煤化工作为煤炭清洁高效利用的重要途径，近年来在我国得到快速发展，但化工重特大事故频发势头尚未得到有效遏制，造成了严重的经济损失和恶劣的社会影响。煤化工生产过程具有规模大、流程长、工况复杂、高温高压、易燃易爆、有毒有害等特点，相关安全管理方法与技术发展缓慢，在实际生产中需要借助传统石化安全管理方法，开发适用于煤化工生产特征的安全管理方法与技术有着重大的现实需求。本文针对煤化工生产过程风险耦合致灾机理、综合风险量化评价、工艺参数预测、动态风险预警等关键问题进行了深入研究，主要取得的研究成果如下：

（1）对我国2008-2022年发生的煤化工安全事故特征规律进行了统计分析，正常生产、检维修作业、受限空间作业和动火作业过程中发生事故较多，占比分别为51%、18%、9%、9%；人、机、物、环、管5类风险因素引发事故的综合占比分别为74%、52%、83%、38%、91%；重特大事故均是由爆炸造成的，且导致的死亡率最高，为8.8人/次。采用文本挖掘技术对事故报告中的事故原因进行文本聚类分析，探究了不同等级、不同类型事故的风险因素耦合关系，重特大事故主要由机械风险因素和物料风险因素耦合作用导致，虽然人员风险因素引发的事故较多，但造成的危害程度相对较小。构建N-K耦合模型定量计算了单因素、双因素、多因素间的正向耦合概率和耦合强度，风险值随耦合因素的增多整体呈上升趋势；机-物、人-机耦合风险值大于部分多因素耦合风险值，机、物与其他因素耦合后会产生较大风险。上述研究揭示煤化工风险因素耦合致灾机理，为后续研究提供了理论基础和数据支撑。

（2）结合国家法规标准和煤化工风险耦合致灾机理，从人、机、物、环、管全维度识别煤化工过程风险因素，通过增加表征风险作为惩罚性指标，构建了包含6个一级风险因素、13个二级风险因素、61个量化指标的风险评价指标体系，为煤化工综合风险评价和预警提供了量化手段。提出NK-SEWM综合赋权法对指标体系进行赋权，结合物元可拓理论和云理论建立可拓云综合风险评价模型，用云关联度函数代替可拓物元理论的固定关联度函数，有效解决了风险评价过程中多指标不相容问题和指标量化、风险等级边界划分造成的模糊性及随机性问题。

（3）根据工艺参数的时序变化特征，建立双向长短期记忆网络（BiLSTM）模型，采用鲸鱼优化算法（WOA）对模型超参数进行自动寻优，并引入修正型转置正态损失函数（MINLF）计算参数偏差造成的预期损失，运用剩余时间理论计算工艺参数异常波动的发生概率，最终构建了包含固有风险和趋势风险的WOA-BiLSTM-MINLF工艺参数预测方法，实现了对工艺参数变化趋势的长期预测，同时将参数波动过程转换为量化的动态风险值，解决了提前预测时间导致预测精度下降的技术难题。结果证明，该预测方法比DCS提前16 min报警，可为操作人员留出充裕的时间采取安全防控措施。

（4）选取具有时效性的煤化工动态风险预警指标，建立最小二乘支持向量机（LSSVM）多分类动态风险预警模型，采用鲸鱼优化算法（WOA）优化模型中核函数和惩罚因子两个超参数，显著提升预警模型的分类性能。通过构建预警模型分类效果的混淆矩阵发现，WOA-LSSVM预警模型的分类准确率为95.83%，误判率为4.17%，漏判率为0，表明该模型能够根据风险预警指标状态准确地预警生产风险等级，为煤化工风险管控提供了一种智能化、信息化的技术手段。

（5）对我国西部一家煤化工企业进行了实证研究。评价结果表明企业整体生产风险为3级中等风险，应依据结果分析对局部风险较大的环节进行及时整改。根据动态风险预警指标值，判断出企业动态风险预警等级为中警，与综合风险评价结果一致。研究结果与企业实际运行情况吻合较好，说明本文构建的安全风险评价方法和动态风险预警模型具有较好的可行性和有效性。

As an important avenue for the clean and efficient utilization of coal, the coal chemical industry has experienced rapid development in China in recent years. However, the frequency occurrence of major chemical accidents has not been effectively curbed, resulting in severe economic losses and adverse social impacts. The coal chemical production process is characterized by large scale, long processes, complex conditions, high temperature and pressure, flammability, explosiveness, and toxicity. The development of related safety management methods and technologies is slow. In practical production, it is need to use traditional petrochemical safety management methods and develop safety management methods and technologies suitable for the characteristics of coal chemical production, which is of great practical need. This paper conducts in-depth research on key issues such as risk coupling disaster-causing mechanisms, comprehensive risk quantification evaluation, process parameter prediction, and dynamic risk early warning in coal chemical production. The main research achievements are as follows:

