论文部分内容阅读
主减速器(简称“主减”)是直升机传动系统的关键部件,它常处于高转速高负荷的恶劣环境下,对其运行状态进行预测,于直升机的安全性来说至关重要。鉴于此,提出了一种离散小波变换(DWT)、Kalman滤波以及Elman神经网络相结合的直升机主减智能状态预测系统:DWT使用“db44”母小波对振动信号进行分解提取特征向量,Kalman滤波对未来各时刻的特征向量进行预测,Elman神经网络对预测值进行故障辨识和分类。在Kalman滤波算法中,提出了一种新的预测算法,并用实验对该算法组成的系统进行验证,结果表明:该Kalman滤波算法预测效果好,更适用于对主减的特征向量进行预测;离散小波变换(DWT)、Kalman滤波以及Elman神经网络相结合组成的智能状态预测系统是可行的,它能很好地对主减的未来状态进行预测。
The main reducer (“main subtraction”) is a key component of a helicopter drive system. It is often under harsh conditions of high speed and high load. Predicting its operating status is crucial for the helicopter’s safety. In view of this, a helicopter main reduction intelligent state prediction system combining DWT, Kalman filter and Elman neural network is proposed: DWT decomposes the vibration signal using “db44 ” mother wavelet to extract eigenvector, Kalman The filter predicts the future eigenvectors of each moment, and the Elman neural network performs fault identification and classification for the predicted values. In the Kalman filter algorithm, a new prediction algorithm is proposed, and the system composed of the algorithm is verified by experiments. The results show that the Kalman filter algorithm has good prediction effect and is more suitable for predicting the main reduced feature vector. Discrete The intelligent state prediction system composed of wavelet transform (DWT), Kalman filter and Elman neural network is feasible and it can predict the future state of the main subtraction.