摘要
针对岩巷掘进机工作时负载多变,动载荷实时识别难度大等问题,提出了一种基于小波包特征能量的神经网络动载荷识别新方法。实时采集截割机构的振动信号、截割电动机的电流及液压缸压力信号,应用小波包分解得到相应信号的特征量,并将其作为神经网络的输入样本,训练神经网络并对网络进行测试。结果表明,动载荷实时识别准确率可达0.93,该识别方法能够满足动载荷实时识别系统的要求。
In order to solve the problems in rock road-header such as changing loads, difficult dynamic load real-time identification, a recognition method based on wavelet packet and neural network was proposed. The vibration signals, the current and hydraulic cylinder pressure signals were collected in real-time. The feature vectors of the corresponding signals, which were chosen as input values for the neural network, were gained through wavelet packets decomposition. It has been shown by experiments that the accuracy rate of dynamic load real-time identification is up to 0.93 and the identification method can meet the requirement of dynamic load real-time identification system.
出处
《煤矿机械》
2015年第3期238-241,共4页
Coal Mine Machinery
基金
国家863计划资源环境技术领域重大项目(2012AA06A405)
高等学校博士学科点博导类专项科研基金(20111402110010)
关键词
岩巷掘进机
小波包
神经网络
动载荷识别
rock roadheader
wavelet packet
neural network
dynamic loading identification
