摘要
为缩短煤矿井下电能故障检测时间、降低漏报率,设计并实现了一种基于改进循环神经网络的煤矿电能故障检测方法。综合使用门控循环单元(GRU)和多层感知器等技术搭建神经网络,学习和提取具有时间先后顺序的采样数据中潜在的相互依赖关系。通过在实验室中进行测试以及在实际生产环境中进行部署后发现,该方法行之有效,极大地缩短了煤矿井下电能故障的检测时间,降低了漏报率。
To shorten the detection time of underground electrical energy fault in coal mine and reduce the false alarm rate,designed and implemented a coal mine electrical energy fault detection method based on an improved recurrent neural network.A neural network was built by using techniques such as gated recurrent units(GRU)and multi-layer perceptron to learn and extract potential interdependencies in sampled data with temporal order.Testing in the laboratory and deployment in actual production environments show that this method is effective,which greatly shortens the detection time of coal mine underground electrical energy fault and reduces the false alarm rate.
作者
张振宇
Zhang Zhenyu(China Coal Research Institute,Beijing 100013,China;Engineering Research Center for Technology Equipment of Emergency Refuge in Coal Mine,Beijing 100013,China;Beijing Mine Safety Engineering Technology Research Center,Beijing 100013,China)
出处
《煤矿机械》
2024年第5期155-157,共3页
Coal Mine Machinery
基金
国家重点研发计划(2021YFB3201905)
煤炭科学技术研究院有限公司新产品新工艺开发项目(2023CG-ZB-11)。
关键词
GRU
循环神经网络
电能故障
故障检测
GRU
recurrent neural network
electrical energy fault
fault detection
