随着我国煤炭开采逐渐向深部延伸，矿用输送带长度逐年递增，带式输送机运行过程因机械损伤、摩擦打滑等原因，导致输送带升温、起火、燃烧等火灾事故频发，输送带燃烧后火焰烟气等在巷道内迅速蔓延，造成巨大的财产损失和人员伤亡。因此，开展高温热源条件下矿用输送带燃烧特性及烟气发展态势预测的研究，可为灾变条件下智能感知预警、避灾路线智能规划、科学救灾及通风系统智能调控提供理论基础与决策依据。

本论文针对前人在矿井阻燃输送带热解动力学特征参数及气体产物类型的本构关系、烟气发展蔓延空间分布特征及预测预警方法等方面研究的不足，围绕矿用阻燃输送带热解燃烧特性、烟气发展时空演化规律和态势预测方法等技术难题，以矿井常用的织物整芯阻燃PVC和PVG输送带为研究对象，采用小尺寸实验、大尺寸实验和机器学习算法预测相结合的研究方法，开展了阻燃输送带热解燃烧、烟气发展蔓延特性及其态势预测的研究，取得以下创新成果：

通过同步热分析-傅里叶红外光谱联用实验，研究不同材质不同拉力强度（680S和1000S）下阻燃输送带在热解过程中的特征温度、质量变化特征、热释放特征和气体产物类型。采用KAS和FWO两种无模式函数法计算不同环境气氛下多升温速率阻燃输送带在热解阶段的表观活化能，确定最概然机理函数，掌握阻燃输送带在热解不同阶段中气体产物类型。得出PVC阻燃性能小于PVG，PVC680S阻燃性能最差，表现为热解初始温度、燃尽温度、放热量和活化能最低。发现输送带最后一阶段热解过程为非典型的凝固相热解机理。确定输送带第一阶段主要为HCl的脱除，PVG同时会释放出氰酸类气体，并产生多烯烃和芳香族化合物；第二阶段为多烯烃链的断裂，产生多种稠环芳烃；第三阶段为剩余C、H产物被剧烈氧化分解的过程。研究结果揭示了矿用阻燃输送带热动力学特征参数及气体、挥发性产物的本质响应机制。

采用锥形量热实验，研究不同阻燃输送带的燃烧特性，掌握外加辐射热强度对阻燃输送带燃烧特性的影响规律。得出输送带点燃时间随外加辐射热强度增加而降低，随阻燃性能增加而增大的规律。发现外加辐射热强度与质量损失、热释放速率、烟生成速率、总烟释放量呈正比关系，阻燃性能和拉伸强度与总烟释放量呈正比关系，与质量损失系数和燃烧效率呈反比关系。确定不同外加辐射热强度下阻燃输送带燃烧残留物表面形态特征，阐述热辐射强度对矿井阻燃输送带燃烧的影响机制，揭示不同阻燃输送带的燃烧机理。

通过大尺寸输送带燃烧蔓延实验，研究阻燃输送带在不同风速下的燃烧蔓延过程中，烟气的气体浓度及温度在巷道横向与竖向的时空分布规律，结合锥形量热实验的燃烧参数，确定燃烧过程中的火源热释放速率。得出巷道上层空间测点烟气温度高于下层测点烟气温度，PVG阻燃输送带各测点烟气温度最高值、到达最高值点所需的时间均大于PVC阻燃输送带，且风速与测点最高温度值呈反比等结论。发现PVG阻燃输送带各测点到达烟气气体浓度最大值所需的时间、在最高值时持续的时间相较于PVC较长，PVG各测点烟气气体浓度增长速率均小于PVC，且风速与烟气气体浓度最大值呈反比。确定风速与热释放速率呈正比关系，且PVG阻燃输送带的热释放速率均大于PVC。掌握矿井阻燃输送带燃烧烟气空间温度场及气体浓度场的时空蔓延扩散规律。

采用CNN-LSTM深度学习算法，建立输送带烟气蔓延温度及气体浓度时空预测模型，通过与CNN、LSTM和BPNN三种单一模型对比验证CNN-LSTM模型的预测性能，确定了最佳预测精度模型。得出四种模型中，CNN-LSTM模型预测PVC和PVG阻燃输送带不同风速下不同测点位置烟气温度及浓度数据效果最好，MAE和RMSE最小，R²最大，性能远优于单一模型。表明本文构建的CNN-LSTM时空预测模型充分挖掘了各测点位置在空间上的信息，减小了预测误差，提升了预测性能，能够实现输送带燃烧过程危险程度的准确研判。

借助本文研究结果，可掌握不同风速条件下不同阻燃输送带燃烧烟气蔓延扩散规律，实现输送带火灾烟气蔓延温度和气体浓度分布的时序预测，对矿井输送带火灾的烟气发展态势预测与科学防控，具有重要的科学价值和实际指导意义。

As China's coal mining gradually extends deeper, the length of mining conveyor belts is increasing year by year. Due to mechanical damage, friction and slipping during the operation of the belt conveyor, those reasons cause the frequent fire accidents, such as heating up, catching fire, and burning of the conveyor belt. After the conveyor belt burns, the flames and smoke spread quickly, causing huge property damage and casualties. Therefore, study on combustion characteristics of mining conveyor belts and situation prediction of smoke development under high-temperature heat source conditions is conducted, it can provide a theoretical basis and decision-making basis for intelligent perception and early warning under disaster conditions, intelligent planning of disaster avoidance routes, scientific disaster relief and intelligent regulation of ventilation systems.

