基于机器人力矩特征信息识别的螺纹传动合格性检验  

Thread Transmission Qualification Inspection Based on Robot Torque Characteristic Information Recognition

作　　者：刘嘉璐 张博 凌宇 叶建伟 张铁[2] 杨明达 LIU Jialu;ZHANG Bo;LING Yu;YE Jianwei;ZHANG Tie;YANG Mingda(Guangzhou Power Supply Bureau,Guangdong Power Grid Co.,Ltd.,Guangzhou Guangdong 510620,China;School of Mechanical and Automotive Engineering,South China University of Technology,Guangzhou Guangdong 510640,China)

机构地区：[1]广东电网有限责任公司广州供电局,广东广州510620 [2]华南理工大学机械与汽车工程学院,广东广州510640

出　　处：《机床与液压》2025年第5期39-46,共8页Machine Tool & Hydraulics

基　　金：南方电网公司科技项目(030167KK52220001)。

摘　　要：面向螺纹传动合格性检验,提出一种基于时间序列模型的检验算法。基于合格性检验机器人系统输出力矩特性,针对开关柜中断路器小车旋进和旋出操作,通过分析断路器小车由实验位置运动到工作位置过程中螺纹传动力矩序列的内在变化规律,建立系统输出力矩的时间序列分析模型,并生成动态阈值。利用最小二乘法和最小信息准则确定该模型阶数,并通过自回归逼近方式对其进行参数辨识。最后,为了验证所提时间序列分析算法的有效性,构建螺纹传动质量检测平台,在卡阻和碰撞状态下,分别对时间序列分析算法、小波理论算法和对称阈值算法进行传动力矩突变检测和斜变检测。结果表明:采用时间序列分析算法对传动力矩进行无突变和斜变经检测时未出现误报情况,识别准确率为100%;与固定阈值算法和小波理论算法相比,该方法检测时延分别缩短了91%和75%。Towards the qualification inspection of threaded fasteners,a testing algorithm based on time series models was proposed.Based on the output torque characteristics of the qualification inspection robotic system,focusing on the screw-in and screw-out operations of the circuit breaker trolley in the switchgear,a time series analysis model for the system′s output torque was established by analyzing the inherent variation patterns of the threaded transmission torque sequence during the movement of the circuit breaker trolley from the test position to the working position,and the dynamic threshold was generated.The order of the model was determined by the least square method and the minimum information criterion,and its parameters were identified by the method of autoregressive approximation.Finally,to validate the effectiveness of the proposed time series analysis algorithm,a threaded transmission quality inspection platform was constructed.Under conditions of jamming and collision,the algorithm was compared with wavelet theory-based and symmetric threshold algorithms for detecting torque mutations and gradients.The results demonstrate that the time series analysis algorithm achieves a 100%accuracy rate in identifying torque without false alarms in both mutation and gradient detection scenarios.Furthermore,compared with fixed threshold and wavelet theory algorithms,the detection latency of the proposed method is reduced by 91%and 75%,respectively.

关 键 词：机器人 螺纹传动合格性检验 时间序列模型 动态阈值 

分 类 号：TP242.2[自动化与计算机技术—检测技术与自动化装置] TH17[自动化与计算机技术—控制科学与工程]