基于挤压-激励注意力机制的CNN-LSTM煤矿用电机异常状态检测

针对环境恶劣和复杂工况导致煤矿用电机故障频发,为保障煤矿用电机的可靠运行,提出一种基于挤压-激励注意力机制的CNN-LSTM煤矿用电机异常状态检测模型。首先引入卷积神经网络(CNN)获取多特征输入空间联系,采用长短期记忆网络(LSTM)提取序列时序变化特征,结合挤压-激励注意力机制(SE)为LSTM层进行自适应权重分配来增强电机定子电流的关键信息提取;然后通过均方根误差(RMSE)对电机定子电流进行残差分析,检测电机定子电流的异常变化;最后,以新疆某大型露天煤矿121带式输送机1#煤矿用电机实时运行状态数据验证所提方法的实用性,结果表明该方法能够精准检测到电机定子电流异常状态,为煤矿用电机可靠运行提供重要依据。

Design and analysis of an electric-locomotive turning system for coal mines

We describe a system for turning the accumulator electric locomotives used in coal mines. This improves transport efficiency and decreases mine construction and production costs. The operating requirements, principles, working features and transmission system are discussed. This system has a simple structure and can be manufactured, transported and assembled easily. It, therefore, can handle accumulator locomotives of middle and small size.