This paper aims at the shortcomings of previous research on the constitutive relationships between pyrolysis kinetics parameters and gas product types, spatial distribution characteristics of smoke development, as well as prediction and warning methods of flame-retardant conveyor belts in mines, surrounding those technical challenges, such as the pyrolysis combustion characteristics, spatiotemporal evolution of smoke development, and prediction and warning methods of mining flame-retardant conveyor belts, the commonly used fabric core flame-retardant PVC and PVG conveyor belts in mines are taken as the research object. The research method combining small-scale experiments, large-scale experiments, and machine learning algorithm prediction is adopted to conduct the study on the pyrolysis and combustion of flame-retardant conveyor belt, the characteristics of smoke development and spread, and its situation prediction. The following innovative results are summarized as follows:

Simultaneous thermal analysis and Fourier infrared spectroscopy combined experiment is adopted to study the characteristic temperature, mass change characteristics, heat release characteristics, and gas product types of flame-retardant conveyor belts during pyrolysis under different materials and tensile strengths (680S and 1000S). Two modeless function methods for KAS and FWO are adopted to calculate the apparent activation energy of flame-retardant conveyor belts with multiple heating rates during pyrolysis under different environmental atmospheres, and the most probable mechanism function is determined, the types of gas products in different stages of pyrolysis of flame-retardant conveyor belts are mastered. We find that PVC has less flame retardancy than PVG, and PVC680S has the worst flame retardant performance because of its lowest initial temperature of pyrolysis, burnout temperature, heat release, and apparent activation energy. Moreover, the final pyrolysis process of the conveyor belt is an atypical solidification phase pyrolysis mechanism. Meanwhile, the first stage of pyrolysis for conveyor belt mainly involves the removal of HCl. PVG conveyor belt releases cyanide gas, and produces polyolefin and aromatic compounds. The second stage is the fracture of polyolefin chains, various polycyclic aromatic hydrocarbons are produced. The third stage is the process where the remaining C and H products are violently oxidized and decomposed. The research results reveal the essential response mechanism between the thermodynamic characteristic parameters and gases and volatile products of the flame-retardant conveyor belts in mines.

Cone calorimetry experiment is used to study the combustion characteristics of different flame-retardant conveyor belts, and the influence of external radiation heat intensity on the combustion characteristics of flame-retardant conveyor belts is mastered. We find that the ignition time of the conveyor belt decreases with the increase of external radiation heat intensity, and it increases with the increase of flame retardancy. Furthermore, the external radiation heat intensity is directly proportional to mass loss, heat release rate, smoke generation rate, and total smoke release amount, the flame retardancy and tensile strength are directly proportional to the total smoke emission, and they are inversely proportional to the mass loss coefficient and combustion efficiency. The surface morphological characteristics of combustion residues for flame-retardant conveyor belt under different external radiation heat intensity are determined, and the mechanism of the influence of thermal radiant intensity on the combustion of mine flame-retardant conveyor belt is clarified, revealing the combustion mechanism of different flame-retardant conveyor belts.

The large-sized combustion and spread experiment for conveyor belt is adopted to study the spatiotemporal distribution of gas concentration and temperature for flame-retardant conveyor belt during combustion spread process in the horizontal and vertical directions of the tunnel under different wind speeds. Combining combustion parameters of cone calorimetry experiment, the heat release rate of the ignition source during the combustion process is determined. We find that the smoke temperature at the measurement point in the upper space of the tunnel is higher than that at the measurement point in the lower layer, in which the highest smoke temperature at each measuring point of flame-retardant conveyor belt for PVG and the time required to reach the highest point are greater than those of flame-retardant conveyor belt for PVC, and the wind speed is inversely proportional to the highest temperature value at the measuring point. Moreover, the required time that each measuring point of flame-retardant conveyor belt for PVG to reach the maximum concentration of smoke, and the duration at the highest value is longer compared to PVC. The growth rate of smoke concentration at each measuring point of PVG is smaller than that of PVC, and the wind speed is inversely proportional to the maximum smoke concentration, it is directly proportional to the heat release rate. Meanwhile, the heat release rate of PVG flame-retardant conveyor belt is higher than that of PVC. The spatiotemporal spread and diffusion laws of the temperature and gas concentration field for the combustion smoke space of flame-retardant conveyor belts in mines are mastered.

