LabVIEW与MATLAB联合编程在制冷系统故障诊断中的应用

制冷系统作为建筑的主要能耗设备，其故障的及时诊断与纠正有助于提高能源利用率，减少对环境的影响。缺乏诊断平台是制约制冷系统故障智能诊断推广应用的重要因素之一。基于MATLAB软件强大的数据处理能力和LabVIEW软件便捷的虚拟仪器开发功能，进行联合编程，实现优势互补，并将粒子群优化（PSO）算法引入最小二乘支持向量机（LSSVM），作为故障诊断的内核，建立了制冷系统故障诊断平台。对离心式冷水机组七类典型故障的诊断实验结果表明：该联合编程方法可行，可以实现界面友好、功能丰富、方便快捷的制冷系统故障诊断；所建立的PSO-LSSVM模型总体诊断正确率高达99.70%，正常和每类故障的诊断正确率均在99%以上。该平台具有一定的开放性和扩展性，后续可以根据需要对诊断算法及其他功能模块进行更改或添加，也可增加不同算法之间的比较分析或集成诊断，改善其灵活性，推进其推广应用。LabVIEW软件的采用，也为故障检测与诊断嵌入系统控制提供了可能。

Application of LabVIEW and MATLAB hybrid programming to fault diagnosis of refrigeration system

Refrigeration system is the major energy consumption equipment in the building. Timely diagnosis and correction of faults is helpful to improve energy efficiency and reduce the environmental impact. The lack of fault diagnosis platform is one of the major factors that limited the popularization and application of intelligent fault diagnosis of refrigeration system. Hybrid programming was adopted by combined MATLAB and LabVIEW with powerful data processing capability and convenient virtual instrument development capability, respectively. Besides, the introduction of particle swarm optimization (PSO) algorithm into the least squares support vector machine (LSSVM) was used as a fault diagnosis kernel to establish a refrigeration system fault diagnosis platform. Diagnostic experiments on seven types of typical faults in the centrifugal chiller showed that the hybrid programming method was feasible and could realize the fault diagnosis of refrigeration system with a friendly interface and abundant functions in a convenient manner. The diagnosis accuracy of PSO-LSSVM model was as high as 99.70%, and the accuracy of normal and each fault exceeded 99%. The platform is open and extensible for the modification and integration of diagnostic algorithms and other functional modules as required. Comparative analysis or integrated diagnosis can be included as well. The adoption of LabVIEW also provides the possibility of embedding fault diagnosis into system control.