一种大储量罐车紧急切断阀失灵检测方法研究  

Research on Large Reserves of Tank Car Emergency Cut-off Valve Failure Detection Method

作　　者：杨怀磊 黄玉萍[1] 

机构地区：[1]郑州旅游职业学院信息工程系,郑州450009

出　　处：《科技通报》2014年第12期85-87,共3页Bulletin of Science and Technology

摘　　要：为了提高大储量液化石油气罐车紧急切断阀失灵检测的精度和效率,提出了基于增强模糊支持向量机的罐车紧急切断阀失灵检测方法。首先通过对大储量液化石油气罐车紧急切断阀的常见失灵类型进行分析,将其样本数据进行训练集和测试集的训练。其次利用模糊支持向量机算法进行罐车紧急切断阀失灵检测模型建立。然后利用粒子群进化算法进行该模型的最优参数、惩罚系数和隶属度的确定,最后利用优化的参数进行支持向量机的分类。通过训练集和测试集对该模型进行仿真实验,仿真实验结果表明,该方法的失灵检测准确率均在95%以上,具有较高的鲁棒性和可靠性。In order to improve the large reserves of liquefied petroleum gas tank car emergency cut-off valve failure detec-tion accuracy and efficiency of fuzzy support vector machine （SVM） was proposed based on enhancement of tank car emer-gency cut-off valve failure detection method. First through to the large reserves of liquefied petroleum gas tank car commonfailure types of emergency cut-off valve were analyzed, and the sample training set and testing set of training data; Second-ly fuzzy support vector machine （SVM） algorithm for tank car emergency cut-off valve failure detection model; Then useparticle swarm evolution algorithm for the optimal parameters of the model, the determination of penalty coefficient andmembership degree; Finally using the optimized parameters of support vector machine （SVM） classification. Through thetraining set and testing set the model simulation, the simulation experimental results show that the failure detection accura-cy of the method are in more than 95%, has high robustness and reliability.

关 键 词：支持向量机 粒子群算法 紧急切断阀 失灵检测 

分 类 号：TP397[自动化与计算机技术—计算机应用技术]