(1) A statistical analysis was conducted on the characteristics of safety accidents in the coal chemical industry in China from 2008 to 2022. Accidents occurred frequently during normal production, inspection and maintenance operations, confined space operations, and hot work operations, accounting for 51%, 18%, 9%, and 9% of incidents, respectively. The combined proportions of accidents caused by five risk factors—human, machine, material, environment, and management—were 74%, 52%, 83%, 38%, and 91%, respectively. Major catastrophic accidents were all caused by explosions, which resulted in the highest fatality rate, at 8.8 deaths per incident. Text mining technology was used to perform text clustering analysis on the causes of accidents in accident reports, exploring the coupling relationships of risk factors in different levels and types of accidents. Major accidents are mainly caused by the coupling of mechanical and material risk factors. Although human risk factors cause many accidents, the harm caused is relatively minor. An N-K coupling model was constructed to quantitatively calculate the forward coupling probability and coupling strength between single, dual, and multiple factors, with risk values increasing with more coupling factors; machine-material and human-machine coupling risks are higher than some multi-factor coupling risks, and coupling with machine and material factors produces greater risks. This research reveals the disaster-causing mechanism of risk factor coupling in the coal chemical industry, providing a theoretical basis and data support for subsequent research.

(2) Combining national regulations and standards and the risk coupling disaster-causing mechanisms of coal chemical processes, risk factors in the coal chemical process were identified from all dimensions of human, machine, material, environment, and management. By adding explicit risk as a punitive indicator, a risk assessment indicator system comprising 6 primary risk factors, 13 secondary risk factors, and 61 quantitative indicators was constructed, providing a quantitative means for comprehensive risk assessment and early warning in the coal chemical industry. An NK-SEWM comprehensive weighting method was proposed to weight the indicator system, and a extension cloud model was established by combining matter-element extension theory and cloud theory, using the cloud correlation function to replace the fixed correlation function of matter-element theory, effectively solving the multi-index incompatibility issues in the risk assessment process and the fuzziness and randomness problems caused by indicator quantification and risk level boundary division.

(3) Based on the temporal variation characteristics of process parameters, a Bidirectional Long Short-Term Memory (BiLSTM) model was established, using the Whale Optimization Algorithm (WOA) to automatically optimize the model's hyperparameters, and introducing the Modified Inverted Normal Loss Function (MINLF) to calculate the expected loss caused by parameter deviations. The residual time theory was used to calculate the probability of abnormal fluctuations in process parameters, ultimately constructing a WOA-BiLSTM-MINLF process parameter prediction method that includes inherent and trend risks, achieving long-term prediction of process parameter trends, and converting parameter fluctuation processes into quantified dynamic risk values, solving the technical difficulty of decreased prediction accuracy due to early prediction timing. The results prove that the prediction method provides alarms 16 minutes earlier than the DCS, allowing operators ample time to take safety precautions.

(4) Dynamic risk early warning indicators with timeliness for the coal chemical industry were selected, and a Least Squares Support Vector Machine (LSSVM) multi-class dynamic risk early warning model was established, using the Whale Optimization Algorithm (WOA) to optimize the two hyperparameters of the kernel function and penalty factor in the model, significantly enhancing the classification performance of the early warning model. Through constructing the classification effect confusion matrix of the early warning model, it was found that the WOA-LSSVM early warning model has a classification accuracy of 95.83%, a misjudgment rate of 4.17%, and a missed detection rate of 0, indicating that the model can accurately warn of production risk levels based on the status of risk early warning indicators, providing an intelligent and informatized technical means for risk control in the coal chemical industry.

(5) An empirical study was conducted on a coal chemical enterprise in western China. The evaluation results show that the overall production risk of the enterprise is at a medium risk level of 3, and based on the analysis of the results, timely rectification should be carried out on parts with higher local risks. According to the dynamic risk early warning indicator values, the dynamic risk early warning level of the enterprise was judged to be medium alert, consistent with the comprehensive risk assessment results. The research results are in good agreement with the actual operation of the enterprise, indicating that the safety risk assessment method and dynamic risk early warning model constructed in this paper have good feasibility and effectiveness.

[1]Wang Y, Gu A, Zhang A. Recent development of energy supply and demand in China, and energy sector prospects through 2030[J]. Energy Policy, 2011, 39(11): 6745–6759.

[2]Fan Y, Xia Y. Exploring energy consumption and demand in China[J]. Energy, 2012, 40(1): 23–30.

[3]Xie KC, Li WY, Zhao W. Coal chemical industry and its sustainable development in China[J]. Energy, 2010, 35: 4349–4355.

[4]Xie KC. Breakthrough and innovative clean and efficient coal conversion technology from a chemical engineering perspective[J]. Chemical Engineering Science: X, 2021, 10: 100092.

[5]陈艾, 焦洪桥, 王秀江, 等. 新型煤化工产业生态化发展的政策分析、技术路径研究[J]. 煤化工, 2023, 51(3): 1–5.

[6]赵一玮, 李冬, 陈楠, 等. 高质量发展要求下黄河中上游煤化工产业环境管理建议[J]. 中国环境管理, 2020, 12(06): 52–57.

