基于时序分析与模糊聚类的变速箱齿轮故障识别

Fault Recognition of Vehicle Transmission Gear Based on Time Series Analysis and Fuzzy Cluster

通过对 L C5 T81变速箱疲劳寿命台架试验采集的齿轮运行状态振动信号进行时间序列分析和特征向量提取 ,并采用模糊聚类分析方法确定变速箱齿轮运行状态特征向量样本的亲疏关系 ,实现了对变速箱齿轮的跑合运行状态、磨损运行状态和故障运行状态的识别与诊断。验证表明 ,基于时间序列分析与模糊聚类分析相结合的故障识别方法能够有效地识别出变速箱齿轮运行状态。

Based on time series analysis and fuzzy cluster analysis, a new method of fault recognition of vehicle transmission gear was set up. By the time series analysis the feature vectors extraction method for the transmission gear working state vibration signal was studied. Based on the fuzzy cluster analysis the similarity relation between the feature vector of the working state transmission gears and the sample feature vector was obtained. Working state of transmission gears was determined according to this similarity relation of the feature vector. This method was used to recognize normal working state, wearing working state and fault working state of LC5T81 transmission gear. The result shows that this method of fault recognition of vehicle transmission gear is effective.