[Objective] To meet the development needs of railway electricity work monitoring services and address issues such as high coupling degree, lengthy deployment time, and high construction and operational costs in conventional signal monitoring system architectures, an architecture design scheme for a SCMS (signal centralized monitoring system) based on cloud computing technology is proposed. [Method] The architecture and network framework of conventional SCMS are introduced. The current application status and existing issues of the SCMS are analyzed. Three service forms and deployment methods provided by cloud computing technology are introduced. Combining the requirements of railway safety production and the nature of SCMS, the proposed SCMS is deployed and tested, and its evaluation indicators are analyzed. [Result & Conclusion] The proposed SCMS architecture is deployed using a private cloud, ensuring the security and controllability of data and assets. The system is designed using distributed architecture and cluster mode, enabling the use of automated scripts to read configuration information, distribute files, and operate processes, thus achieving the automated deployment of SCMS. Test results demonstrate that the architecture design scheme of SCMS based on cloud computing technology can optimize resource allocation and flexibly deploy the SCMS, exhibiting stable and reliable performance.

In view of the current switch equipment monitoring methods is single, and the existing monitoring and detection methods cannot meet the requirements of intelligent and efficient operation and maintenance of equipment in complex environments, a switch working condition monitoring system based on magnetic grid ruler is designed. It monitors the whole process of the key mechanical parts of the turnout, in addition, the big data, machine learning and other technologies are used to diagnose and warn the common diseases of the turnout junction,so as to provide data support for the integrated analysis and regulation of the combination of track and communication ＆ signaling.

重载铁路电务设备智能运维技术,是保障重载列车安全高效运行的重要手段.依托朔黄铁路,开展重载铁路电务设备智能运维技术研究.运用新型传感器、北斗卫星定位、边缘计算等新技术,构建以地面设备动态监测、检测车自动检测相结合的电务设备智能检测监测体系,实现电务设备检测监测全面化、自动化、智能化和集成管理;采用BIM、高精度地图、大数据、人工智能等先进技术,建立集设备全寿命周期管理、检测监测一体化管理、状态评估与运维决策、安全生产管理以及环境监测管理五大功能模块的电务设备智能运维系统.系统实现了设备资产数字化、检测监测一体化、决策评估智能化、检修作业标准化、环境监测可视化,探索了重载铁路电务设备运维管理新模式.

针对传统重载铁路电务设备运维能力低,导致运维作业精准度不高的问题,提出基于深度学习的重载铁路电务设备智能运维系统.首先采用监测运维一体化采集电务设备运维数据;然后基于大数据技术,将递归神经网络RvNN、树形长短期记忆网络Tree-LSTM和树形卷积神经网络TBCNN三个树形神经网络进行融合,并与循环神经网络及衔生算法相结合,构建一个基于联锁逻辑时序与深度学习相结合的设备运维模型;最后通过构建模型进行联锁故障诊断和设备状态评估.实验结果表明,在二分类任务中,本模型的故障诊断准确率高达98.54％;在多分类任务中的诊断准确率为89.13％,对比于单一的树形结构模型和BP神经网络模型,多分类任务中的诊断率分别高出了15％和20％.系统应用发现,该系统能够进行重载铁路电务设备运维状态预测和准确评估,实现了电务设备全生命周期管理和自动化运维.

针对某城市轨道交通线路全电子联锁系统道岔模块死机故障导致道岔无法正常操动的问题,提出了一种道岔模块复位控制系统的设计方案.当道岔模块死机时,通过人工或自动的方式,远程控制道岔模块进行快速复位,以缩短故障处理时间,提高列车运行准点率.

针对电务实操考核中经常出现的器材损坏、培训效率低、培训记录不全等问题,提出了一种电务实操智能考核系统设计方案.通过与电务段实训基地的配套使用,能够实现智能化考核、智能化演练,并能对考核和演练的全过程进行记录.该系统已应用在长春电务段实训基地,取得了很好的效果.

According to the maintenance requirement of the SICAS interlocking system of Guangzhou metro Line 1,we designed a test based on centralized signal monitoring and give the design principle,key technologies applied in the system,and major system functions.

According to the maintenance requirements of signal equipment after the large-scale speed-up campaign,a set of intelligent analysis and fault diagnosis system were designed based on microcomputer monitoring.The design principle,key technologies,and realized functions of the system were given.