基于LS―SVM制冷系统故障诊断.docVIP

基于LS―SVM制冷系统故障诊断 　　摘要：为了提高制冷系统故障诊断速度及准确性，提出了基于最小二乘支持向量机（LSSVM）的制冷系统故障诊断模型，并采用ASHRAE制冷系统故障模拟实验数据进行模型训练与验证.对一台 90冷吨（约316 kW）的离心式冷水机组的7类制冷循环典型故障进行了实验.研究结果表明，LS-SVM模型对制冷系统七类故障的总体诊断正确率比 支持向量机（SVM）诊断模型、误差反向传播（BP）神经网络诊断模型分别提高0.12%和1.32%；尽管对个别局部故障（冷凝器结垢、冷凝器水流量不足、制冷剂含不凝性气体）的诊断性能较SVM模型的略有下降，但对系统故障的诊断性能均有较大改善，特别是对制冷剂泄漏/不足故障；诊断耗时比SVM模型减少近一半，快速性亦有所改善.可?，LSSVM模型在制冷系统故障诊断中具有良好的应用前景. 关键词：制冷系统； 故障诊断； 最小二乘支持向量机； 误差反向传播； 支持向量机 中图分类号： TB 65文献标志码： A Abstract：In order to improve the fault diagnosis speed and accuracy for refrigeration system，a fault diagnosis model based on least squares support vector machine（LSSVM） was proposed.American Society of Heating，Refrigerating，and Airconditioning Engineering（ASHRAE） refrigeration system fault simulation data was used for the model training and validation.The experiments of a centrifugal chiller of 90 tons with seven types of typical faults were conducted.The results showed that the overall diagnostic accuracy of LSSVM model for seven types of faults increased by 0.12% and 1.32% respectively，compared with support vector machine（SVM） diagnosis model and error backpropagation（BP） neural network model.Although diagnostic performance of LSSVM model for individual componentlevel fault（ConFoul/ReduCF/NonCon） was low slightly compared with SVM model，the diagnosis performance for systemlevel were greatly improved，especially for refrigerant leakage or lack of refrigerant.The diagnosis time of LSSVM model reduced nearly half than that of SVM model.At the same time，its rapidity improved.Therefore，LSSVM diagnostic model had good application in the fault diagnosis of refrigeration system. Keywords：refrigeration system； fault diagnosis； least squares support vector machine； error backpropagation； support vector machine 暖通空调系统（HVAC）“带障”运行将导致系统工作效率降低，无法达到预期制冷效果，维修成本、能耗及温室气体排放量都将增加.文献[1]显示，制冷系统因故障而导致的能耗增加达15%～30%，有效的故障检测与诊断有助于及时恢复HVAC系统正常运行，缩短设备带障运行的时间，减少制冷系统设备磨损乃至温室气体的释放等，从而节约能源，保护环境，并减少不必要的经济损失. 目前，在故障诊断方面应用最多的是神经网络方法，如误差反向传播（BP）神经网络、径向基函