[7]Yang QL. Comparison of coal mine safety management in China and the United States and its inspiration to China’s coal mine safety management[J]. Technology and Innovation Management, 2016, 04: 365–370.

[8]Chen C, Reniers G. Chemical industry in China: The current status, safety problems, and pathways for future sustainable development[J]. Safety Science, 2020, 128: 104741.

[9]谢克昌. 加强关键技术创新, 推进煤化工高端化[C]. 2019第八届中国国际煤化工发展论坛. 北京: 中国石油和化学工业联合会, 2019.

[10]胡迁林, 赵明. “十四五”时期现代煤化工发展思考[J]. 中国煤炭, 2021, 47(3): 2–8.

[11]杨红琳. 当代化工安全技术智能化发展趋势思考[J]. 中国安全科学学报, 2022, 32(1): 213–214.

[12]韩红梅, 朱彬彬, 龚华俊, 等. 现代煤化工行业“十三五”回顾和“十四五”发展展望[J]. 化学工业, 2021, 39(1): 17–20.

[13]承奇, 郭燕秋, 张礼敬, 等. 石化企业危险化工工艺风险等级评估指标体系研究[J]. 中国安全生产科学技术, 2011, 07(10): 89–92.

[14]徐振刚. 中国现代煤化工近25年发展回顾·反思·展望[J]. 煤炭科学技术, 2020, 48(8): 1–25.

[15]唐宏青. 现代煤化工新技术[M]. 北京: 化学工业出版社, 2009.

[16]Yu K, Zhou LJ, Hu C, et al. Analysis of Influencing Factors of Occupational Safety and Health in Coal Chemical Enterprises Based on the Analytic Network Process and System Dynamics[J]. Processes, 2019, 7(1): 53–53.

[17]胡瑾秋, 张立强, 张来斌. 石油化工装置长周期运行风险超早期精确预警方法[J]. 石油学报(石油加工), 2019, 35(3): 527–533.

[18]Wu X, Chen G. Global primary energy use associated with production, consumption and international trade[J]. Energy Policy, 2017, 111: 85–94.

[19]Abotah R, Daim TU. Towards building a multi perspective policy development framework for transition into renewable energy[J]. Sustainable Energy Technologies and Assessments, 2017, 21: 67–88.

[20]Yang J, Wu J, He T, et al. Energy gases and related carbon emissions in China[J]. Resources, Conservation and Recycling, 2016, 113: 140–148.

[21]Wei XF, Wang JG, Ding YJ. Progress and Development Trend of Clean and Efficient Coal Utilization Technology[J]. Bulletin of Chinese Academy of Sciences, 2019, 34: 409–416.

[22]Li J, Hu SY. History and future of the coal and coal chemical industry in China[J]. Resources, Conservation and Recycling, 2017, 124: 3–24.

[23]新浪财经. 2023年中国煤化工行业企业大数据全景图谱[EB/OL]. [2022–11–29]. https://finance.sina.cn/2022-11-29/detail-imqmmthc6395968.d.html

[24]张胜利, 焦洪桥, 杨靖华, 等. 碳中和背景下现代煤化工产业生态链布局和创新发展路径[J]. 中国煤炭, 2022, 48(8): 7–13.

[25]中国煤炭工业协会. 煤炭工业“十四五”现代煤化工发展指导意见 [EB/OL]. [2021–06–09]. http://www.coalchina.org.cn/

[26]高珩. 浅析煤化工的产业现状及发展前景[J]. 当代化工研究, 2020, 17: 10–11.

[27]周亚飞, 刘茂. 化工园区重大事故风险分析[J]. 防灾减灾工程学报, 2011, 31(1): 68–74.

[28]许铭, 多英全, 吴宗之. 化工园区安全规划发展历史回顾[J]. 中国安全科学学报, 2008, 18(8): 140–149.

[29]中国标准化研究院. 企业安全生产标准化基本规范: GB/T 33000–2016[S]. 北京: 中国标准出版社, 2016.

[30]罗云. 风险分析与安全评价[M]. 化学工业出版社, 2004.

[31]Khan F, Rathnayaka S, Ahmed S. Methods and models in process safety and risk management: Past, present and future[J]. Process safety and environmental protection, 2015, 98: 116–147.

[32]孙亚菲, 王永刚, 杨建森, 等. 民航空管系统运行质量监督系统的分析与设计[J]. 计算机应用与软件, 2015, 32(4): 104–108.

[33]Khan FI, Abbasi SA. Techniques and methodologies for risk analysis in chemical process industries[J]. Construction and Building Materials, 1998, 11(4): 261–277.

[34]张双甫, 周旻. 定性和定量评估在航行安全风险分析中的应用[J]. 大连海事大学学报, 2010, 36(S1): 1–3.

[35]张其国, 张中俭, 丁继新. 基于改进的安全检查表法的尾矿库安全现状定量评价[J]. 安全与环境工程, 2018, 25(05): 121–126.

