| 157 | 0 | 128 |
| 下载次数 | 被引频次 | 阅读次数 |
[目的]准确预测在役防火涂料的耐火性能是消防安全评估工作中的重点和难点之一。[方法]借助傅里叶变换红外光谱仪，对不同酸腐蚀周期的膨胀型钢结构防火涂料试样进行测试，发现涂料中的主要成分会与盐酸溶液发生双向迁移并产生反应，酸腐蚀24 h时该反应达到峰值。利用主成分分析法对红外光谱数据进行降维处理的结果表明不同酸腐蚀周期试样存在可分性。联合使用S-G卷积平滑和标准正态变化(SNV)法对数据进行预处理，并对常用的析)、LW-PLSC (局部加权偏最小二乘分类)、K-ELM(核极限学习机)、SVM(支持向量机)等4种分类方法进行评估。[结果]测试集平均分类准确度高达94.25%，最优模型分类准确率可达100%。[结论]将红外光谱技术与机器学习结合起来能快速、准确、定量地评估钢结构防火涂料的实验室酸腐蚀周期，为实现防火涂料实际耐火性能的预测提供可行的新方法。
Abstract：[Objective] Accurately predicting the fire resistance of in-service fireproof coatings is one of the key and difficult problems in fire safety assessment. [Method] The samples of intumescent fireproof coatings for steel structures subjected to different cycles of acid corrosion were analyzed by using Fourier-transform infrared spectrometer. It was found that the main components in the paint migrated toward and reacted with the hydrochloric acid solution, which also migrated inside the coating, and the reaction peaked when the coating was corroded by hydrochloric acid for 24 hours.The dimensionality of infrared spectral data was reduced by principal component analysis, which showed that the samples subjected to different cycles of acid corrosion were separable. The S-G convolution smoothing and SNV method were used to preprocess the data, and four commonly used classification methods i.e. PLS-DA(partial least squaresdiscriminant analysis), LW-PLSC(locally weighted–partial least square classification), K-ELM(kernel extreme learning machine), and SVM(support vector machine) were evaluated. [Result] The average classification accuracy of the test set was as high as 94.25%, and the classification accuracy of the optimized model was up to 100%. [Conclusion] The combination of infrared spectroscopic analysis technology with machine learning can quickly, accurately, and quantitatively evaluate the cycle number of acid corrosion in laboratory corresponding to the fire-retardant coating for steel structures, providing a new and feasible method for predicting the actual fire resistance of fire-retardant coatings.
[1] H??LER D, MUND M, DAUS L H, et al. Durability of intumescent coatings and recommendations for test concepts for a working life of more than 10 years[J]. Fire Safety Journal, 2024, 146:104173.
[2] MACIULAITIS R, GRIGONIS M, MALAISKIENE J. The impact of the aging of intumescent fire protective coatings on fire resistance[J].Fire Safety Journal, 2018, 98:15-23.
[3]王国辉,王俊胜,刘丹,等.冻融循环试验对膨胀型防火涂料防火性能及热稳定性的影响[J].涂层与防护, 2018, 39(11):33-37, 44.WANG G H, WANG J S, LIU D, et al. Effects of freeze-thaw cycle test on fire retardancy and thermal stability of intumescent fire retardant coatings[J]. Coating and Protection, 2018, 39(11):33-37, 44.
[4]刘成杰,许丽颖,吴慧焱,等.钢结构防火涂料老化阶段耐火性能快速检测方法探讨[J].广东化工, 2023, 50(18):153-155, 91.LIU C J, XU L Y, WU H Y, et al. Discussion on rapid detection method of fire resistance in the weathering stage of steel structure fireproof coating[J]. Guangdong Chemical Industry, 2023, 50(18):153-155, 91.
[5] WANG J B, WANG G J. Influences of montmorillonite on fire protection, water and corrosion resistance of waterborne intumescent fire retardant coating for steel structure[J]. Surface and Coatings Technology, 2014, 239:177-184.
[6] LIU Z T, DAI M Q, HU Q H, et al. Effect of microencapsulated ammonium polyphosphate on the durability and fire resistance of waterborne intumescent fire-retardant coatings[J]. Journal of Coatings Technology and Research, 2019, 16(1):135-145.
