摘要
系统阐述人工智能(Artificial Intelligence,AI)大模型在城市轨道交通场景的应用实践,聚焦于AI大模型技术与轨道交通领域的深度融合,突破传统人工巡检与经验判断导致的效率瓶颈。通过预训练与有监督微调,模型能够有效适配轨道交通的具体应用场景,并结合私域知识库构建与系统集成,形成智能工作辅助体系,为行业智能化转型提供可行的技术路径。应用路径包括:构建统一的专题知识库与问答资源,提升模型的专业化水平;开发智能运维助手,实现故障预测与实时决策支持;优化视频分析等重复性任务的执行准确率。
This paper systematically elaborates on the application practices of large-scale AI(Artificial Intelligence)models in urban rail transit scenarios,focusing on the deep integration of AI large model technology with the rail transit field.It aims to overcome the efficiency bottlenecks caused by traditional reliance on manual inspections and experience-based judgments.The model can eff ectively adapt to specific application scenarios in rail transit through pre-training and supervised fine-tuning.With the support of private knowledge bases and system integration,the model can create an intelligent work assistance system,providing a feasible technical path for the intelligent transformation of the industry.The application pathways include:build a unified thematic knowledge base and Q&A resources to enhance the model's professional level;develop an intelligent operation and maintenance assistant to achieve fault prediction and real-time decision-making support;and improve the execution accuracy of repetitive tasks such as video analysis.
作者
刘伟林
Liu Weilin(Shenzhen Metro Operation Group Co.,Ltd.,Shenzhen 518000,China)
出处
《铁路通信信号工程技术》
2025年第12期113-117,130,共6页
Railway Signalling & Communication Engineering
基金
深圳市级交通港航类政府投资项目(2018-440300-54-01-719465)
深圳地铁运营集团有限公司科研项目(SZJC-0030/2025)。
关键词
人工智能
轨道交通
大模型
运营效率
视频分析
智能调度
artificial intelligence
rail transit
large model
operational efficiency
video analytics
intelligent scheduling
