基于气压特征融合的包裹碰撞与挤压识别

段晨昊, 朱磊, 张媛, 杜艳平

包装工程(技术栏目) ›› 2026, Vol. 47 ›› Issue (17) : 223-233.

PDF(1831 KB)
PDF(1831 KB)
包装工程(技术栏目) ›› 2026, Vol. 47 ›› Issue (17) : 223-233. DOI: 10.19554/j.cnki.1001-3563.2026.17.023
自动化与智能化技术

基于气压特征融合的包裹碰撞与挤压识别

  • 段晨昊1a,b, 朱磊1a,b*, 张媛1a,b, 杜艳平1b
作者信息 +

Recognition of Parcel Collision and Squeeze Based on Air Pressure Feature Fusion

  • DUAN Chenhao1a,b, ZHU Lei1a,b*, ZHANG Yuan1a,b, DU Yanping1b
Author information +
文章历史 +

摘要

目的 为解决非可视物流运输中包裹破损追溯困难、内部缓冲结构失效评估缺乏客观依据的问题,提供一种低成本、抗干扰的包裹受力状态与受损方位精准监测手段。方法 搭建模拟真实物流包裹的准密闭腔体实验平台,在其内部对称部署高频动态与低频静态气压传感器,采集单侧撞击、双侧挤压等典型工况下的气压信号。首先,利用基于物理能量分布先验的自适应算法对原始高噪声信号进行清洗。随后,将对齐后的多模态气压序列输入所构建的压力上下文卷积循环神经网络(PC-CRNN)中,通过残差卷积提取空间特征并结合双向长短期记忆网络解析时序演变规律,最终完成受力状态的识别。结果 该系统有效克服了传统方法难以辨识“双侧连续挤压”与“单侧强撞击”的混淆局限。在包含多种受力状态的复杂工况下,系统综合识别准确率达到88.95%。尤其是针对破坏性极强的“双侧挤压”工况,其识别准确率较无时序特征提取的基线模型大幅提升了23.8%。结论 双尺度气压特征融合与时空深度学习架构成功实现了包裹定向碰撞与挤压的高精度辨识,为物流安全监测提供了一种低成本、高鲁棒性且支持多点受力分析的技术方案。

Abstract

The work aims to propose a low-cost and anti-interference precise monitoring method for the force state and damage location of packages to deal with the problems of difficult traceability of package damage and the lack of an objective basis for assessing the failure of internal buffer structures in non-visual logistics transportation. A quasi-sealed chamber experimental platform that simulated real logistics packages was built, and high-frequency dynamic and low-frequency static barometric pressure sensors were symmetrically deployed inside it to collect pressure signals under typical working conditions such as single-side impact and double-side compression. Firstly, the original high-noise signal was cleaned through an adaptive algorithm based on the prior physical energy distribution. Then, the aligned multimodal pressure sequence was input into the constructed pressure-context convolutional recurrent neural network (PC-CRNN). This network utilized residual convolutions to extract spatial energy fingerprints and employed a bidirectional long short-term memory (Bi-LSTM) network to parse the temporal pulse features of multi-point impacts. Experimental results demonstrated that this system effectively solved the confusion of traditional methods in distinguishing between "bilateral continuous squeezes" and "unilateral strong impacts". The comprehensive recognition rate under complex working conditions reached 88.95%. Notably, the recognition accuracy for bilateral squeezes increased by 23.8% compared with a baseline model lacking temporal feature extraction. The dual-scale pneumatic feature fusion and spatio-temporal deep learning architecture have successfully achieved high-precision identification of package directional collision and compression, providing a low-cost, highly robust technical solution for logistics safety monitoring and multi-point force analysis.

关键词

物流安全监测 / 双尺度气压感知 / 时空特征融合 / 压力上下文卷积循环神经网络(PC-CRNN) / 双向长短期记忆网络(Bi-LSTM)

Key words

logistics safety monitoring / dual-scale air pressure sensing / spatiotemporal feature fusion / pressure-context convolutional recurrent neural network (PC-CRNN) / bidirectional long short-term memory (Bi-LSTM)

