机械传动系统含偏差频率分量协同检测与分解方法  

Cooperative Detection and Decomposition Approach of Deviation-included Frequency Components for Mechanical Transmission Systems

作　　者：于小洛 杨阳[2] 杜明刚[2] 何清波 彭志科[1,3] YU Xiaoluo;YANG Yang;DU Minggang;HE Qingbo;PENG Zhike(State Key Laboratory of Mechanical System and Vibration,Shanghai Jiao Tong University,Shanghai 200240;China North Vehicle Research Institute,Beijing 100072;School of Mechanical Engineering,Ningxia University,Yinchuan 750021)

机构地区：[1]上海交通大学机械系统与振动国家重点实验室,上海200240 [2]中国北方车辆研究所,北京100072 [3]宁夏大学机械工程学院,银川750021

出　　处：《机械工程学报》2024年第15期100-112,共13页Journal of Mechanical Engineering

基　　金：基础研究资助项目(MKF20210013)。

摘　　要：传动系统故障诊断对于保障装备可靠运行具有重要作用,信号分解是传动系统故障特征提取的重要技术手段。由于传动误差、零部件惯性、频率分辨率等,传动系统振动信号的部分分量实测频率往往与根据运动学关系计算的理论频率存在偏差,且分量的存在性判定问题通常被忽略。上述局限导致主流的信号分解方法在处理含偏差频率分量时,面临分解精度及物理可解释性方面的共同挑战。对此,提出一种针对传动系统含偏差频率分量的协同检测与分解方法,借助信号分量间固有的谐波关系对含偏差频率分量进行协同检测,并判定其存在性;协同检测的结果能够进一步作为信号分解的基础,分解精度相比未经协同检测的现有方法大幅提升。通过三个实际应用案例验证了所提方法对平稳工况及非平稳工况下的传动系统信号特征提取的有效性。Fault diagnosis of transmission systems plays an essential role in ensuring the reliable operation of equipment,and signal decomposition is an important technique in fault feature extraction.Due to the existence of transmission error,component inertia,and frequency resolution,the tested frequency of the vibration signal component often deviates from the theoretical one calculated according to the kinematic relationship,and the existence determination of the components is usually ignored.The above limitations result in common challenges for the mainstream signal decomposition approaches in terms of decomposition accuracy and physical interpretability for decomposing deviation-included frequency components.Therefore,a deviation-included frequency component cooperative detection and decomposition approach is proposed to detect the deviation-included frequency components and determine their existence.The cooperative detection results can be further used as the basis of signal decomposition,and the decomposition accuracy is greatly improved compared with the existing methods without cooperative detection.Finally,three practical application cases demonstrate the effectiveness of the proposed method in transmission system signal characteristics extraction under stationary and non-stationary conditions.

关 键 词：协同检测 故障诊断 传动系统 信号分解 

分 类 号：TG156[金属学及工艺—热处理]