CNN-LSTM deep learning algorithm is used to establish a spatiotemporal prediction model for smoke spread temperature and concentration for conveyor belts. The prediction performance of CNN-LSTM model is verified by comparing with three single models, including CNN, LSTM and BPNN, the best prediction accuracy model is determined. In four models, the CNN-LSTM model has the best prediction effect on the smoke temperature and concentration data at different measurement points and wind speeds for PVC and PVG flame-retardant conveyor belts, and the smallest for MAE and RMSE with the largest R². The performance of CNN-LSTM is the best than that of a single model, indicating that the constructed CNN-LSTM spatiotemporal prediction model in this paper fully explores the spatial information of each measurement point position, reduces prediction errors, improves prediction performance, and it can realize the accurate judgment for danger degree during the combustion process of conveyor belts.

Based on the research results of this paper, we can grasp the smoke spread and diffusion laws of different flame-retardant conveyor belts under different wind speed conditions, realize time series prediction of smoke spread temperature and gas concentration distribution on conveyor belt, which has important scientific value and practical guiding significance for the situation prediction of smoke development and scientific prevention and control of combustion for mining conveyor belt.

[1]王德明. 煤矿热动力灾害及特性[J]. 煤炭学报, 2018, 43(1): 137-142.

[2]赵志强. 矿用带式输送机的使用及故障分析[J]. 当代化工研究, 2019, (12): 17-18.

[3]王烨. 矿井胶带火灾烟流蔓延特性与监测方法研究[D]. 西安: 西安科技大学, 2021.

[4]尤光星, 温达, 矦贵海, 等. 超支化膨胀型阻燃剂的合成及其在PVG输送带中的应用研究[J]. 塑料工业, 2014, 42(10): 17-20+35.

[5]周艺婷. 皮带巷灾变风流自动调控技术研究[D]. 唐山: 华北理工大学, 2020.

[6]郑宝莲. 煤矿外因火灾的原因及其防控[J]. 科技创新导报, 2017, 14(2): 55+57.

[7]黑龙江龙煤集团鸡西杏花煤矿"11·20"井下火灾事故抢险救援纪实[J]. 中国应急管理, 2015, 107(11): 56-57.

[8]重庆松藻煤矿"9·27"重大火灾事故37名公职人员被追责问责[J]. 消防界, 2021, 7(07): 44-45.

[9]李志宪, 陈江. 日本三井三池有明煤矿井下特大火灾事故[J]. 工业安全与防尘, 1990, (3): 42+21.

[10]驹井武, 井清武弘, 蒋时才, 等. 矿用运输胶带在井巷内燃烧特性的研究[J]. 矿业安全与环保, 1990, (2): 47-52.

[11]Miranda R, Pakdel H, Roy C, et al. Vacuum pyrolysis of commingled plastics containing PVC II. Product analysis[J]. Polymer Degradation and Stability, 2001, 73(1).

[12]Bart J. Polymer/additive analysis by flash pyrolysis techniques[J]. Journal of Analytical and Applied Pyrolysis, 2001, 58-59.

[13]Song ZC. Experimental study of kinetic properties of pyrolysis for conveyor belt in coal mine[J]. Journal of China University of Mining and Technology, 2005, 15(4).

[14]Singh S, Wu C, Williams PT. Pyrolysis of waste materials using TGA-MS and TGA-FTIR as complementary characterisation techniques[J]. Journal of Analytical & Applied Pyrolysis, 2012, 94.

[15]Cepeliogullar O, Puetuen AE. Thermal and kinetic behaviors of biomass and plastic wastes in co-pyrolysis[J]. Energy Conversion & Management, 2013, 75.

[16]Lago RM, Gomes JPDM, Santos EJM, et al. Pyrolysis of rubber wastes from the mining industry for the production of fuels: pilot scale studies[J]. Polímeros, 2017, 27: 42-47.

[17]Januszewicz K, Kazimierski P, Suchocki T, et al. Waste rubber pyrolysis: Product yields and limonene concentration[J]. Materials, 2020, 13(19): 1-14.

[18]Jing PA, Hong JB, Tai PQ, et al. Transformation and kinetics of chlorine-containing products during pyrolysis of plastic wastes-science direct[J]. Chemosphere, 2021, 284(12).

[19]范韫, 胡连桃, 邓建. 矿井运输胶带热解动力特性的实验研究[J]. 中国矿业大学学报, 1998, 27(3): 3.

[20]赵艳萍, 刘申友, 张和平, 等. 聚氯乙烯电缆绝缘材料热解机理与动力学参数研究[J]. 消防科学与技术, 2012, 31(4): 329-332.

[21]毕松梅, 刘焕章, 谢艳霞, 等. 叠层阻燃输送带用高效弹性体阻燃配方及机理研究[J]. 塑料工业, 2014, 42(5): 77-82.

[22]杨明辉, 陈祎, 刘金和, 等. PVC热解特性及Cl的析出过程研究[J]. 工业加热, 2020, 49(5): 44-47.

[23]卫德俊. 复杂矿井火灾烟气蔓延三维仿真研究[J]. 能源与环保, 2020, 42(8): 55-59.

[24]Komai W, Takehiro I, Shicai J. Study on the combustion characteristics of conveyor belt in mine roadway[J]. Coal Engineer, 1990, 20(2): 45-50.

[25]Perera IE, Litton CD. Impact of air velocity on the detection of fires in conveyor belt haulageways[J]. Fire Technology, 2012, 48(2).