[36]姚桂莹. HAZOP分析技术在催化重整装置的应用[J]. 石油化工安全环保技术, 2014, 30(01): 17–22.

[37]Song YL, Bai GL, Zhou W, et al. Application of hazard and operability study in liquefied natural gas receiving terminal[J]. Chemical Engineering (China), 2012, 40(2): 74–78.

[38]王喜梅, 陈建宏, 杨珊. 基于粒计算的尾矿库安全评价不确定性研究[J]. 中国安全生产科学技术, 2018, 14(01): 42–48.

[39]Cozzani V, Salzano E. The quantitative assessment of domino effect caused by overpressure: Part II. Case studies[J]. Journal of Hazardous Materials, 2004, 107(3): 81–94.

[40]宁寻安, 周剑波, 李仕文, 等. 环氧氯丙烷生产装置的火灾爆炸危险性分析与评价[J]. 广东工业大学学报, 2018, 28(2): 1–5.

[41]秦华礼, 桂焱. 蒙德法评价方法的计算机软件开发及应用[J]. 化肥设计, 2016, 54(04): 33–36.

[42]Sarbayev M, Yang M, Wang H. Risk assessment of process systems by mapping fault tree into artificial neural network[J]. Journal of Loss Prevention in the Process Industries, 2019, 60: 203–212.

[43]都心爽, 武玮, 张金阳, 等. 基于三角直觉模糊故障树的工业管道安全评价方法与应用[J]. 西北大学学报(自然科学版), 2021, 51(4): 632–642.

[44]姚成玉, 王传路, 陈东宁, 等. 连续时间T–S动态故障树分析方法[J]. 机械工程学报, 2020, 56(10): 244–256.

[45]Keller W, Modarres M. A historical overview of probabilistic risk assessment development and its use in the nuclear power industry: a tribute to the late Professor Norman Carl Rasmussen[J]. Reliability Engineering & System Safety, 2005, 89(3): 271–285.

[46]Karasan A, Ilbahar E, Celi S, et al. A new risk assessment approach: Safety and Critical Effect Analysis (SCEA) and its extension with Pythagorean fuzzy sets[J]. Safety Science, 2018, 108: 173–187.

[47]Tian Z, Chen Q, Zhang T. A method for assessing the crossed risk of construction safety[J].Safety Science, 2022, 146: 105531.

[48]程玉龙, 罗云, 师立晨, 等. 化工园区Na–Tech风险定量评价方法[J]. 化工进展, 2019, 38(S1): 283–287.

[49]潘李冬, 郑宇, 郑娟, 等. 基于 BN–FCEM 的化工企业安全风险分析方法[J]. 安全与环境工程, 2023, 30(5): 1–10+18.

[50]刘庆龙, 曲秋影, 赵东风, 等. 基于多源异构数据融合的化工安全风险动态量化评估方法[J]. 化工学报, 2021, 72(3): 1769–1777.

[51]贾梅生, 陈国华, 胡昆. 化工园区多米诺事故风险评价与防控技术综述[J]. 化工进展, 2017, 36(04): 1534–1543.

[52]Zavari M, Shahhosseini V, Ardeshir A. Multi-objective optimization of dynamic construction site layout using BIM and GIS[J]. Journal of Building Engineering, 2022, 52: 104518.

[53]Kolar Z, Chen H, Luo X. Transfer learning and deep convolutional neural networks forsafety guardrail detection in 2D images[J]. Automation in construction, 2018, 89: 58–70.

[54]Wang ZZ, Chen C. Fuzzy comprehensive Bayesian network-based safety risk assessment for metro construction projects[J]. Tunnelling and Underground Space Technology, 2017, 70: 330–342.

[55]王福生, 高湛翔, 董宪伟, 等. 基于物元可拓理论的化工企业风险评价[J]. 安全与环境学报, 2021, 21(06): 2401–2406.

[56]Roessler J, Cheng W, Hayes JB, et al. Evaluation of the leaching risk posed by the beneficial use of ammoniated coal fly ash[J]. Fuel, 2016, 184: 613–619.

[57]王璐. 煤制烯烃工程火灾风险探讨[J]. 现代化工, 2014, 34(4): 14–16.

[58]谭新蕊. 基于火灾风险评估的煤化工企业火灾保险研究[D]. 沈阳: 沈阳航空航天大学, 2020.

[59]秦博. 煤化工有害气体安全监控物联网系统的设计与实现[D]. 成都: 电子科技大学, 2016.

[60]冉丽君, 梁鹏, 罗霂, 等. 我国现代煤化工面临的环保困境及对策建议[J]. 环境保护, 2017, 45(01): 39–41.

[61]Yao M, Fang YX, Tang WL, Zhou JJ. Study on safety behavior planning theory and control strategies for coal chemical workers[J]. Safety Science, 2020, 128: 104726.

[62]许雅, 庞集华. 某煤化工企业煤气化装置火灾爆炸事故模拟[J]. 消防科学与技术, 2022, 41(9): 1247–1250.

