目的 针对包装机轴承早期故障难检测的问题,研究蝴蝶-粒子群混合算法（Butterfly-Particle Swarm Hybrid Algorithm,BOA-PSO）与双向长短时记忆网络（Bidirectional Long Short Term Memory Network,BiLSTM）的包装机轴承故障诊断方法,以提升设备运维效率。方法 鉴于BiLSTM参数设置对最终诊断结果影响较大的问题,通过融合蝴蝶优化算法（Butterfly Optimization Algorithm,BOA）和粒子群算法（Particle Swarm Optimization,PSO）的优势,提出BOA-PSO算法,实现BiLSTM参数自适应选取,从而更快更准识别包装轴承故障状态。结果 BOA-PSO-BiLSTM对7种轴承状态的平均诊断精度达到了98.43%,在优化算法相同、识别方法不同的对比中,较其余4种方法至少提高了2.57%;在优化算法不同、模式识别方法相同的对比中,较其余4种优化BiLSTM的方法至少提高3%。结论 通过BOA-PSO-BiLSTM可有效解决包装机轴承在复杂工况下的诊断难题,为包装机轴承预知性维护提供了智能诊断方案。
Abstract
To address the difficult early detection of faults in packaging machine bearings, the work aims to study a fault diagnosis method for packaging machine bearings based on the butterfly-particle swarm hybrid algorithm (BOA-PSO) and bidirectional long short-term memory network (BiLSTM) to improve the efficiency of equipment operation and maintenance. Considering the significant impact of BiLSTM parameter settings on the final diagnosis results, the BOA-PSO algorithm was proposed by integrating the advantages of the butterfly optimization algorithm (BOA) and the particle swarm optimization (PSO) algorithm to achieve adaptive selection of BiLSTM parameters, thereby enabling faster and more accurate identification of packaging bearing fault states. The average diagnosis accuracy of BOA-PSO-BiLSTM for 7 bearing states reached 98.43%. In the comparison with the same optimization algorithm but different recognition methods, it was at least 2.57% higher than the other four methods. In the comparison with different optimization algorithms but the same pattern recognition method, it was at least 3% higher than the other four optimized BiLSTM methods. BOA-PSO-BiLSTM can effectively solve the diagnosis problems of packaging machine bearings in complex working conditions at an early stage, providing an intelligent diagnosis solution for predictive maintenance of packaging machine bearings.
关键词
双向长短时记忆网络 /
蝴蝶优化算法 /
粒子群算法 /
包装机轴承 /
故障诊断
Key words
bidirectional long short-term memory /
butterfly optimization algorithm /
particle swarm optimization /
packing machine bearing /
fault diagnosis
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 车畅, 李明辉, 马晨佩, 等. 包装机深沟球轴承多模态智能诊断模型与自适应参数优化研究[J]. 包装工程, 2025, 46(19): 247-257.
CHE C, LI M H, MA C P, et al.Multimodal Intelligent Diagnosis Model and Adaptive Parameter Optimization of Deep Groove Ball Bearings in Packaging Machines[J]. Packaging Engineering, 2025, 46(19): 247-257.
[2] 方群玲, 马智宇, 张锐, 等. MDCNet轴承智能故障诊断方法研究[J]. 包装工程, 2023, 44(9): 218-223.
FANG Q L, MA Z Y, ZHANG R, et al.Intelligent Fault Diagnosis Method of Bearing Based on MDCNet[J]. Packaging Engineering, 2023, 44(9): 218-223.
[3] 张秋昕, 周进军, 陈锋, 等. 基于DS-PCA模型的包装设备滚动轴承故障诊断方法研究[J]. 包装工程, 2023, 44(9): 224-231.
ZHANG Q X, ZHOU J J, CHEN F, et al.Fault Diagnosis Method of Rolling Bearings of Packaging Equipment Based on DS-PCA Model[J]. Packaging Engineering, 2023, 44(9): 224-231.
[4] 张璇, 龚廷恺, 李志农, 等. 基于MOMEDA结合软阈值处理的行星轮轴承故障诊断[J]. 振动与冲击, 2026, 45(4): 286-295.
ZHANG X, GONG T K, LI Z N, et al.Fault Diagnosis of Planetary Bearings Based on MOMEDA with Soft Threshold Processing[J]. Journal of Vibration and Shock, 2026, 45(4): 286-295.
[5] 赵孝礼, 姚建勇, 邓文翔, 等. 基于GTCNs的航空发动机高速轴承智能故障诊断[J]. 振动、测试与诊断, 2026, 46(1): 70-76.
