摘要
对工业机械设备的状态进行及时监测,可以极大程度地降低企业因工业设备故障带来的损失。以传统的设备状态监测方法为基础,结合机器学习中的SVM支持向量机算法与Clara聚类算法,首次提出了基于SVM-Clara模型的机械设备状态监测方法。通过仿真实验,首先得到了SVM-Clara模型的最高效率为96.9%,最佳训练数据量为7000,此时模型最高的聚类效率为32.07%;接着再与传统的三种机械设备状态监测方法进行横向对比,得出SVM-Clara模型的理论准确率为95.8333%,证明了基于SVM-Clara模型的机械设备状态监测方法的准确性与高效性。
Timely monitoring of the state of industrial machinery and equipment can greatly reduce the loss caused by industrial equipment failure.In this paper,based on the traditional equipment condition monitoring method,combining the SVM support vector machine algorithm and Clara clustering algorithm in machine learning,a mechanical equipment condition monitoring method based on SVM-CLARA model is proposed for the first time.Through the simulation experiment, the maximum efficiency of SVM-CLARA model is 96.9%,and the optimal training data amount is 7000.At this time,the maximum clustering efficiency of the model is 32.07%.Then compared with the three traditional mechanical equipment state monitoring methods,the theoretical accuracy of SVM-CLARA model is 95.8333%,which proves the accuracy and efficiency of the mechanical equipment state monitoring method based on SVM-CLARA model.
出处
《工业控制计算机》
2022年第1期47-49,51,共4页
Industrial Control Computer