[7]丁浩晗,田嘉伟,谢祯奇,等.机器学习和大数据在食品领域的应用[J].食品与发酵工业, 2024, 50(24):353-361.DING H H, TIAN J W, XIE Z Q, et al. Applications of machine learning and big data in food industry[J]. Food and Fermentation Industries,2024, 50(24):353-361.
[8]魏超杰,解宏鑫,王伟,等.基于同步辐射X射线荧光光谱与机器学习的非靶标金属组学预测PVC纳米塑料对水稻毒性的浓度依赖性[J].中国无机分析化学, 2024, 14(8):1015-1021.WEI C J, XIE H X, WANG W, et al. Prediction of the concentrationdependence toxicity of pvc nanoplastics to rice through non-targeted metallomics with synchrotron radiation x-ray fluorescence and machine learning[J]. Chinese Journal of Inorganic Analytical Chemistry, 2024,14(8):1015-1021.
[9]黄梓萱,李芷瑶,朱怡亭,等.表面增强拉曼光谱在细菌耐药性评价方面的研究进展[J].化学通报, 2024, 87(5):550-558, 549.HUANG Z X, LI Z Y, ZHU Y T, et al. Research advances in SERS technology for accessing bacterial drug resistance[J]. Chemistry, 2024,87(5):550-558, 549.
[10]胡爱琴.防火材料及灭火剂快速检测技术研究[D].北京:北京化工大学, 2013.HU A Q. Study on rapid quality evaluation of fire extinguishing agent and fire protection material[D]. Beijing:Beijing University of Chemical Technology, 2013.
[11]薛岗,宋文琦,李树超.基于近红外光谱技术的钢结构防火涂料品牌鉴别方法研究[J].光谱学与光谱分析, 2015, 35(1):104-107.XUE G, SONG W Q, LI S C. Study on discrimination of varieties of fire resistive coating for steel structure based on near-infrared spectroscopy[J]. Spectroscopy and Spectral Analysis, 2015, 35(1):104-107.
[12] ZHANG Y M, SONG W R, ZHAO S Y, et al. Brand identification of transparent intumescent fire retardant coatings using portable Raman spectroscopy and machine learning[J]. Vibrational Spectroscopy, 2022,122:103428.
[13]岳鑫,朱智超,宋惟然,等.高光谱成像与短视频成像结合机器学习应用于检测防火涂料的一致性[J].涂料工业, 2024, 54(3):46-53.YUE X, ZHU Z C, SONG W R, et al. Fire retardant coatings identification using hyperspectral imaging and short video imaging combined with machine learning[J]. Paint&Coatings Industry, 2024,54(3):46-53.
[14]吕志文.漆膜成分现场检测技术[D].保定:华北电力大学, 2020.LYU Z W. On-site detection technology of paint composition[D].Baoding:North China Electric Power University, 2020.
[15]王霁.纺织类企业用膨胀型钢结构防火涂料的老化及性能研究[D].天津:天津工业大学, 2015.WANG J. Study on the aging and performance of the intumescent fire retardant coatings for steel structures used in textile industries[D].Tianjin:Tiangong University, 2015.
[16]李晓康,王霁,徐志胜.酸腐蚀条件下膨胀型防火涂料性能变化研究[J].安全与环境学报, 2017, 17(6):2228-2233.LI X K, WANG J, XU Z S. Study on the performance change of the fire retardant intumescent coating due to the acid corrosion[J]. Journal of Safety and Environment, 2017, 17(6):2228-2233.
[17]陈博文.基于傅里叶红外光谱技术的翅荚木人工林木材材性预测[D].福州:福建农林大学, 2022.CHEN B W. Prediction of wood properties of zenia insignis chun plantation based on Fourier transform infrared spectroscopy[D].Fuzhou:Fujian Agriculture and Forestry University, 2022.
[18]吴菲.基于冠层光谱的设施番茄水分监测诊断模型研究[D].扬州:扬州大学, 2022.WU F. Studies on water monitoring and diagnosis models of greenhouse tomato based on canopy spectrum[D]. Yangzhou:Yangzhou University, 2022.
