基于SMO算法的中央空调螺杆式压缩机故障自动检测技术  

Automatic Fault Detection Technology for Screw Compressor of Central Air Conditioner Based on SMO Algorithm

作　　者：李振坡[1] 李永新[1] LI Zhen-po;LI Yong-xin(Capital Normal University,Beijing 100048 China)

机构地区：[1]首都师范大学,北京100048

出　　处：《自动化技术与应用》2022年第6期21-25,共5页Techniques of Automation and Applications

摘　　要：中央空调螺杆式压缩机的工作马达和压缩机各个零件被密封在一个机箱中,操作不当很有可能使压缩机电机出现过载的情况而发生故障。因此,提出了基于SMO算法的中央空调螺杆式压缩机故障自动检测技术。故障检测装置由PCI总线安装在工控计算机内,并集合Labview软件和PCI1710驱动程序,实现对故障的特征提取,根据不同的故障特征,设置其相对应的特征向量,组成训练样本和测试样本。通过一对一分类算法建立多元分类器,测试样本,根据测试结果判断压缩机出现故障的部位,实现对压缩机故障自动检测。实验以首都师范大学螺杆机机组维修数据为样本进行实验分析,结果表明所提方法具有较高诊断效率且性能良好,适合压缩机的故障识别。The working motor and various parts of the compressor of screw compressor of the central air conditioner are sealed in a box. Improper operation may cause the overload of the compressor motor and cause the failure. Therefore, the automatic fault detection technology of screw compressor in central air conditioning based on SMO algorithm is proposed. The fault detection device is installed in the industrial control computer by PCI bus, and integrates LabVIEW software and pci1710 driver to realize fault feature extraction. According to different fault features, the corresponding feature vectors are set to form training samples and test samples. Through the one-to-one classification algorithm to establish a multi classifier, test samples, according to the test results to determine the fault location of the compressor, realize the automatic detection of compressor fault. The experimental results show that the proposed method has high diagnosis efficiency and good performance, and is suitable for compressor fault identification.

关 键 词：SMO算法 特征提取 压缩机故障检测 一对一方法 支持向量机 

分 类 号：TP206.3[自动化与计算机技术—检测技术与自动化装置]