摘要
为了对液化石油气(LPG)公路运输罐车储罐系统故障进行准确、全面的诊断,通过利用故障模式影响分析方法(FMEA)构建储罐系统故障模式及故障特征指标,根据日常检测数据构造训练样本,运用径向基函数(RBF)神经网络对网络进行训练建立诊断模型并利用诊断模型对罐车故障进行诊断。经验证:诊断结果与实际情况相符合。因此,基于FMEA与RBF神经网络所构建的模型可以用于危险化学品汽车罐车储罐系统的故障诊断。
In order to diagnose the fault of the LPG tank for road transportation accurately and roundly,the failure mode and failure index system were founded based on the method of FMEA.Then training samples were constructed according to the daily test data by which the RBF neural network was trained and the diagnosis model was built.At last,to verify the correctness,the model was applied to diagnose the tank＇s fault.The results show that the diagnosis fault of the model is consistent with the actual fault of tank.So the model based on FMEA and RBF neural network is applicable to the fault diagnosis of hazardous chemical tank.
出处
《中国安全科学学报》
CAS
CSCD
北大核心
2011年第1期99-104,共6页
China Safety Science Journal
基金
广西壮族自治区教育厅科研基金资助(200808MS021)