ZHAO X L, YAO J Y, DENG W X, et al.Intelligent Fault Diagnosis of Aero-Engine High-Speed Bearings Using Graph Transfer Convolutional Networks under Fluctuating Working Conditions[J]. Journal of Vibration, Measurement & Diagnosis, 2026, 46(1): 70-76.
[6] 何勇, 刘晓玲. 基于参数自适应FMD和SDAE的变负载下轴承故障诊断[J]. 振动与冲击, 2026, 45(2): 189-200.
HE Y, LIU X L.Fault Diagnosis of Bearings under Varying Loads Based on Parameter Adaptive FMD and SDAE[J]. Journal of Vibration and Shock, 2026, 45(2): 189-200.
[7] 冀相冰, 朱艳辉, 徐啸, 等. 基于注意力机制的包装命名实体识别[J]. 包装工程, 2019, 40(15): 24-29.
JI X B, ZHU Y H, XU X, et al.Packaging Named Entity Recognition Based on Attention Mechanism[J]. Packaging Engineering, 2019, 40(15): 24-29.
[8] 蒋开正, 吕丽平. 自适应变异麻雀算法优化LSTM变速箱故障诊断[J]. 噪声与振动控制, 2023, 43(4): 129-134.
JIANG K Z, LYU L P.Gearbox Fault Diagnosis Based on LSTM Optimized by Adaptive Mutation Sparrow Search Algorithm[J]. Noise and Vibration Control, 2023, 43(4): 129-134.
[9] 任义, 倪瑶, 栾方军, 等. 基于时空特征融合VFMCNN-BiLSTM滚动轴承故障诊断方法[J]. 机床与液压, 2026, 54(2): 7-17.
REN Y, NI Y, LUAN F J, et al.Rolling Bearing Fault Diagnosis Method Based on VFMCNN-BiLSTM Fusion of Spatiotemporal Feature[J]. Machine Tool & Hydraulics, 2026, 54(2): 7-17.
[10] 郭明军, 陈昕昀, 石淇, 等. 基于NRBO-Transformer- BiLSTM的柔性薄壁轴承细粒度故障诊断[J]. 现代制造工程, 2025(11): 136-145.
GUO M J, CHEN X Y, SHI Q, et al.Fine-Grained Fault Diagnosis of Flexible Thin-Walled Bearings Based on NRBO-Transformer-BiLSTM[J]. Modern Manufacturing Engineering, 2025(11): 136-145.
[11] 李清, 吴杏, 周晓君. 基于优化双向长短时记忆网络的刀具磨损状态识别[J]. 机械设计与研究, 2023, 39(4): 147-151.
LI Q, WU X, ZHOU X J.Tool Wear State Recognition Based on Optimization Bidirectional Long Short Term Memory Network[J]. Machine Design & Research, 2023, 39(4): 147-151.
[12] 萧彦, 任江波, 何亚坤, 等. 基于多传感器成像与CNN的电网继保故障检测[J]. 传感器与微系统, 2026, 45(3): 172-178.
XIAO Y, REN J B, HE Y K, et al.Power Grid Relay Protection Fault Detection Based on Multi-Sensor Imaging and CNN[J]. Transducer and Microsystem Technologies, 2026, 45(3): 172-178.
[13] 张丰麟, 张松, 白克强, 等. 基于改进蝴蝶算法的排气压力自抗扰控制参数优化[J]. 机床与液压, 2025, 53(18): 104-111.
ZHANG F L, ZHANG S, BAI K Q, et al.Optimization Parameters of Exhaust Pressure Active Disturbance Rejection Control Based on Improved Butterfly Algorithm[J]. Machine Tool & Hydraulics, 2025, 53(18): 104-111.
[14] 董金瑞, 邹光明, 刘源泂, 等. 基于改进蝴蝶算法的机械臂轨迹规划[J]. 组合机床与自动化加工技术, 2025(9): 67-73.
DONG J R, ZOU G M, LIU Y J, et al.Trajectory Planning of Robotic Arm Based on Improved Butterfly Algorithm[J]. Modular Machine Tool & Automatic Manufacturing Technique, 2025(9): 67-73.
[15] 杨敖, 李宏峰, 柳晓菁, 等. 基于IPSO模糊PID的切纸机后挡规电液伺服系统跟踪控制设计[J]. 包装工程, 2025, 46(23): 243-252.
YANG A, LI H F, LIU X J, et al.Tracking Control Design of Electro-Hydraulic Servo System for Paper Cutter Backgauge Based on IPSO-Fuzzy PID[J]. Packaging Engineering, 2025, 46(23): 243-252.
基金
河南省科技攻关项目（262102320321）; 河南省科技攻关项目（252102211048）
PDF(918 KB)
渝公网安备 50010702501716号 渝ICP备15012534号-2
