基于多元融合技术在辅控监控系统中的研究及应用  

Research and Application of Multi-Modality Fusion Technology in Auxiliary Monitoring Systems

作　　者：宗起振 詹妍 曾清旋 Zong Qi-zhen;Zhan Yan;Zeng Qing-xuan(Nanjing Guodian Nanzi Urban Rail Transit Engineering Co.,Ltd.,Nanjing 210032,Jiangsu Province,China)

机构地区：[1]南京国电南自轨道交通工程有限公司,江苏南京210032

出　　处：《科学与信息化》2025年第7期54-56,共3页Technology and Information

摘　　要：本文围绕牵引变电所现有运维场景,在分析研究传统牵引变电所运维模式向无人化、数字化和智能化转型过程中普遍存在的问题,提出以计算机视觉为核心,对物联网、大数据、人工智能及三维可视化等技术进行多元融合,同时提高图像识别的鲁棒性和准确率,实现以多维度呈现变电所运维场景,对变电所全方位多角度运行状态实时监测、运维设备互联互通,对多源异构数据进行融合、报警分级管理及多级联动控制,从而进一步提高变电所设备的全状态感知力与控制力,有效降低运维成本,增强安全生产保障能力,提高运检精益管理水平。This paper focuses on the existing operation and maintenance(O&M)scenarios of traction substations.By analyzing the common issues encountered during the transition from traditional O&M models to unmanned,digital,and intelligent modes,a multi-modality fusion approach is proposed.This approach centers on computer vision and integrates technologies such as the Internet of Things(IoT),big data,artificial intelligence,and 3D visualization.It enhances the robustness and accuracy of image recognition,enabling multi-dimensional presentation of substation O&M scenarios.It achieves real-time monitoring of the substation's operational status from multiple perspectives,interconnectivity of O&M equipment,and fusion of multi-source heterogeneous data.Additionally,it supports alarm grading management and multi-level collaborative control.This ultimately improves the comprehensive perception and control capabilities of substation equipment,reduces O&M costs,enhances safety production assurance,and improves the lean management level of operation and inspection.

关 键 词：多元融合 三维可视化 图像识别 牵引变电所 辅助监控 

分 类 号：U224[交通运输工程—道路与铁道工程]