引用本文

导出引用
段晨昊, 朱磊, 张媛, 杜艳平. 基于气压特征融合的包裹碰撞与挤压识别[J]. 包装工程. 2026, 47(17): 223-233 https://doi.org/10.19554/j.cnki.1001-3563.2026.17.023
DUAN Chenhao, ZHU Lei, ZHANG Yuan, DU Yanping. Recognition of Parcel Collision and Squeeze Based on Air Pressure Feature Fusion[J]. Packaging Engineering. 2026, 47(17): 223-233 https://doi.org/10.19554/j.cnki.1001-3563.2026.17.023
中图分类号: TB48   

参考文献

[1] 蒲丽琼. 数字经济背景下电子商务物流效率优化策略探析[J]. 电子商务评论, 2024, 13(3): 4363-4369
PU L Q.Exploration of Optimization Strategies for E-Commerce Logistics Efficiency Under the Digital Economy Background[J]. E-Commerce Review, 2024, 13(3): 4363-4369.
[2] 杨慧敏, 高小雯, 李瑞涛, 等. 基于改进YOLOv8n 的快递包裹缺陷检测方法研究[J]. 电子测量技术, 2026, 49(3): 66-76.
YANG H M, GAO X W, LI R T, et al.(2026). Research on Defect Detection Method for Express Parcels Based on Improved YOLOv8n[J]. Electronic Measurement Technology, 2026, 49(3): 66-76.
[3] 伍瑾, 卢富德, 王彪, 等. 易损件在连续跌落冲击载荷作用下的动力学响应[J]. 振动与冲击, 2023, 42(20): 275-279.
WU J, LU F D, WANG B, et al.Dynamic Response of Vulnerable Part Under the Continuous Drop Impact Loads. Journal of Vibration and Shock, 2023, 42(20): 275-279.
[4] 贾杏歌, 孙伟, 周俊丽, 等. 液晶显示屏包装件跌落仿真分析及结构优化[J]. 包装工程, 2021, 42(1): 143-149.
JIA X G, SUN W, ZHOU J L, et al.Drop Simulation Analysis and Structural Optimization of LCD Package. Packaging Engineering, 2021, 42(1): 143-149.
[5] 刘凯旋, 郑华栋, 陈鹿民, 等. 经济型快递物流安检标记机器人设计与研究[J]. 包装工程, 2022, 43(9): 163-170.
LIU K X, ZHENG H D, CHEN L M, et al.Design and Research of an Economical Courier Security Marking Robot. Packaging Engineering, 2022, 43(9): 163-170.
[6] PARK J, LEE S, KIM S, et al.Stress Analysis on Palletized Unit-Load of Various Pallet Stacking Patterns in Corrugated Packages[J]. Korean Journal of Packaging Science & Technology, 2024, 30(3): 241-248.
[7] 巩桂芬, 张鼎乾, 江轲, 等. 微拱滚筋结构对三片罐罐身抗压强度影响的试验与仿真研究[J]. 包装工程, 2025, 46(5): 235-241.
GONG G F, ZHANG D Q, JIANG K, et al.Experiments and Simulations on the Compressive Strength of Three-Piece Can Body by Micro-Arch Reinforcement Structure. Packaging Engineering, 2025, 46(5): 235-241.
[8] 陈新发, 叶天, 李玉龙, 等. 基于构型设计的包装容器缓冲效应数值分析[J]. 北京理工大学学报, 2019, 39(4): 352-358.
CHEN, X F, YE T, LI Y L, et al.Numerical Simulation on Cushion Effect of Packaging Container with Configuration Design[J]. Transactions of Beijing Institute of Technology, 2019, 39(4): 352-358.
[9] SEQUEIRA J S.IMU Networks for Trajectory Reconstruction in Logistics Applications[J]. Sensors, 2023, 23(18): 7838.
[10] DING A, ZHANG Y, ZHU L, et al.Intelligent Recognition of Rough Handling of Express Parcels Based on CNN-GRU with the Channel Attention Mechanism[J]. Journal of Ambient Intelligence and Humanized Computing, 2023, 14(2): 973-990.
[11] QIN F, ZHANG Y, ZHU L, et al.Multilayer Perceptron with Analysis of Variance for Detection of Rough Handling in the Small Household Appliances Express Service[C]// 2023 6th International Conference on Electronics Technology. New York: ICET, 2023: 1193-1198.
[12] ZHANG Y, YANG W, HUANG H, et al.A Lightweight Framework for Recognizing Rough Handling Behaviours in Parcel Logistics[J]. Measurement, 2026(259): 119693.
