【摘要】 自动机射击动作时因激振力的作用会产生一定方向和频率的冲击振动,构件的裂纹或松动等故障会影响到其响应成分的频率能量特性;针对自动机实射动作冲击响应振动信号,利用小波分析快速进行信噪分离,频域范围内采用功率谱分析结合小波包分解对各频段能量谱分析。根据振动信号时域峰值和时刻,频域能量的变化和分布,给出故障诊断层使用的状态特征向量,并用比例梯度动量共轭算法训练的神经网络模型进行自动机状态定位与故障识别。更多还原

【Abstract】 Automatic machine will have some impact on the direction and frequency of vibration due to the role of the exciting force in the shooting action,Cracks or loose of the components will affect the frequency and energy of its response to the action;Wavelet analysis can be used in the live-fire vibration signals to get the signal noise separation quickly,Power spectrum analysis with wavelet packet decomposed the energy spectrum of the frequency band.The state of fault diagnosis using the feature vector was given according to the time domain vibration signal peak and moment,frequency and distribution of energy change,Momentum conjugate gradient algorithm for the proportion of trained neural network model was put forward for automatic positioning machine status and fault identification.更多还原