[26]Yuan L, Mainiero RJ, Rowland JH, et al. Numerical and experimental study on flame spread over conveyor belts in a large-scale tunnel[J]. Journal of Loss Prevention in the Process Industries, 2014, 30.

[27]程远平, 季经纬, 李增华. 矿用输送带与木材燃烧性能的对比实验研究[J]. 煤炭学报, 2000, 25(6): 624-627.

[28]周浩生, 周琥, 陆继东. 弱氧条件下橡胶的燃烧机理[J]. 燃烧科学与技术, 2001, (2): 186-190.

[29]王德明, 程远平, 周福宝, 等. 矿井火灾火源燃烧特性的实验研究[J]. 中国矿业大学学报, 2002, 31(1): 30-33.

[30]潘德祥, 王玉怀, 马尚权, 等. 矿用输送带、风筒与坑木燃烧特性的实验研究[J]. 煤炭科学技术, 2005, 33(7): 53-54.

[31]王玉怀, 马尚权, 潘德祥. 煤矿井下可燃材料燃烧特性的试验研究[J]. 华北科技学院学报, 2006, (01): 1-4.

[32]季经纬, 程远平. PVC输送带点燃温度判据及应用[J]. 燃烧科学与技术, 2006, (05): 438-441.

[33]马洪亮, 周心权. 矿井火灾燃烧特性曲线的研究与应用[J]. 煤炭学报, 2008, 33(7): 780-783.

[34]宫传播, 盛睿, 肖尚军. 塑料, 橡胶氧指数试验方法的比较[J]. 中国新技术新产品, 2020, (4): 77-79.

[35]肖国强, 张鹏宇, 马砺, 等. 矿用胶带火灾燃烧特性预测模型[J]. 湖南科技大学学报, 2021, 36(3): 9-15.

[36]Brehob EG, Kim CI, Kulkarni AK. Numerical model of upward flame spread on practical wall materials[J]. Fire Safety Journal, 2001, 36(3).

[37]Consalvi JL, Pizzo Y, Porterie B. Numerical analysis of the heating process in upward flame spread over thick PMMA slabs[J]. Fire Safety Journal, 2008, 43(5).

[38]Linteris G, Gewuerz L, Mcgrattan K. Modeling solid sample burning[J]. Fire Safety Science, 2005, 8.

[39]Wong CK, Dembsey NA, Alston J, et al. A multi-component dataset framework for validation of CFD flame spread models[J]. Journal of Fire Protection Engineering, 2013, 23(2).

[40]Atreya A, Baum HR. A model of opposed-flow flame spread over charring materials[J]. Proceedings of the Combustion Institute, 2002, 29(1).

[41]周延, 王德明, 周福宝. 水平巷道火灾中烟流逆流层长度的实验研究[J]. 中国矿业大学学报, 2001, 30(5): 446-448.

[42]Rowland JH, Smith AC. Flammability of wider conveyor belts using large-scale fire tests[J]. Transactions of Society for Mining, Metallurgy, and Exploration, 2011, 330.

[43]Ying ZL, Bo L, Ingason H. Study of critical velocity and backlayering length in longitudinally ventilated tunnel fires[J]. Fire Safety Journal, 2011, 45(6-8).

[44]Wang HY. Numerical and theoretical evaluations of the propagation of smoke and fire in a full-scale tunnel[J]. Fire Safety Journal, 2012, 49.

[45]黄刚. 煤矿井下胶带火灾烟气流动及分布规律研究[D]. 廊坊: 华北科技学院, 2018.

[46]Edwards JC, Hwang CC. CFD Modeling of Fire Spread Along Combustibles in a Mine Entry[C]. 2006 SME Annual Meeting and Exhibit, Missouri, 2006.

[47]Lowndes IS, Silvester SA, Giddings D, et al. The computational modelling of flame spread along a conveyor belt[J]. Fire Safety Journal, 2007, 42(1).

[48]Hu LH, Fong NK, Yang LZ, et al. Modeling fire-induced smoke spread and carbon monoxide transportation in a long channel: Fire dynamics simulator comparisons with measured data[J]. Journal of Hazardous Materials, 2007, 140(1-2).

[49]Hu LH, Tang F, Yang D, et al. Longitudinal distributions of CO concentration and difference with temperature field in a tunnel fire smoke flow[J]. International Journal of Heat & Mass Transfer, 2010, 53(13-14).

[50]Ali SM, RaghavanV, Rangwala AS. Numerical analysis of flame heating on arbitrarily oriented condensed fuel surfaces[J]. Fire Safety Journal, 2012, 49.

[51]Sojoudi A, Afshin H, Farhanieh B, et al. Large eddy simulation of smoke flow in a real road tunnel fire using FDS[C]. Proceedings of the 4th International Conference on Computational Methods, Australia, 2012.

[52]Brahim K, Mourad B, Afif EC, et al. Control of smoke flow in a tunnel[J]. Journal of Applied Fluid Mechanics, 2013, 6: 49-60.

[53]刘汝正. 细水雾防治胶带火灾的数值模拟和实验研究[D]. 太原: 太原理工大学, 2016.

