-
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。Abstract: Based on advanced fire sensors, the Bayesian network model of recognizing fire size and category intelligently was established, six fire characteristic parameters including upper temperature, lower temperature, CO concentration, CO2 concentration, O2concentration 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 fireexperimental 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 antiship 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.
-
Key words:
- ship /
- compartment fire /
- Bayesian network /
- intelligent identification /
- fire sensor /
- fire size /
- fire category
-
图 14 图 14小火识别概率验证结果
Figure 14. Verification results of recognition probabilities of small fire
图 15 中火识别概率验证结果
Figure 15. Verification results of recognition probabilities of medium fire
图 16 大火识别概率验证结果
Figure 16. Verification results of recognition probabilities of large fire
图 18 固体火识别概率验证结果
Figure 18. Verification results of recognition probabilities of solid fire
图 19 油料火识别概率验证结果
Figure 19. Verification results of recognition probabilities of fuel oil fire
表 11 传感器失效时验证试验的识别正确率
Table 11. Recognition accuracy rates of verification test with failure sensor
下载: 导出CSV
-
[1] LEBLANC D. Fire environments typical of navy ships[D]. London: Worcester Polytechnic Institute, 1998. [2] CALABRESE F, CORALLO A, MARGHERITA A, et al. A knowledge-based decision support system for shipboard damage control[J]. Expert Systems with Applications, 2012, 39(9): 8204-8211. doi: 10.1016/j.eswa.2012.01.146 [3] OWRUTSKY J C, STEINHURST D A, MINOR C P, et al. Long wavelength video detection of fire in ship compartments[J]. Fire Safety Journal, 2006, 41(4): 315-320. doi: 10.1016/j.firesaf.2005.11.011 [4] GOTTUK D T, LYNCH J A, ROSE-PEHRSSON S L, et al. Video image fire detection for shipboard use[J]. Fire Safety Journal, 2006, 41(4): 321-326. doi: 10.1016/j.firesaf.2005.12.007 [5] ROSE-PEHRSSON S L, SHAFFER R E, HART S J, et al. Multi-criteria fire detection systems using a probabilistic neural network[J]. Sensors and Actuators B: Chemical, 2000, 69(3): 325-335. doi: 10.1016/S0925-4005(00)00481-0 [6] KUO H C, CHANG H K. A real-time shipboard fire-detection system based on grey-fuzzy algorithms[J]. Fire Safety Journal, 2003, 38(4): 341-363. doi: 10.1016/S0379-7112(02)00088-7 [7] ROSE-PEHRSSON S L, HART S J, STREET T T, et al. Early warning fire detection system using a probabilistic neural network[J]. Fire Technology, 2003, 39(2): 147-171. doi: 10.1023/A:1024260130050 [8] WANG S J, JENG D L, TSAI M T. Early fire detection method in video for vessels[J]. Journal of Systems and Software, 2009, 82(4): 656-667. doi: 10.1016/j.jss.2008.09.025 [9] WILKINS D C, SNIEZEK J A, TETEM PA, et al. The DC-SCS supervisory control system for ship damage control: volume 1—design overview[R]. Washington DC: Naval Research Laboratory, 2001. [10] ROSE-PEHRSSON S L, MINOR C P, STEINHURST D A, et al. Volume sensor for damage assessment and situational awareness[J]. Fire Safety Journal, 2006, 41(4): 301-310. doi: 10.1016/j.firesaf.2005.12.005 [11] MINOR C P, JOHNSON K J, ROSE-PEHRSSON S L. A full-scale prototype multisensor system for fire detection and situational awareness[C]//SPIE. Proceedings of SPIE 6571, Multisensor, Multisource Information Fusion: Architectures, Algorithms, and Applications. Bellingham: SPIE, 2007: 11-12. [12] 李卡麟. 