目的 针对传统温度阈值监测方法在复杂工况与动态负载下适应性不足,且无法预警早期故障的问题,提出一种基于反向传播神经网络的肘杆机构温度异常监测方法。方法 采集模切机肘杆机构多工况运行数据,构建正常运行的温度变化数据集。利用反向传播神经网络强大的非线性拟合能力,建立温度特征与复杂工况的映射模型。以模型预测温度与实际监测温度的残差作为异常判据,实现故障的早期预警。结果 在测试集中温度预测误差±1 ℃的样本占总样本的99.39%,基于该预测结果设定的动态阈值,能够满足温度预测的准确度需求,且温度异常监测模型能够有效识别温度异常,误报率和漏报率分别为0.17%和3.11%。结论 基于反向传播神经网络的模切机肘杆机构温度异常监测方法,将温度监测从传统的阈值检测模式转化为温度行为的动态监测,从而在复杂动态工况下实现对早期故障的敏锐感知,显著提升了监测系统的自适应能力与工况鲁棒性。
Abstract
The work aims to propose a BP neural network-based method for monitoring temperature anomalies in toggle mechanisms to address the limitations of traditional temperature threshold monitoring, namely, poor adaptability under complex operating conditions and dynamic loads, and the inability to detect early-stage faults. Operational data from toggle mechanism of a die-cutting machine under diverse conditions were collected to construct a dataset of normal temperature variations. Leveraging the BP network's strong nonlinear fitting capability, a mapping model was established between temperature behaviors and operational parameters. The prediction error was used as an anomaly indicator to enable early fault detection. Experimental results showed that 99.39% of test samples exhibited prediction errors within ±1 °C. The resulting dynamic thresholds met the required accuracy for temperature prediction, and the model effectively identified anomalies, achieving a false positive rate of 0.17% and a false negative rate of 3.11%. This approach shifts temperature monitoring from static thresholding to dynamic behavioral modeling, enabling sensitive identification of early-stage faults under variable operating conditions, significantly enhancing system adaptability and robustness.
关键词
神经网络 /
肘杆机构 /
异常监测 /
模切机
Key words
neural network /
toggle mechanism /
anomaly monitoring /
die-cutting machine
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] ZHANG J X, ZHAO H, CHEN K D, et al.Dexterous Hand towards Intelligent Manufacturing:A Review of Technologies, Trends, and Potential Applications[J]. Robotics and Computer-Integrated Manufacturing, 2025, 95:103021.
[2] ZHANG H T, SEMUJJU S D, WANG Z C, et al.Large Scale Foundation Models for Intelligent Manufacturing Applications:A Survey[J]. Journal of Intelligent Manufacturing, 2026, 37(1):119-170.
[3] 刘名轩, 刘东, 韦树远, 等. 高速全清废模切机关键技术与研究进展[J]. 绿色包装, 2025(2):22-26.
LIU M X, LIU D, WEI S Y, et al.Key Technologies and Research Progress of High-Speed Full Waste Removal Die-Cutting Machines[J]. Green Packaging, 2025(2):22-26.
[4] SCAIFE A D.Improve Predictive Maintenance through the Application of Artificial Intelligence:A Systematic Review[J]. Results in Engineering, 2024, 21:101645.
[5] MALLIORIS P, AIVAZIDOU E, BECHTSIS D.Predictive Maintenance in Industry 4.0:A Systematic Multi-Sector Mapping[J]. CIRP Journal of Manufacturing Science and Technology, 2024, 50:80-103.
[6] 韦树远, 关景果, 陈光, 等. 大幅面高速重载模切机动平台传动肘杆分析及设计[J]. 包装工程, 2023, 44(17):206-212.
WEI S Y, GUAN J G, CHEN G, et al.Analysis and Design of Transmission Elbow Bar for Large Format High-Speed and Heavy-Duty Die-Cutting Mobile Platform[J]. Packaging Engineering, 2023, 44(17):206-212.
[7] 韦树远, 陈光, 李鹏, 等. 高速重载模切机动平台运动分析及结构优化设计[J]. 包装工程, 2023, 44(7):264-269.