[54]马砺, 范晶, 任立峰. 地铁站台不同排烟模式下的烟气流动数值模拟[J]. 西安科技大学学报, 2017, 37(5): 643-648.

[55]Jeinsch T, Sader M, Noack R, et al. A robust model-based information system for monitoring and fault detection of large scale belt conveyor systems[C]. World Congress on Intelligent Control & Automation, 2002: 55-59.

[56]Fromme CC, Stager DJ, Pilarski TE, et al. Conveyor belt inspection system and method[Z]. US, 2006.

[57]Fu LQ, Wu XJ. Visual monitoring system of steel-cord conveyor belt[J]. Advanced Materials Research, 2012.

[58]Yang Y, Miao C, Li X, et al. On-line conveyor belts inspection based on machine vision[J]. Optik-International Journal for Light and Electron Optics, 2014, 125(19).

[59]Miao C, Li J. The conveyor belt longitudinal tear on-line detection based on improved SSR algorithm[J]. Journal for Light and Electronoptic, 2016, 127(19): 8002-8010.

[60]Liu W, Qiao T, Yan G, et al. Research on visible light and infrared vision real-time detection system for conveyor belt longitudinal tear[J]. Iet Science Measurement & Technology, 2016, 10(6).

[61]Qiao T, Chen L, Pang Y, et al. Integrative binocular vision detection method based on infrared and visible light fusion for conveyor belts longitudinal tear[J]. Measurement, 2017, 110.

[62]Yu B, Qiao T, Zhang H, et al. Dual band infrared detection method based on mid-infrared and long infrared vision for conveyor belts longitudinal tear[J]. Measurement, 2018, 120: 140-149.

[63]Xianguo L, Lifang S, Zixu M, et al. Laser-based on-line machine vision detection for longitudinal rip of conveyor belt[J]. Optik, 2018, 168: 360-369.

[64]Hou C, Qiao T, Zhang H, et al. Multispectral visual detection method for conveyor belt longitudinal tear[J]. Measurement, 2019, 143: 246-257.

[65]Hou C, Qiao T, Qiao M, et al. Research on audio-visual detection method for conveyor belt longitudinal tear[J]. IEEE Access, 2019, 7.

[66]Yang Y, Hou C, Qiao T, et al. Longitudinal tear early-warning method for conveyor belt based on infrared vision[J]. Measurement, 2019, 147.

[67]Yang R, Qiao T, Pang Y, et al. Infrared spectrum analysis method for detection and early warning of longitudinal tear of mine conveyor belt[J]. Measurement, 2020, 165(12).

[68]黄民, 顾玉华, 魏任之. 矿用胶带输送机火灾监测系统的研究[J]. 煤炭学报, 2002, 27(1): 78-82.

[69]黄民, 魏任之. 矿用钢绳芯带式输送机实时工况监测与故障诊断技术[J]. 煤炭学报, 2005, 30(2): 245-250.

[70]孙继平, 任慧, 吕朝辉. 矿井红外火灾图像处理方法的研究[J]. 煤炭科学技术, 2007, (6): 64-66.

[71]刘庆伟, 徐迎曦, 李传锋, 等. 基于模糊算法的胶带输送机故障诊断系统的研究[J]. 工矿自动化, 2008, (6): 99-102.

[72]邓军, 肖旸, 陈晓坤, 等. 矿井火灾多源信息融合预警方法的研究[J]. 采矿与安全工程学报, 2011, 28(4): 638-643.

[73]李凌云, 牛跃伟, 熊鲲. 带式输送机故障分析与预测系统的研发[J]. 港口装卸, 2012, (5): 40-41.

[74]陈薇, 李鑫. 基于最小二乘支持向量机的大型带式输送机建模[J]. 第三十一届中国控制会议论文集A卷, 2012: 5.

[75]赵肖敏, 孟文俊, 张建水, 等. 基于BP神经网络的带式输送机故障诊断研究[J]. 起重运输机械, 2014, (2): 87-90.

[76]李占利, 赵文博. 基于视觉计算的胶带运输机跑偏监测[J]. 煤矿安全, 2014, 45(5): 118-121.

[77]白宏磊, 张明泽, 王克成. 长距离带式输送机托辊的故障检测[J]. 辽宁科技大学学报, 2015, 38(01): 41-45.

[78]Chen G, Zhang X, Wang ZJ, et al. Robust support vector data description for outlier detection with noise or uncertain data[J]. Knowledge-Based Systems, 2015, 90(9).

[79]马宏伟. 带式输送机托辊红外图像分割与定位算法[J]. 西安科技大学学报, 2017, 37(6): 892-898.

[80]邵晨. 基于噪声信号技术的皮带机托辊故障监测[J]. 仪表技术, 2017, (08): 7-11+46.

[81]刘伟力, 乔铁柱. 矿用输送带纵向撕裂检测系统研究[J]. 工矿自动化, 2017, 43(2): 78-81.

[82]赵敬林. 煤矿带式运输机故障预测系统的研究[D]. 徐州: 中国矿业大学, 2019.

