针对滚动轴承故障诊断中存在的故障类型多样且有效特征难以选取等问题,提出了一种基于商空间和支持向量机的智能诊断模型。该模型利用商空间的粒化分层思想将输入样本按照不同等价关系进行粒化分层,并对每一粒度层的时域和频域特征进行约简,然后将每一层约简后的特征向量输入到支持向量机进行状态识别,最后对各粒度层状态识别结果加权融合得到最终结果。利用轴承全寿命试验数据对该模型进行验证,识别精度达到96.92%。
Abstract
Aiming at the problems in rolling bearing fault diagnosis that the types of faults are various, and effective features are difficult to select, an intelligent diagnosis model is proposed based on quotient space and support vector machine (SVM). Firstly, based on the stratification idea of quotient space, the model granulates input samples into different granular layers according to different equivalence relations, then time domain and frequency domain features are reduced to obtain the sensitive feature set of each granular layer. Secondly, the sensitive feature set of each layer is inputted into SVM for fault identification. Finally, the final diagnosis result is obtained by weighted fusion of the fault recognition results of each granularity layer. The model is applied to process the bearing run-to-failure test data, and the recognition accuracy reaches 96.92%, indicating the validity and practicability of the model.
关键词
计量学 /
智能诊断 /
商空间 /
滚动轴承 /
支持向量机
Key words
metrology /
intelligent diagnosis /
quotient space /
rolling bearing /
support vector machine
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
［1］孟宗, 邢婷婷, 张圆圆, 等. 基于关联维数和线段聚类的滚动轴承故障诊断［J］. 计量学报, 2019,40(1): 100-105.
Meng Z, Xing T T, Zhang Y Y, et al. Fault diagnosis of rolling bearing based on correlation dimension and segment lustering［J］. Acta Metrologica Sinica, 2019,40(1): 100-105.
［2］时培明, 梁凯, 赵娜, 等. 基于分形维数和GA-SVM的风电机组齿轮箱轴承故障诊断［J］. 计量学报, 2018,39(1): 61-65.
Shi P M, Liang K, Zhao N, et al. Fault diagnosis of wind turbine gearbox bearing based on fractal dimension and GA-SVM［J］. Acta Metrology Sinica, 2018,39(1): 61-65.
［3］吐松江·卡日, 高文胜, 张紫薇, 等. 基于支持向量机和遗传算法的变压器故障诊断［J］. 清华大学学报(自然科学版), 2018, 58(7): 623-629.
Kari T S J, Gao W S, Zhang Z W, et al. Power transformer fault diagnosis based on a support vector machine and a genetic algorithm［J］. Journal of Tsinghua University (Science and Technology), 2018, 58(7): 623-629.
［4］孟宗, 刘东, 岳建辉, 等. 基于DEMD局部时频熵和SVM的风电齿轮箱故障诊断方法研究［J］. 计量学报, 2017, 38(4): 449-452.
Meng Z, Liu D, Yue J H, et al. Wind power gear box fault diagnosis based on DEMD local frequency entropy and SVM［J］. Acta Metrology Sinica, 2017: 38(4): 449-452.
［5］Yao Y Y. A partition model of granular computing［J］. Transactions on Rough Sets I, 2004, 3100: 232-253.
［6］张玲, 张钹. 问题求解的理论及应用:第2版［M］. 北京: 清华大学出版社, 2007.
［7］苗夺谦, 王国胤, 刘清, 等. 粒计算: 过去、现在与展望［M］. 北京: 科学出版社, 2007.
［8］郭庆文, 王国胤, 张清华. 多粒度的图像检索方法研究［J］. 计算机科学, 2013, 40(1): 298-301.
Guo Q W, Wang G Y, Zhang Q H. Research on multi-granularity image retrieval method［J］. Computer Science, 2013, 40(1): 298-301.
［9］张清华, 郭永龙, 薛玉斌. 概率统计下的多粒度搜索算法研究［J］. 模式识别与人工智能, 2015, 28(5): 422-428.
Zhang Q H, Guo Y L, Xue Y B. Multi-granularity search algorithm based on probability statistics［J］. Pattern Recognition and Artificial Intelligence, 2015, 28(5): 422-428.
［10］梁远银. 基于商空间的粒计算及其在数据挖掘中的应用［D］. 成都: 四川师范大学, 2013.
［11］张玲, 张钹. 模糊商空间理论(模糊粒度计算方法)［J］. 软件学报, 2003, 14(4): 770-776.
Zhang L, Zhang B. Theory of fuzzy quotient space (Methods of Fuzzy Granular Computing)［J］. Journal of Software, 2003, 14(4): 770-776.
［12］杨纶标, 高英仪. 模糊数学原理及应用［M］. 广州: 华南理工大学出版社, 2008: 66-77.
［13］张清华, 王国胤, 刘显全. 分层递阶的模糊商空间结构分析［J］. 模式识别与人工智能, 2008, 21(5): 627-634.
Zhang Q H, Wang G Y, Liu X Q. Hierarchical structure analysis of fuzzy quotient space［J］. Pattern Recognition and Artificial Intelligence, 2008, 21(5): 627-634.
［14］王国胤. Rough集理论与知识获取［M］. 西安: 西安交通大学出版社, 2001.
［15］李超. 基于距离评估和粗糙集理论的流型特征选择方法［D］. 吉林: 东北电力大学, 2010.
［16］Chen S Y, Wang W, Zuylen H V. Construct support vector machine ensemble to detect traffic incident［J］. Expert Systems with Applications, 2009, 36: 10976-10986.
基金
国家自然科学基金(51505415); 河北省自然科学基金(E2017203142); 秦皇岛市科技支撑计划(201602A025)
PDF(592 KB)