[63]刘兴辉, 刘旭. 现代煤化工装置隐患排查治理实施[J]. 当代化工研究, 2022, 06: 132–134.

[64]秦洪浪. 煤化工机电设备在线振动故障检测系统智能化研究[J]. 工业加热, 2021, 50(6): 62–65.

[65]Zhang SH, Dai ZH, Xu JL, et al. Sensitivity and safety boundary Analysis of Opposed Multi–Burner Coal Water Slurry Gasification System[J]. International Journal of Energy Research, 2020, 44(3): 2278–2288.

[66]齐东川. 基于支持向量机的水煤浆气化工艺安全评价模型及系统设计[D]. 重庆: 重庆科技学院, 2021.

[67]张丽婷. 基于HAZOP–FTA的煤化工生产过程风险管控研究[D]. 天津: 天津理工大学, 2014.

[68]孙超, 王旭, 王丹丹. 煤化工行业风险管理对策研究[J]. 煤矿安全, 2018, 49(8): 282–284.

[69]乔元, 王建立. 现代煤制油化工安全管理体系的探索与实践[J]. 中国安全科学学报, 2021, 31(S1): 34–38.

[70]Zhang L, Sun Z, Dai S, et al. Safety risk evaluation of China coal chemical industry on the basis of fuzzy mathematics theory[C]. The 2nd International Conference on Computing and Data Science, 2021: 1–6.

[71]徐晓东. 基于BP神经网络的煤化工园区安全预警机制研究[J]. 中国煤炭, 2014, 40(11): 13–17+39.

[72]闫星月. 基于结构方程模型的化工园区风险因素分析[D]. 天津: 天津理工大学, 2022.

[73]Khakzad N, Kha F, Amyotte P. Dynamic risk analysis using bow–tie approach[J]. Reliability Engineering & System Safety, 2012, 104: 36–44.

[74]罗云, 黄西菲, 许铭. 安全生产科学管理的发展与趋势探讨[J]. 中国安全生产科学技术, 2016, 12(10): 5–11.

[75]张道斌. 企业安全生产预警预报机制建设研究[J]. 中国安全科学学报, 2013, 23(09): 154–158.

[76]邱丕群. 高校投入产出分析与预警系统研究[J]. 统计与信息论坛, 1998, (03): 16–20.

[77]裴兴旺, 李文龙, 刘怡君. 钢筋混凝土排架结构厂房再生利用施工安全风险预警[J/OL]. 安全与环境学报, 1–11[2024–04–03]. https://doi.org/10.13637/j.issn.1009-6094. 2023.1470.

[78]李爽, 李丁炜, 犹梦洁, 等. 基于BN–ELM的煤矿瓦斯安全态势预测方法[J]. 系统工程, 2020, 38(03): 132–140.

[79]He R, Li X, Chen G, et al. Generative adversarial network–based semi–supervised learning for real–time risk warning of process industries[J]. Expert Systems with Applications, 2020, 150: 113244.

[80]Tian W, Wang S, Sun S, e al. Intelligent prediction and early warning of abnormal conditions for fluid catalytic cracking process[J]. Chemical Engineering Research and Design, 2022, 181: 304–320.

[81]张良. 基于系统风险熵的化工园区风险态势预测预警研究[D]. 广州: 华南理工大学, 2016

[82]王春利. 石化装置异常工况监测预警与操作指导系统研发与应用[J]. 中国安全生产, 2015, 2: 58–59.

[83]李兴华, 王媛媛, 袁子龙. 基于新一代信息技术的化工园区气体泄漏监测预警系统[J]. 中国安全生产科学技术, 2021, 17(S1): 10–14.

[84]Chen A, Jiao HQ, Wang XJ, et al. Policy analysis and technical path research on the ecological development of the new coal chemical industry[J]. Coal Chemical Industry, 2023, 51(3): 1–5.

[85]Tian WD, Hu MG, Li CK. Fault prediction based on dynamic model and grey time series model in chemical processes[J]. Chinese Journal of Chemical Engineering, 2014, 22(6): 643–650.

[86]Amin MT, Khan F, Ahmed S, et al. A data–driven Bayesian network learning method for process fault diagnosis[J]. Process Safety and Environmental Protection, 2021, 150: 110–122.

[87]Yuan XF, Ou C, Wang YL, et al. A novel semi–supervised pre–training strategy for deep networks and its application for quality variable prediction in industrial processes[J]. Chemical Engineering Science, 2020, 217: 115509.

[88]Zhong K, Han M, Han B. Data–Driven Based Fault Prognosis for Industrial Systems: A Concise Overview[J]. IEEE–CAA Journal of Automatica Sinica, 2020, 7: 330–345.

[89]Yao K. Bayesian inference with uncertain data of imprecise observations[J]. Communications in Statistics–Theory and Methods, 2022, 51(15): 5330–5341.

[90]Qi F, Zhang L, Zhou K, et al. Early Waning for Manufacturing Supply Chain Resilience Based on Improved Grey Prediction Model[J]. Sustainability, 2022, 14(20): 215–226.