[83]宋新扩. 带式输送机故障分析及其诊断系统设计研究[D]. 徐州: 中国矿业大学, 2019.

[84]贾英新, 靳晔, 王勇, 等. 基于视觉检测带式输送机输送带纠偏系统[J]. 煤矿机械, 2019, 40(9): 62-64.

[85]徐小力. 机电设备故障预警及安全保障技术的发展[J]. 设备管理与维修, 2015, (8): 7-10.

[86]马硕, 焦现炜, 田柯文, 等. 故障预测技术发展与分类[J]. 四川兵工学报, 2013, 34(2): 92-95.

[87]谢世龙. 基于非线性滤波的故障诊断研究与应用[D]. 青岛: 青岛理工大学, 2016.

[88]Wu WJ, He YG, Duan JJ. Application of improved grey model residual modified fusion algorithm in dissolved gas forecasting[J]. The Journal of Engineering, 2019, (23): 9128-9132.

[89]杨阔. 基于预测模型与专家系统的挖掘机液压系统故障诊断技术研究[D]. 吉林: 吉林大学, 2020.

[90]Jian Y, Mei L. Fault diagnosis of excavator hydraulic system based on expert system[J]. Lecture Notes in Electrical Engineering, 2011, 122.

[91]卢传奇. 基于HHT和模糊C均值聚类的液压泵故障诊断研究[D]. 秦皇岛: 燕山大学, 2012.

[92]张一朦. 基于粗糙集与模糊神经网络的滚动轴承故障诊断研究[D]. 大连: 大连交通大学, 2018.

[93]王媛彬. 煤矿烟雾的计算机视觉检测方法[J]. 辽宁工程技术大学学报, 2016, 11: 1230-1234.

[94]骆建彬. 矿冶设备典型零部件寿命预测及维修决策研究[D]. 赣州: 江西理工大学, 2018.

[95]Sun H, Wu J, Yang H. Hybrid SVM and ARIMA model for failure time series prediction based on EEMD[J]. International Journal of Performability Engineering, 2019, 15(4): 1161-1170.

[96]Wang Y, Wang Y, Dang L. Video detection of foreign objects on the surface of belt conveyor underground coal mine based on improved SSD[J]. Journal of Ambient Intelligence and Humanized Computing, 2020: 1-10.

[97]丁永生. 计算智能: 理论、技术与应用[M]. 科学出版社, 2004.

[98]Sunphorka S, Chalermsinsuwan B, Piumsomboon P. Artificial neural network model for the prediction of kinetic parameters of biomass pyrolysis from its constituents-scienceDirect[J]. Fuel, 2017, 193.

[99]Wen X, Jian S, Wang J. Prediction models of calorific value of coal based on wavelet neural networks[J]. Fuel, 2017, 199.

[100]Yang T, Asanjan A, Faridzad M, et al. An enhanced artificial neural network with a shuffled complex evolutionary global optimization with principal component analysis[J]. Information Sciences, 2017: 302-316.

[101]唐彤彤, 杨维, 邵小桃. 基于PSO-GA的Kriging插值法建立透地通信分层地层媒质模型[J]. 煤炭学报, 2016, 41(11): 2913-2920.

[102]Mahmood HA, Adam NM, Sahari BB, et al. Development of a particle swarm optimisation model for estimating the homogeneity of a mixture inside a newly designed CNG-H2-AIR mixer for a dual fuel engine: An experimental and theoretic study[J]. Fuel, 2018, 217.

[103]刘明霞, 游晓明, 刘升. 基于聚度的自适应动态混沌蚁群算法[J]. 计算机工程与应用, 2019, 55(3): 15-22.

[104]于卓熙, 靳雨佳. 基于改进人工鱼群算法的回归模型参数估计[J]. 统计与决策, 2018, 34(22): 75-77.

[105]Eissa MM, Ali AA, Abdel-Latif KM. A frequency control technique based on decision tree concept by managing thermostatically controllable loads at smart grids-ScienceDirect[J]. International Journal of Electrical Power & Energy Systems, 2019, 108.

[106]Protic M, Shamshirband S, Petkovi D. Forecasting of consumers heat load in district heating systems using the support vector machine with a discrete wavelet transform algorithm[J]. Energy, 2015, 87: 343-351.

[107]Han IS, Chung CB. Performance prediction and analysis of a PEM fuel cell operating on pure oxygen using data-driven models: A comparison of artificial neural network and support vector machine[J]. International Journal of Hydrogen Energy, 2016, 41(24).

[108]于繁华, 刘仁云. 计算智能技术及其工程应用[M]. 科学出版社, 2010.

[109]高蕊, 蒋仲安, 董枫. 基于MapObject的矿井火灾动态最佳救灾路线数学模型和算法[J]. 北京科技大学学报, 2008, 30(7): 705-709＋755.

[110]樊荣, 侯媛彬, 张代, 等. 基于神经网络的带式输送机火灾预警算法[J]. 工矿自动化, 2012, 38(9): 70-74.

[111]黄萍, 李兵磊. 基于加权聚合粒子群算法的矿井火灾救援研究[J]. 福州大学学报, 2016, 44(6): 868-873.

