基于概率神经网络的风洞设备故障预测诊断方法

Fault Forecast and Diagnosis of Wind Tunnel Equipment Based on Probabilistic Neural Network

针对风洞设备故障征兆与故障原因之间的非线性关系,提出基于概率神经网络的风洞设备故障预测诊断方法。利用概率神经网络强大的自主学习能力和较强的模式识别能力,来预测诊断风洞设备的故障原因,通过故障样本对概率神经网络进行训练,并对待测样本进行故障预测诊断。结果表明:概率神经网络能满足故障诊断快速和准确的要求,故障预测诊断精度较高,适用于在线检测,具有实际应用价值。

According to the nonlinear mapping relationship between fault symptom and wind tunnel equipment faults, fault forecast and diagnosis method was presented which is based on probabilistic neural network（PNN）. By using the powerful self-learning ability and strong pattern recognition capability, the cause of the fault for the wind tunnel equipment is predicted. The sample of the fault is established and the PNN is trained based on the symptom diagnosis. The test sample is used to fault forecast and diagnosis. The result shows that PNN can meet the requirement for fast diagnosis rate and high diagnosis precision during fault diagnosis process, so PNN can be used in the real time diagnosis with application value.