[91]Abid A, Khan MT, Iqbal J. A review on fault detection and diagnosis techniques: basics and beyond[J]. Artificial Intelligence Review, 2021, 54 (5): 3639–3664.

[92]Yin Z, Hou J. Recent advances on SVM based fault diagnosis and process monitoring in complicated industrial processes[J]. Neurocomputing, 2016, 174: 643–650.

[93]Liu Y, Yang C, Gao ZL, et al. Ensemble deep kernel learning with application to quality prediction in industrial polymerization processes[J]. Chemometrics and Intelligent Laboratory Systems, 2018, 174: 15–21.

[94]Liu LS, Chen LQ, Wang ZL, et al. Early fault detection of planetary gearbox based on acoustic emission and improved variational mode decomposition[J]. IEEE Sensors Journal, 2021, 21 (2): 1735–1745.

[95]Yao L, Ge ZQ. Deep learning of semisupervised process data with hierarchical extreme learning machine and soft sensor application[J]. IEEE Transactions on Industrial Electronics, 2018, 65(2): 1490–1498.

[96]Dorgo G, Palazoglu A, Abonyi J. Decision trees for informative process alarm definition and alarm–based fault classification[J]. Process Safety and Environmental Protection, 2021, 149: 312–324.

[97]Mozaffari S, Javadi S, Moghaddam HK, et al. Forecasting groundwater levels using a hybrid of support vector regression and particle swarm optimization[J]. Water Resources Management, 2022, 36(6): 1955–1972.

[98]Xu C, Chen X, Zhang L. Predicting river dissolved oxygen time series based on stand–alone models and hybrid wavelet–based models[J]. Journal of Environmental Management, 2021, 295: 113085.

[99]Xu J, Zhao Y, Chen H, et al. ABC–GSPBFT: PBFT with grouping score mechanism and optimized consensus process for flight operation data–sharing[J]. Informing Science, 2023, 624: 110–127.

[100]Yuan XF, Li L, Shardt YAW, et al. Deep learning with spatiotemporal attention–based LSTM for industrial soft sensor model development[J]. IEEE Transactions on Industrial Electronics, 2021, 68(5): 4404–4414.

[101]Yang W, Liu W, Gao Q. Prediction of dissolved oxygen concentration in aquaculture based on attention mechanism and combined neural network[J]. Mathematical Biosciences and Engineering, 2023, 20(1): 998–1017.

[102]陈樑, 朱君烨, 金龙, 等. 时序数据驱动的化工过程风险动态预警研究[J]. 安全与环境学报, 2023, 23(10): 3491–3501.

[103]Ma M, Mao Z. Deep–convolution–based LSTM network for remaining useful life prediction[J]. IEEE Transactions on Industrial Informatics, 2021, 17(3): 1658–1667.

[104]Kumari K, Dey P, Kumar C, et al. UMAP and LSTM based fire status and explosibility prediction for sealed–off area in underground coal mine[J]. Process Safety and Environmental Protection, 2021, 146: 837–852.

[105]Quddus N, Yu M, Tamim N, et al. Risk assessment of class 3 (PG II & III) hazardous materials in transportation[J]. Process Safety Progress, 2018, 37(3): 376–381.

[106]Halim SZ, Yu M, Escobar H, et al. Towards a causal model from pipeline incident data analysis[J]. Process Safety and Environmental Protection, 2020, 143: 348–360.

[107]Liew WT, Adhitya A, Srinivasan R. Sustainability trends in the process industries: A text mining–based analysis[J]. Computers in Industry, 2014, 65(3): 393–400.

[108]Hatipkarasulu Y. Project level analysis of special trade contractor fatalities using accident investigation reports[J]. Journal of safety research, 2010, 41(5): 451–457.

[109]Brown DE. Text mining the contributors to rail accidents[J]. IEEE Transactions on Intelligent Transportation Systems, 2015, 17(2): 346–355.

[110]中国标准化委员会. 企业职工伤亡事故分类: GB/T 6441–1986[S]. 北京:中国标准出版社, 1986.

[111]中华人民共和国国务院. 生产安全事故报告和调查处理条例[EB/OL]. [2007–04–19]. http://www.gov.cn/ziliao/flfg/2007-04/19/content_589264.htm

[112]中国标准化委员会. 生产过程危险和有害因素分类与代码: GB/T 13861–2022[S]. 北京: 中国标准出版社, 2022.

[113]张璐, 许开立, 葛及, 等. 基于DEMATEL–ANP–可拓云模型的铜冶炼企业安全生产风险评价[J]. 安全与环境工程, 2023, 30(01): 1–8.

[114]中国标准化委员会. 企业职工伤亡事故调查分析规则: GB/T 6442–1986[S]. 北京:中国标准出版社, 1986.

[115]Cheng CW, Leu SS, Cheng YM, et al. Applying data mining techniques to explore factors contributing to occupational injuries in Taiwan’s construction industry[J]. Accident Analysis & Prevention, 2012, 48: 214–222.