[13] POZOS-ESTRADA O.Influence of Entrapped Air on Hydraulic Transients During Rapid Closure of a Valve Located Upstream and Downstream of an Air Pocket in Pressurised Pipes[J]. Water, 2025, 17(7): 927.
[14] MARYKOVSKIY Y, CLARK T, DEPARDAY J, et al.Architecting a Digital Twin for Wind Turbine Rotor Blade Aerodynamic Monitoring[J]. Frontiers in Energy Research, 2024(12): 1428387.
[15] CANNARD C, DELORME A, WAHBEH H.HRV and EEG Correlates of Well-Being Using Ultra-Short, Portable, and Low-Cost Measurements[J]. Progress in Brain Research, 2024(287): 91-109.
[16] BARBER S, DEPARDAY J, MARYKOVSKIY Y, et al.Development of a Wireless, Non-Intrusive, MEMS-Based Pressure and Acoustic Measurement System for Large-Scale Operating Wind Turbine Blades[J]. Wind Energy Science, 2022(7): 1383-1398.
[17] MALLEGNI N, MOLINARI G, RICCI C, et al.Sensing Devices for Detecting and Processing Acoustic Signals in Healthcare[J]. Biosensors, 2022, 12(10): 835.
[18] LIU W, CHEN J, YE H, et al.Multifunctional Sensors Made with Conductive Microframework and Biomass Hydrogel for Detecting Packaging Pressure and Food Freshness[J]. ACS Applied Materials & Interfaces, 2024, 16(8): 10785-10794.
[19] GUPTA P, YOUNG A, RAO A. Investigating Cargo Loss in Logistics Systems Using Low-Cost Impact Sensors[EB/OL]. (2022-01-02)[2025-08-29]. https://arxiv.org/abs/2201.00301.
[20] PATRIZI G, CARRATÙ M, CIANI L, et al.Analysis of Inertial Measurement Units Performances Under Dynamic Conditions[J]. IEEE Transactions on Instrumentation and Measurement, 2023(72): 1-13.
[21] HOLMVALL M, HOLMGREN M.Applied Design of Corrugated Board Properties to Mitigate Puncturing in Parcel Handling Distribution Chains[J]. BioResources, 2025, 20(3): 5377-5397.
[22] FRANK B.Corrugated Box Compression—A Literature Survey[J]. Packaging Technology and Science, 2014, 27(2): 105-128.
[23] AGUILAR-GONZÁLEZ, A., MEDINA SANTIAGO, A. Road Event Detection and Classification Algorithm Using Vibration and Acceleration Data[J]. Algorithms, 2025, 18(3): 127.
[24] ZHANG X, SONG Z, PARK D M, et al.An IMU-Based Machine Learning System for Container Collision Position Identification[J]. Journal of Marine Science and Engineering, 2025, 13(6): 1144.
[25] 孙德强, 李想, 郝乾崇, 等. 基于纸张本构模型的瓦楞纸板边压强度仿真分析[J]. 包装工程, 2025, 46(01): 155-162.
SUN D Q, LI X, HAO Q C, et al.Simulation and Analysis of Edge Compression Strength of Corrugated Paper Fiberboard Based on Original Paper Constitutive Model. Packaging Engineering, 2025, 46(1): 155-162.
[26] 李国志, 李元吉, 孙德强, 等. 基于缓冲包装系统的气体采样设备改进设计与验证[J]. 包装工程, 2025, 46(3): 245-252.
LI G Z, LI Y J, SUN D Q., et al.Improved Design and Verification of Gas Sampling Equipment Based on Cushion Packaging System. Packaging Engineering, 2025, 46(3): 245-252.
[27] HORVATH S, KOVACS J G.Real-Time Product Weight Estimation Based on Internal Pressure Monitoring in Injection Molding[J]. Polymer Engineering & Science, 2025, 65(4): 1693-1701.
[28] 刘博文, 龙仁荣, 张庆明, 等. 舱内爆炸角隅汇聚反射冲击波超压特性研究[J]. 爆炸与冲击, 2023, 43(1): 35-51.
LIU B W, LONG R R, ZHANG Q M, et al.Study on the Corner Overpressure Characteristics of Concentrated Reflected Shock Wave Due to Internal Blast in Cabin. Explosion and Shock Waves, 2023, 43(1): 35-51.

基金

北京印刷学院科研平台建设项目(KYCPT202507)

PDF(1831 KB)

Accesses

Citation

Detail

段落导航
相关文章

/

〈 〉