光纤传感器感温性能分析修正

消防科学与技术 ›› 2022, Vol. 41 ›› Issue (10): 1425-1429.

光纤传感器感温性能分析修正

王岚1，孟弘融2,3，曲冠华3,4，任蕾3,4

（1.应急管理部天津消防研究所，天津 300381；2.天津大学国际工程师学院，天津 300072；3.天津大学天津市建筑物理环境与生态技术重点实验室，天津 300072；4.天津大学建筑学院，天津 300072）

出版日期:2022-10-15 发布日期:2022-10-15
通讯作者: 应急管理部消防救援局重点攻关项目（2020XFZD02）
作者简介:作者简介:王岚(1980- )，女，河南洛阳人，应急管理部天津消防研究所助理研究员，主要从事防灾减灾领域理论与技术研究，天津市南开区卫津南路110号，300381。

Analysis and correction of temperature sensing performance of optical fiber sensor

Wang Lan1, Meng Hongrong2,3, Qu Guanhua3,4, Ren Lei3,4

(1. Tianjin Fire Science and Technology Research Institute of MEM, Tianjin 300381, China; 2. International School of Engineers, Tianjin University, Tianjin 300072, China; 3. Tianjin Key Laboratory of Building Physical Environment and Ecological Technology, Tianjin University, Tianjin 300072, China; 4. School of Architecture, Tianjin University, Tianjin 300072, China)

Online:2022-10-15 Published:2022-10-15

摘要/Abstract

摘要： 摘要：为提升光纤传感器在建筑火灾预警中的适用性，本研究针对单模、多模普通光纤开展基于常用实际火灾升温曲线的高温试验以及感温性能分析，进而遴选感温性能更适用于建筑火灾条件下的传感器。同时，基于试验数据建立升温误差修正模型，以提升测温精度。结果表明：在测温稳定性方面，多模光纤传感器远优于单模光纤传感器，多模光纤传感器感温性能在建筑火灾预警中适用性更好，且更具有修正价值，初步建立的升温误差修正模型使其测温平均绝对误差降低至13.12 ℃。利用分布式光纤进行高精度建筑火灾地图绘制，成为火灾监测预警、疏散救援以及结构倒塌预测的技术关键。

关键词: 关键词：分布式光纤, 火灾经验模型, 感温性能, 升温曲线

Abstract: Abstract: Based on the current emergencies in the field of firefighting and rescue, an intelligent matching algorithm for firefighting and rescue expertise is proposed. The algorithm calculates the semantic relationship between case description and firefighting knowledge based on natural language processing and attention mechanism, realizing the matching of related firefighting and rescue knowledge. First learn semantic information at sentence granularity level based on natural language processing, then learns semantic information at word granularity level based on the attention mechanism, and finally, based on the interaction of two levels of semantic information, the relationship between sentences is inferred based on the local differences of the information. Experimental results show that the algorithm has excellent performance. At the same time, it can understand sentences more accurately from the two granularities of words and sentences, realize intelligent matching of firefighting and rescue professional knowledge based on case descriptions.

Key words: Key words: fire rescue, intelligent matching, natural language processing, attention mechanism, semantic information

王岚, 孟弘融, 曲冠华, 任蕾, . 光纤传感器感温性能分析修正[J]. 消防科学与技术, 2022, 41(10): 1425-1429.

Wang Lan, Meng Hongrong, Qu Guanhua, Ren Lei, . Analysis and correction of temperature sensing performance of optical fiber sensor[J]. Fire Science and Technology, 2022, 41(10): 1425-1429.