[112]Apriori算法在预测矿井火灾事故中的应用[J]. 能源与环保, 2018, 40(08): 78-81.

[113]张美金, 田宇驰, 方志朋. 矿井主运输系统火灾预测的RS-SVM模型[J]. 测控技术, 2018, 37(9): 29-32.

[114]徐劭懿, 李梅, 毛善君, 等. 带约束条件的煤矿火灾避灾路线算法研究[J]. 煤炭科学技术, 2018, 46(5): 173-179+159.

[115]苗续芝, 陈伟, 毕方明, 等. 基于改进FOA-SVM的矿井火灾图像识别[J]. 计算机工程, 2019, 45(4): 267-274.

[116]Hao XL, Liang H. A multi-class support vector machine real-time detection system for surface damage of conveyor belts based on visual saliency[J]. Measurement, 2019, 146: 125-132.

[117]郑凡帆. 基于深度学习的机械设备故障预警与诊断技术研究[D]. 北京: 北京化工大学, 2020.

[118]韩涛, 黄友锐, 张立志, 等. 基于图像识别的带式输送机输煤量和跑偏检测方法[J]. 工矿自动化, 2020, 46(4): 17-22.

[119]张鹏宇. 不同封堵条件下巷道火灾烟气温度预测研究[D]. 西安: 西安科技大学, 2020.

[120]曾宪奎, 宗殿瑞. 国内外输送带技术现状及发展趋势[J]. 特种橡胶制品, 2007, 28(5): 47-52.

[121]安伟, 宋长江, 袁陆海. 煤矿用输送带的选型和使用[J]. 中国橡胶, 2009, (4): 36-39.

[122]刘丁丁. PVC, PVG高强力整芯阻燃输送带及其带芯设计[J]. 当代化工研究, 2019, (4): 172-174.

[123]宋长江. 不同工艺条件下的PVC, PVG阻燃带的质量性能对比分析[C]. 中国橡胶工业协会橡胶工业骨架材料中外技术论坛暨骨架材料专业委员会会员大会, 2007.

[124]丁攀攀. 输送带技术现状及发展趋势[J]. 橡胶科技市场, 2010, (7): 5-9.

[125]Richter F, Rein G. A multiscale model of wood pyrolysis in fire to study the roles of chemistry and heat transfer at the mesoscale[J]. Combustion and Flame, 2020, 216.

[126]Jiang LX, He HH, Sun JJ, et al. Application of genetic algorithm to pyrolysis of typical polymers[J]. Fuel Processing Technology, 2015, 138.

[127]Vyazovkin S, Burnham AK, Criado JM, et al. ICTAC kinetics committee recommendations for performing kinetic computations on thermal analysis data[J]. Thermochimica Acta, 2011, 520(1-2).

[128]White JE, Catallo WJ, Legendre BL. Biomass pyrolysis kinetics: A comparative critical review with relevant agricultural residue case studies[J]. Journal of Analytical & Applied Pyrolysis, 2011, 91(1).

[129]Aboulkas A, Harfi KE, Bouadili AE. Thermal degradation behaviors of polyethylene and polypropylene. Part I: Pyrolysis kinetics and mechanisms[J]. Energy Conversion and Management, 2010, 51(7).

[130]王蔚. 聚氯乙烯电缆火灾特性及其影响因素研究[D]. 合肥: 中国科学技术大学, 2008.

[131]Akahira T, Sunose T. Method of determining activation deterioration constant of electrical insulating materials[J]. Res Rep Chiba Inst Technol (Sci Technol), 1971, 16.

[132]Kissinger HE. Reaction kinetics in differential thermal analysis[J]. Analytical Chemistry, 1957, 29(11).

[133]Flynn JH, Wall LA. A quick, direct method for the determination of activation energy from thermogravimetric data[J]. Journal of Polymer Science: Polymer Letters Edition, 1966, 4(5): 323-328.

[134]Ozawa T. A new method of analyzing thermogravimetric data[J]. Bulletin of the Chemical Society of Japan, 1965, 38(11).

[135]Friedman HL. Kinetics of thermal degradation of char-forming plastics from thermogravimetry. Application to a phenolic plastic[J]. Journal of Polymer Science Polymer Symposia, 2010, 6(1).

[136]Sánchez-Jiménez PE, Pérez-Maqueda LA, Perejón A, et al. Generalized master plots as a straightforward approach for determining the kinetic model: The case of cellulose pyrolysis[J]. Thermochimica Acta, 2013, 552(2).

[137]Coats AW, Redfern JP. Kinetic analysis from thermogravimetric data[J]. Nature, 1964, 68(201).

[138]Criado JM. Kinetic analysis of DTG data from master curves[J]. Thermochimica Acta, 1978, 24(1).

[139]Lesnikovich A, Levchik S. A method of finding invariant values of kinetic parameters[J]. Journal of Thermal Analysis & Calorimetry, 1983, 27(1).

[140]Hurley MJ, Gottuk DT, Hall Jr JR, et al. SFPE handbook of fire protection engineering[M]. Berlin: Springer, 2015.