WEI S Y, CHEN G, LI P, et al.Motion Analysis and Structure Optimization Design of Moving Platform of High-Speed Heavy-Duty Die-Cutting Machine[J]. Packaging Engineering, 2023, 44(7):264-269.
[8] DEVASENAN M, MADHAVAN S.Thermal Intelligence:Exploring AI's Role in Optimizing Thermal Systems-a Review[J]. Interactions, 2024, 245(1):282.
[9] BAMROONGSHAWGASAME T, ZHANG X, LI Q L.Emerging Trends and Applications in Thermal Imaging Using Infrared Detectors:A Review[J]. IEEE Sensors Journal, 2026, 26(2):1520-1532.
[10] ZHU J, WANG B T, JI J, et al.Fiber Optic Array Temperature Field Measurement System for High-Temperature Turbine Blades[J]. IEEE Transactions on Instrumentation and Measurement, 2024, 73:7007208.
[11] HIDALGO-LÓPEZ J A, ROMERO-SÁNCHEZ J, FERNÁNDEZ-RAMOS R, et al. A Low-Cost, High-Accuracy Temperature Sensor Array[J]. Measurement, 2018, 125:425-431.
[12] STEFANIDIS G D, MERCI B, HEYNDERICKX G J, et al.CFD Simulations of Steam Cracking Furnaces Using Detailed Combustion Mechanisms[J]. Computers & Chemical Engineering, 2006, 30(4):635-649.
[13] 王文斌. 高精度测长机温度场建模与热变形补偿技术研究[J]. 仪器仪表用户, 2026, 33(2):20-22.
WANG W B.Research on Temperature Field Modeling and Thermal Deformation Compensation Technology for High-Precision Length Measuring Machines[J]. Instrumentation Customer, 2026, 33(2):20-22.
[14] PANG P, ZHENG J, ZHAO Y H, et al.Thermal-Vibration Correlation Study for High-Temperature Superconducting Maglev Intelligent Monitoring Based on back Propagation Neural Network Analysis[J]. Superconductor Science and Technology, 2024, 37(2):025011.
[15] AL-GABALAWY M, ELMETWALY A H, YOUNIS R A, et al.Temperature Prediction for Electric Vehicles of Permanent Magnet Synchronous Motor Using Robust Machine Learning Tools[J]. Journal of Ambient Intelligence and Humanized Computing, 2024, 15(1):243-260.
[16] ZHU M R, YANG Y, FENG X B, et al.Robust Modeling Method for Thermal Error of CNC Machine Tools Based on Random Forest Algorithm[J]. Journal of Intelligent Manufacturing, 2023, 34(4):2013-2026.
[17] 李学娟. 高压开关柜无线温度监测与热故障预警技术[J]. 煤矿机电, 2022, 43(3):64-67.
LI X J.Wireless Temperature Monitoring and Thermal Fault Early Warning Technology of High Voltage Switchgear[J]. Colliery Mechanical & Electrical Technology, 2022, 43(3):64-67.
[18] LI W F, WANG Y Q.Intelligent Temperature Control Method of Instrument Based on Fuzzy PID Control Technology[J]. International Journal of Advanced Computer Science and Applications, 2024, 15(1):927.
[19] 韩冰, 孙允伟, 何茜儒. 基于温度智能传感的变压器安装状态监测与运行优化方法[J]. 电气技术与经济, 2026(2):394-397.
HAN B, SUN Y W, HE Q R.Transformer Installation State Monitoring and Operation Optimization Method Based on Intelligent Temperature Sensing[J]. Electrical Equipment and Economy, 2026(2):394-397.
[20] YANG Z Q, LIU B B, ZHANG Y R, et al.Intelligent Sensing of Thermal Error of CNC Machine Tool Spindle Based on Multi-Source Information Fusion[J]. Sensors, 2024, 24(11):3614.
基金
2025年支持企业加强研发能力建设项目（25YFYFFG00580）
PDF(1736 KB)
渝公网安备 50010702501716号 渝ICP备15012534号-2
