基于隐马尔科夫模型的牵引座状态识别

牵引座是机车中连接车体和转向架的重要部件,承受并传递着机车的纵向力,因此牵引座的状态影响着机车的安全。针对牵引座正常、小裂纹故障及大裂纹故障这3种状态,提出了一种基于隐马尔科夫模型的状态识别方法:首先对机车牵引座运行的加速度信号进行特征提取,选取敏感的特征量组成隐马尔科夫模型的训练集与测试集;然后进行状态识别;最后以最大的似然概率对应的状态作为识别结果。结果表明,针对牵引座的3种不同状态,该方法的识别结果比K邻近算法有更高的识别率。

State recognition of traction seat based on hidden Markov model

The traction seat is an important part of the locomotive connecting the car body and the bogie.It bears and transmits the longitudinal force of the locomotive.Therefore,the state of the traction seat affects the safety of the whole locomotive.Aiming at three states of normal traction,small crack fault and large crack fault,a state recognition method based on hidden Markov model was proposed.Firstly,the acceleration signal of locomotive traction seat was extracted by feature extraction,and the training set and test set of hidden Markov model were composed of sensitive feature variables for state recognition.Finally,the state corresponding to the maximum likelihood probability was used as the recognition result.The results show that the recognition rate of this method is higher than that of K-NearestNeighbor algorithm for three different states of traction seat.