[19] GALVAO R K H, ARAUJO M C U, JOSéG E, et al. A method for calibration and validation subset partitioning[J]. Talanta, 2005, 67(4):736-740.
[20]张娟.基于SPXY-WT-CARS算法的草莓糖度近红外光谱检测研究[J].食品与发酵科技, 2020, 56(6):136-139, 142.ZHANG J. Research on the detection of strawberry sugar content by NIR based on SPXY-WT-CARS algorithm[J]. Food and Fermentation Sciences&Technology, 2020, 56(6):136-139, 142.
[21]李盼盼.甲氨蝶呤对胶原诱导性关节炎大鼠血清代谢谱的影响及机制研究[D].太原:山西医科大学, 2023.LI P P. Effect and mechanism of methotrexate on serum metabolic profile in collagen-induced arthritis rats[D]. Taiyuan:Shanxi Medical University, 2023.
[22]胡昆.基于气相色谱法结合化学计量学识别微量植物油的研究[D].北京:中国人民公安大学, 2023.HU K. Study on the identification of trace vegetable oils based on gas chromatography combined with chemometrics[D]. Beijing:People’s Public Security University of China, 2023.
[23] YUAN X F, ZHOU J, WANG Y L. A spatial-temporal LWPLS for adaptive soft sensor modeling and its application for an industrial hydrocracking process[J]. Chemometrics and Intelligent Laboratory Systems, 2020, 197:103921.
[24]卢泳,王洛临,张建军,等.麸炒白术中白术内酯I、II、III含量近红外快速分析方法的建立[J].中华中医药杂志, 2021, 36(9):5599-5605.LU Y, WANG L L, ZHANG J J, et al. Establishment of NIRS rapid method for analysis on content of atractylenolide I, II, III in rhizoma atractylodis macrocephalae processed by stir-frying with bran[J].China Journal of Traditional Chinese Medicine and Pharmacy, 2021,36(9):5599-5605.
[25]乔威豪,安葳鹏,赵雪菡,等.基于改进极限学习机的煤与瓦斯突出预测研究[J].矿业研究与开发, 2024, 44(5):98-105.QIAO W H, AN W P, ZHAO X H, et al. Research on coal and gas outburst prediction based on improved extreme learning machine[J].Mining Research and Development, 2024, 44(5):98-105.
[26]谢百亨,马晋芳,周泳欣,等.高光谱结合哈里斯鹰优化核极限学习机鉴别化橘红胎切片年份[J].光谱学与光谱分析, 2024, 44(5):1494-1500.XIE B H, MA J F, ZHOU Y X, et al. Identification of citri grandis fructus immaturus based on hyperspectral combined with HHO-KELM[J]. Spectroscopy and Spectral Analysis, 2024, 44(5):1494-1500.
[27] DING S F, SHI Z Z, TAO D C, et al. Recent advances in support vector machines[J]. Neurocomputing, 2016, 211:1-3.
[28]李福祥,王雪,张驰,等.基于边界点的支持向量机分类算法[J].陕西理工大学学报(自然科学版), 2022, 38(3):30-38.LI F X, WANG X ZHANG C, et al. Support vector machine classification algorithm based on boundary points[J]. Journal of Shaanxi University of Technology(Natural Science Edition), 2022, 38(3):30-38.
[29]李召桐.支持向量机发展历程及其应用[J].信息系统工程, 2024(3):124-126.LI Z T. The development history and applications of support vector machines[J]. China CIO News, 2024(3):124-126.
[30]李娜娜,王勇,周林,等.基于特征重要度二次筛选的DDoS攻击随机森林检测方法[J].计算机科学, 2021, 48(增刊1):464-467, 476.LI N N, WANG Y, ZHOU L, et al. DDoS attack random forest detection method based on secondary screening of feature importance[J]. Computer Science, 2021, 48(Suppl.1):464-467, 476.
基本信息:
DOI：10.19289/j.1004-227x.2025.01.006
中图分类号:TQ637;O657.33;TP181
引用信息:
[1]岳鑫,刘培东,王今等.基于傅里叶变换红外光谱与机器学习的防火涂料酸腐蚀周期预测研究[J].电镀与涂饰,2025,44(01):37-44.DOI:10.19289/j.1004-227x.2025.01.006.
基金信息: