基于人工智能的基建工程质量通病智能识别研究与应用
Research and Application of Intelligent Identification of Common Quality Problems of Infrastructure Engineering Based on Artificial Intelligence
王旭东 1阎玮 2陈伟1
作者信息
- 1. 山西益通电网保护自动化有限责任公司,山西 太原 030008
- 2. 国网山西省电力公司建设部,山西 太原 030008
- 折叠
摘要
随着人工智能技术的日趋成熟,其应用也不断增加,针对海量的数据资源也亟须应用新技术进行数据的深度挖掘分析.通过将人工智能技术引入基建工程现场质量通病业务场景中,应用基于深度学习的目标检测模型,构建智能识别应用实现工程现场质量通病的智能识别与告警,并通过样本增强技术及结果去重技术等相关技术,提高模型识别精度,提升目标检测效率.该项目的研究使得质量通病问题由传统的人工巡检和监控检查向自动化、智能化的机器识别转变,有效提升了工程现场质量通病监管效率和智能化水平,为安全、优质、高效地完成电网建设任务奠定了坚实的基础.
Abstract
As artificial intelligence technology matures,the application of artificial intelligence is also increasing,and it is urgent to apply new technologies for in-depth data mining and analysis for massive data resources.Through the introduction of artificial intelli-gence technology into the business scenarios of common quality problems of infrastructure engineering site,intelligent identification ap-plication to realize intelligent identification and alarm of common quality problems on the engineering site is constructed by applying object detection model based on deep learning.What's more,model identification accuracy and target detection efficiency are im-proved by sample enhancement technology and result de-duplication technology.The research of this project has transformed the com-mon quality problems from the traditional manual inspection and monitoring inspection to the automatic and intelligent machine identi-fication,effectively improving the supervision efficiency and intelligent level of the common quality problems on the project site,which lays a solid foundation for the safe,high-quality and efficient completion of the power grid construction tasks.
关键词
人工智能/智能识别/质量通病/电网建设Key words
artificial intelligence/intelligent identification/common quality problems/power grid construction引用本文复制引用
出版年
2024
NETL
NSTL
万方数据