[141]Quintiere JG. Fundamentals of fire phenomena-火灾学基础[M]. 北京: 化学工业出版社, 2009.

[142]Quintiere JG. A theoretical basis for flammability properties[J]. Fire & Materials, 2010, 30(3).

[143]ISO I. 5660-1: 2015 Reaction-to-fire test-Heat release, smoke production and mass loss rate-Part 1: Heat release rate (cone calorimeter method) and smoke production rate (dynamic measurement), Geneva[M]. Switzerland: International Organization for Standardization, 2015.

[144]ASTM E. 1354-14, Standard test method for heat and visible smoke release rates for materials and products using an oxygen consumption calorimeter[M]. USA: The American Society of Testing Materials, 2017.

[145]Huggett C. Estimation of rate of heat release by means of oxygen consumption measurements[J]. Fire and Materials, 1980, 4(2).

[146]黄松. 低压条件下飞机典型内饰材料燃烧特性研究[D]. 成都: 中国民用航空飞行学院, 2019.

[147]Ignition handbook-principles and applications to fire safety engineering, fire investigation, risk management and forensic science[J]. Fire Safety Journal, 2004, 39(3): 255-256.

[148]曲芳, 孙玉昭, 王志. 典型汽车内饰材料的燃烧特性研究[J]. 消防科学与技术, 2020, 39(6).

[149]Hopkins D, Quintiere JG. Material fire properties and predictions for thermoplastics[J]. Fire Safety Journal, 1996, 26(3).

[150]Schartel B, Hull TR. Development of fire-retarded materials-Interpretation of cone calorimeter data[J]. Fire and Materials, 2007, 31(5).

[151]李梦、邓帮林、黄元培. 汽车座椅材料的燃烧特性研究[J]. 消防科学与技术, 2020, 39(7).

[152]曲芳, 吴鹏, 王志. 环氧树脂复合材料的热解特性及动力学研究[J]. 消防科学与技术, 2018, 37(11).

[153]Petrella RV. The assessment of full-scale fire hazards from cone calorimeter data[J]. Journal of Fire Sciences, 1994, 12(1).

[154]马丽. 4种铺地材料的燃烧及产烟特性研究[J]. 中国新技术新产品, 2021, (15).

[155]王勇, 胡晓, 祈闻, 等. 阻燃PC-ABS燃烧性能的锥形量热仪研究[J]. 云南大学学报, 2005, (3): 484-487.

[156]王甘雨. 纵向通风条件下大跨度隧道火灾烟气多维运动特性研究[D]. 杭州: 浙江大学, 2022.

[157]Hubel DH, Wiesel TN. Receptive fields binocular interaction and functional architecture in the cat's visual cortex[J]. Journal of Physiology, 1962, 160(1): 106-154.

[158]Fukushima K. Neocognitron: a self organizing neural network model for a mechanism of pattern recognition unaffected by shift in position[J]. Biological Cybernetics, 1980, 36(4): 193-202.

[159]LeCun Y, Bottou L, Bengio Y, et al. Gradient based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11): 2278-2324.

[160]Krizhevsky A, Sutskever II, Hinton G. Imagenet classification with deep convolutional neural networks[C]. Proceedings of the Advances in Neural Information Processing Systems Lake Tahoe, USA, 2012: 1097-1105.

[161]贾瑶. 基于时空特征分析的网约车短时需求预测研究[D]. 北京: 北京交通大学, 2019.

[162]Zeiler MD, Fergus R. Visualizing and understanding convolutional networks[C]. In: Fleet D, Pajdla T, Schiele B, Tuytelaars T (eds) Computer Vision-ECCV 2014 ECCV 2014 Lecture NNotes in Computer Science, 2014: 8689+8818-8833.

[163]Charu C, Aggarwal IBM T.J. Convolutional neural networks[M]. In: Neural Networks and Deep Learning Springer, Cham Research Center Watson, 2018: 315-371.

[164]王彦卷. 基于CNN-LSTM和气象要素关联的气温时空预测[D]. 银川: 宁夏大学, 2022.

[165]Hochreiter S, Schmidhuber J. Long short-term memory[J]. Neural Computation, 1997, 9(8).

[166]Langkvist M, Karlsson L, Loutfi A, et al. Areview of unsupervised feature learning and deep leaming for time-series modeling[J]. Pattern Recognition Letters, 2014, 42(1): 11-24.

[167]Graves A, Jaitly N, Mohamed A. Hybrid speech recognition with deep bidirectional LSTM[C]. Automatic Speech Recognition and Understanding Olomouc: IEEE, 2013: 273-278.

[168]胡家伟. 基于深度学习的温度场预测方法及其在惯导平台中的应用[D]. 长沙: 湖南大学, 2021.

[169]宋耀宇. 基于深度神经网络的大气污染物分布预测[D]. 郑州市: 华北水利水电大学, 2020.

[170]程万伟. 时间序列缺失值插补方法研究[D]. 长沙: 湖南大学, 2018.

[171]Cover T, Hart P. Nearest neighbor pattern classification[J]. IEEE Transactions on Information Theory, 2003, 13(1).