论文部分内容阅读
振动信号包含信息丰富,反应状态直接,振动法是压缩机气阀故障诊断常用方法。振动信号的特征参数繁多,特征向量选择是否合理对故障诊断结果准确性影响很大。研究ReCorre方法对气阀振动信号特征参数进行优化选择,再通过神经网络进行分类识别。实例表明,基于特征优化的模糊神经网络分类识别结果正确率高,识别结果受数据来源影响小,是一种较好的气阀故障诊断方法。
Vibration signal contains rich information, the reaction state is direct, vibrating method is compressor valve common fault diagnosis method. The characteristic parameters of vibration signals are numerous, and the selection of eigenvectors is reasonable and has a great influence on the accuracy of fault diagnosis. ReCorre method is used to optimize the characteristic parameters of the valve vibration signals and then classified and identified by neural network. The example shows that the fuzzy neural network based on feature optimization has a high correctness of classification and recognition, and the recognition result is less affected by the data source. So it is a good method to diagnose the valve fault.