[116]Allahyari M, Pouriyeh S, Assefi M, et al. A brief survey of text mining:classification, clustering and extraction techniques[J]. arXiv preprint arXiv, 2017, 1707.02919v2.

[117]刘堂卿, 罗帆. 空中交通安全风险构成及耦合关系分析[J]. 武汉理工大学学报(信息与管理工程版), 2012, 34(01): 93–97.

[118]李新春, 韩继磊, 乔万贯. 煤矿事故风险因子耦合机理研究[J]. 煤矿安全, 2015, 46(08): 240–242.

[119]Frenken K. A complexity approach to innovation networks. The case of the aircraft industry (1909–1997)[J]. Research Policy, 2000, 29(2): 257–272.

[120]吴贤国, 吴克宝, 沈梅芳, 等. 基于N–K模型的地铁施工安全风险耦合研究[J]. 中国安全科学学报, 2016, 26(04): 96–101.

[121]卢春雪, 陈杰, 张涛. 关于焦化企业所涉危险工艺的分析[J]. 工业安全与环保, 2016, 42(3): 51–52.

[122]Sun C, Wang X, Wang DD. Risk Management Countermeasures of Coal Chemical Industry[J]. Safety in Coal Mines, 2018, 49: 282–284.

[123]孙越. 煤尘爆炸性的实验研究[J]. 煤炭与化工, 2017, 7: 62–63.

[124]Tulashie SK, Addai EK, Annan JS. Exposure assessment, a preventive process in managing workplace safety and health, challenges in Ghana[J]. Safety science, 2016, 84(3): 210–215.

[125]Jahangiri M, Hoboubi N, Rostamabadi A, et al. Human Error Analysis in a Permit to Work System: A Case Study in a Chemical Plant[J]. Safety and Health at Work, 2016, 7(1): 6–11.

[126]Su T, Lin P, Shu Y, et al. Analysis of the Multi–Relationships and Their Structures for Safety Culture[J]. Open Journal of Safety Science and Technology, 2012, 2(3): 89–97.

[127]桂强. 新时期化工企业的安全生产与管理分析[J]. 科技创新与应用, 2017, 2: 268–269.

[128]Petitta L, Probst TM, Barbaranelli, C. Disentangling the roles of climate and safety culture: Multi–level effects on the relationship between supervisor enforcement and safety compliance [J]. Accident Analysis & Prevention, 2017, 99: 77–89.

[129]陈学东, 范志超 ,陈永东, 等. 我国高端压力容器设计制造与维护技术进展[J]. 机械工程学报, 2023, 59(20): 18–33.

[130]邵爱新, 孙宁, 张晓曦, 等. 基于危险化学品重大危险源辨识分级的大型油储库外部安全防护距离研究[J]. 南开大学学报(自然科学版), 2021, 54(06): 81–89.

[131]中国标准化委员会. 危险货物分类和品名编号: GB/T 6944–2012[S]. 北京:中国标准出版社, 2012.

[132]张任栋, 张礼敬, 陶刚. 模糊推理–组合赋权的化工工艺本质安全度研究[J]. 中国安全科学学报, 2023, 33(08): 149–155.

[133]严朝阳, 孙娇, 李鑫. 储运系统设备及管道布置的优化[J]. 化工设计通讯, 2020, 46(4): 116+124.

[134]王丹丹. 煤化工行业风险分析及保险研究[D]. 沈阳: 沈阳航空航天大学, 2017.

[135]中华人民共和国住房与城乡建设部. 建筑设计防火规范: GB/T 50016–2014[S]. 北京:中国计划出版社, 2014.

[136]Goh YM, Love PED, Stagbouer G, et al. Dynamics of safety performace and culture: a group model building approach[J]. Accident Analysis & Prevention, 2012, 48: 118–125.

[137]Choi GH. Effectiveness of indirect safety regulation on industrial machines and devices in Korea[J]. Procedia engineer, 2014, 84: 122–125.

[138]陈立杰. 基于安全评价的煤矿企业风险管理对策[J]. 煤矿安全, 2015, 46(1): 236–237.

[139]付长亮, 张爱民. 现代煤化工生产技术[M]. 北京: 化学工业出版社, 2009.

[140]李聿营. 基于风险的检验(RBI)在GE水煤浆气化装置中的应用[J]. 石油和化工设备, 2012, 11: 44–48.

[141]Nonaka I. The knowledge–creating company[M]. Harvard Business Review Press, 2008.

[142]Lukic D, Littlejohn A, Margaryan A. A framework for learning from incidents in the workplace[J]. Safety Science, 2012, 50(4): 950–957.

[143]Heinrich HW, Stone RW. Industrial Accident Prevention[J]. Social Service Review, 1931, 5(2): 323–324.

[144]刁家久. 打造新型化工职业病全生命周期康复疗养体系[J]. 中国软科学, 2022, (S1): 276–279.