基于二叉树的LS-WSVM模型在早期火灾分类上的研究[D]. 汕头: 汕头大学, 2010.LI Ka-lin. Research on LS-WSVM based on binary tree in early fire multi-class classification[D]. Shantou: Shantou University, 2010. (in Chinese) [13] 庄哲民, 李卡麟, 张新蜂, 等. 用于早期火灾分类的非线性决策树支持向量机[J]. 火灾科学, 2009, 18(4): 206-211. doi: 10.3969/j.issn.1004-5309.2009.04.004ZHUANG Zhe-min, LI Ka-lin, ZHANG Xin-feng, et al. Nonlinear decision tree support vector machine for early fire classification[J]. Fire Safety Science, 2009, 18(4): 206-211. (in Chinese) doi: 10.3969/j.issn.1004-5309.2009.04.004 [14] 孙福志, 于军琪, 杨柳. 火灾识别中RS-SVM模型的应用[J]. 计算机工程与应用, 2010, 46(3): 198-200. doi: 10.3778/j.issn.1002-8331.2010.03.061SUN Fu-zhi, YU Jun-qi, YANG Liu. Application of RS-SVM model for fire identification[J]. Computer Engineering and Application, 2010, 46(3): 198-200. (in Chinese) doi: 10.3778/j.issn.1002-8331.2010.03.061 [15] 赵亚琴. 基于模糊神经网络的火灾识别算法[J]. 计算机仿真, 2015, 32(2): 369-373. doi: 10.3969/j.issn.1006-9348.2015.02.080ZHAO Ya-qin. Forest fire recognition algorithm based on fuzzy neural network[J]. Computer Simulation, 2015, 32(2): 369-373. (in Chinese) doi: 10.3969/j.issn.1006-9348.2015.02.080 [16] KIM J H, LATTIMER B Y. Real-time probabilistic classification of fire and smoke using thermalimagery for intelligent firefighting robot[J]. Fire Safety Journal, 2015, 72: 40-49. doi: 10.1016/j.firesaf.2015.02.007 [17] AKHTAR M J, UTNE I B. Human fatigue's effect on the risk of maritime groundings—a Bayesian network modeling approach[J]. Safety Science, 2014, 62: 427-440. doi: 10.1016/j.ssci.2013.10.002 [18] GROIS E, WILKINS D C, EARMAN I, et al. The DC-SCS supervisory control system for ship damage control: volume4—intelligent reasoning[R]. Washington DC: Naval Research Laboratory, 2001. [19] BAKSH A A, KHAN F, GADAG V, et al. Network based approach for predictive accident modelling[J]. Safety Science, 2015, 80: 274-287. doi: 10.1016/j.ssci.2015.08.003 [20] 刘志军, 纪卓尚, 林焰. 基于贝叶斯网络的船舶机舱火灾风险分析研究[J]. 中国造船, 2010, 51(3): 199-205. doi: 10.3969/j.issn.1000-4882.2010.03.024LIU Zhi-jun, JI Zhuo-shang, LIN Yan. Fire risk analysis in ship engine room based on Bayesian networks[J]. Shipbuilding of China, 2010, 51(3): 199-205. (in Chinese) doi: 10.3969/j.issn.1000-4882.2010.03.024 [21] WILLIAMS F W, SCHEFFEY J L, HILL S A, et al. Postflashover fires in shipboard compartments aboard ex-USS Shadwell: phase V—fire dynamics[R]. Washington DC: Naval Research Laboratory, 1999. [22] WILLIAMS F W, TATEM P A, XUAN N, et al. Results of1998DC-ARM/ISFE demonstration tests[R]. Washington DC: Naval Research Laboratory, 2000. [23] HOOVER J B, WHITEHURST C L, CHANG E B, et al. Final report on fire performance of shipboard electronic space materials[R]. Washington DC: Naval Research Laboratory, 2006. [24] HOOVER J B, BAILEY J L, WILLAUER H D, et al. Evaluation of submarine hydraulic system explosion and fire hazards[R]. Washington DC: Naval Research Laboratory, 2005. [25] WONG J T, GOTTUK D T, ROSE-PETHRSSON S L, et al. Results of multi-criteria fire detection system tests[R]. Washington DC: Naval Research Laboratory, 2000. -
点击查看大图
图(20) / 表(11)
计量
- 文章访问数: 414
- HTML全文浏览量: 73
- PDF下载量: 940
- 被引次数: 0
