基于概率神经网络的风机故障诊断  

Fault Diagnosis of Fan Based on Probabilistic Neural Network

作　　者：李铁军[1] 朱成实[1] 吕营[1] 王丹[1] 王学平[1] 

机构地区：[1]沈阳化工学院机械工程学院,沈阳110142

出　　处：《煤矿机械》2007年第10期187-189,共3页Coal Mine Machinery

摘　　要：针对风机常见故障征兆与故障类型之间的非线性映射关系,结合专家知识建立了风机系统故障知识库,提出了基于PNN神经网络的风机故障诊断方法,结果表明该方法能克服BP算法诊断过程中容易陷入局部极小的缺点,并能满足故障诊断的快速性和准确性要求,适用于在线检测,具有实际应用价值。Based on nonlinear mapping relationship between fault symptom and fan faults, probabilistic neural network （PNN） approach was presented for fault diagnosis. Then fault features were extracted from fan failures and the extracted features were regarded as fault symptom eigenvector. Fault diagnosis model and fault diagnosis algorithm were given using probabilistic neural network. The result shows that probabilistic neural network can overcome the limitation of local imrinitesimal of BP, and can meet the requirement for fast diagnosis rate and high diagnosis precision during fault diagnosis process, so probabilistic neural network can be used in the real time diagnosis. And it shows that the fault diagnosis based on probabilistic neural network is useful.

关 键 词：PNN神经网络 故障诊断 风机 

分 类 号：TP11[自动化与计算机技术—控制理论与控制工程] TH43[自动化与计算机技术—控制科学与工程]