典型舱室火灾智能识别的贝叶斯网络模型  

Bayesian network model for intelligent recognition of typical compartment fire

作　　者：谢田华[1,2] 林焰[1] 杨祖耀[3] 金良安[2] 

机构地区：[1]大连理工大学船舶工程学院,辽宁大连116024 [2]海军大连舰艇学院航海系,辽宁大连116018 [3]海军装备研究院标准规范研究所,上海200235

出　　处：《交通运输工程学报》2016年第2期90-99,共10页Journal of Traffic and Transportation Engineering

基　　金："十二五"国防预研项目(4010403010208);武器装备军内科研重点项目(2014HJ0030)

摘　　要：基于先进传感器,建立了火灾大小和类型智能识别的贝叶斯网络模型,上层温度、下层温度、CO浓度、CO2浓度、O2浓度和遮光度等6个火灾特征参数为识别模型的输入变量,火灾大小和类型为输出变量,并推导了输入变量与输出变量之间的关系。分别在住舱、指挥室、机舱和机库等4种典型舱室模拟了床垫火、电缆火、油池火和喷射火等4种火源,利用CFAST软件得到了2 880组模拟样本数据,对模型参数进行了训练,并根据全尺度火灾试验数据对训练后的识别模型进行了验证。验证结果表明:在火灾传感器数据完整时,对小火、中火和大火状态的平均识别正确率分别为88.0%、95.0%、85.7%,对固体火和油料火的平均识别正确率分别为90.2%、81.5%;在火灾损害严重或武器打击致使单个传感器失效的情况下,对火灾大小和类型的平均识别正确率分别为82.4%、82.7%,比火灾传感器数据完整时分别降低8.1%、2.8%。可见,识别模型具有良好的识别能力和鲁棒性,可应用于舰船损管监控系统,为指挥员选择最有效的灭火方法和战术提供实时的决策支持。Based on advanced fire sensors, category intelligently was established, the Bayesian network model of recognizing fire size and six fire characteristic parameters including upper temperature, lower temperature, CO concentration, CO2 concentration, O2 concentration and light obscuration were considered as the input variables of the model, fire size and category were considered as the output variables of the model, and the relationship between input and output variables was deduced. Four kinds of fire sources including mattress fire, cable fire, pool fire and spill fire were simulated in living room, combat center, engine room and hangar respectively, 2 880 groups of simulated sample data were generated by using CFAST software, the parameters of the model were trained, and the trained recognition model was validated by using full-scale fire experimental data. Validation result indicates that when the data of fire sensors are intact, the average recognition accuracy rates are 88.0%, 95.0% and 85.7% for small, medium and large fires respectively, and the average recognition accuracy rates are 90.2% and 81.5% for solid and oil fires respectively. When one sensor can＇t work because of serious damage or the hit of anti- ship weapon, the average recognition accuracy rates are 82.4% and 82.7% for fire＇s size and category, only 8.1% and 2.8 % less than the rates with integrated fire sensors data respectively. So, the proposed model has good recognition ability and robustness, and can be integrated into ship damage control supervisory system （DCSS） to assist commanders in timely selecting the most effective firefighting methods and tactics. 11 tabs, 20 figs, 25 refs.

关 键 词：舰船工程 舱室火灾 贝叶斯网络 智能识别 火灾传感器 火灾大小 火灾类型 

分 类 号：U661.7[交通运输工程—船舶及航道工程]