ensuring the safety of personnel and e-quipment.In order to improve management efficiency and ensure security,a disturbance determination method based on BA-PNN algorithm and digital twin technology was proposed.First,data were collected through sensors,the data were analyzed and prepro-cessed.Subsequently,the traditional PNN model and BA-PNN disturbance judgment model were built,and the latter aims at the optimization of algorithm recognition rate.BA-PNN model was integrated into twin platform by digital twin technology.Finally,through simulation and result analysis,compared with the model effect before optimization,the recognition effect of the model is significantly improved,while reflecting the characteristics of digital twins,proving the effectiveness of the method.

probabilistic neural networkbat algorithmdigital twindisturbance event

张若语、胡友民、吴波、杨晔、秦峻峰

展开 >

华中科技大学机械科学与工程学院,武汉 430070

概率神经网络 蝙蝠算法 数字孪生 扰动事件

2024

现代制造工程
北京机械工程学会 北京市机械工业局技术开发研究所

现代制造工程

CSTPCD北大核心
影响因子:0.374
ISSN:1671-3133
年,卷(期):2024.(3)
  • 16