[145]Park S, Kook H, Seok H, et al. The Negative Impact of Long Working Hours on Mental Health in Young Korean Workers[J]. PloS one, 2020, 15(8): e0236931.

[146]姚敏, 邵俊杰. 世界级煤制油化工基地创新与实践[M]. 北京: 中国石化出版社, 2019.

[147]张运东, 赵东星. 国际煤制合成天然气技术的专利格局[J]. 石油科技论坛, 2009, 28(4): 59–62.

[148]刘兰翠. 基于模糊Modular神经网络理论的开滦煤矿系统安全评价[D]. 唐山: 河北理工学院, 2002.

[149]李敏睿, 王海清, 姚竣瀚, 等. 安全联锁设备SIL验证中的参数估计偏差分析[J]. 中国安全生产科学技术, 2021, 17(06): 24–29.

[150]中华人民共和国住房和城乡建设部. 石油化工企业设计防火标准(2018年版): GB/T 50160–2008[S]. 北京: 中国计划出版社, 2018.

[151]中华人民共和国工业和信息化部. 压力容器中化学介质毒性危害和爆炸危险度分类: HG/T 20660–2017[S]. 北京: 科学技术文献出版社, 2017.

[152]许文, 张毅民. 化工安全工程概论[M]. 北京: 化学工业出版社, 2010.

[153]张麦秋, 李平辉. 化工生产安全技术第2版[M]. 北京: 化学工业出版社, 2014.

[154]中国标准化委员会. 危险化学品重大危险源辨识: GB/T 1828–2018[S]. 北京: 中国标准出版社, 2018.

[155]中华人民共和国应急管理部. 危险化学品重大危险源监督管理暂行规定[EB/OL]. [2011–08–08].https://www.mem.gov.cn/gk/gwgg/agwzlfl/zjl_01/201108/t20110808_233783.shtml

[156]Cozzani V, Antonioni G, Landucci G, et al. Quantitative assessment of domino and NaTech scenarios in complex industrial areas[J]. Journal of Loss Prevention in the Process Industries, 2014, 28(28): 10–22.

[157]黄孔星, 陈国华, 曾涛, 等. 化工园区Na–Tech事件定量风险评价与防控体系评述[J]. 化工进展, 2019, 38(07): 3482–3494.

[158]Krausmann E, Cruz AM, Affeltranger B. The impact of the 12 May 2008 Wenchuan earthquake on industrial facilities[J]. Journal of Loss Prevention in the Process Industries, 2010, 23(2): 242–248.

[159]王洪德, 崔铁军. 化工园区初始火灾爆炸引发连锁事故概率研究[J]. 安全与环境学报, 2011, 11(04): 158–163.

[160]朱小光. 安全标志识别培训及不同培训方式的有效性研究[J]. 中国安全科学学报. 2013, 23(7): 128–131.

[161]金玉文. 石化企业噪声危害现状评价与分析[J]. 职业卫生与病伤, 2016, 8: 195–198.

[162]Liu T, Lin CC, Huang KC, Chen YC. Effects of noise type, noise intensity, and illumination intensity on reading performance[J]. Applied Acoustics, 2017, 120(13): 70–74.

[163]张文, 林长喜, 彭永臻. 现代煤化工废水近零排放技术集成与优化建议[J]. 环境工程, 2021, 39(11): 41–45+109.

[164]Xiong RH, Wei C. Current status and technology trends of zero liquid discharge at coal chemical industry in China[J]. Journal of Water Process Engineering, 2017, 19: 346–351.

[165]Xiang D, Qian Y, Man Y, et al. Techno–economic analysis of the coal–to–olefins process in comparison with the oil–to–olefins process[J]. Applied Energy, 2014, 113: 639–647.

[166]Liu XD, Jin ZW, Jing YH, et al. Review of the characteristics and graded utilisation of coal gasification slag[J]. Chinese Journal of Chemical Engineering, 2021, 35: 92–106.

[167]刘丹, 朱卫昌, 李墨潇, 等. 利益相关者视角下化工安全关键致因网络分析[J]. 中国安全科学学报, 2023, 33(11): 59–66.

[168]罗帆, 佘廉. 企业组织管理预警系统评价指标的权重及综合评价[J]. 中国软科学, 2000, (07): 116–118.

[169]李峰, 储胜利, 程宗华. 石油石化重大工艺安全事故应急机制浅析[J]. 中国安全生产科学技术, 2014, 10(02): 126–130.

[170]Bourassa D, Gauthier F, Abdul–Nour G. Equipment failures and their contribution to industrial incidents and accidents in the manufacturing industry[J]. International Journal of Occupational Safety and Ergonomics Jose, 2016, 2(1): 131–141.

[171]程启月. 评测指标权重确定的结构熵权法[J]. 系统工程理论与实践, 2010, 30(7): 1225–1228.

[172]杨春燕, 蔡文. 可拓工程研究[J]. 中国工程科学, 2000, 12: 90–96.

[173]李德毅, 孟海军, 史雪梅. 隶属云和隶属云发生器[J]. 计算机研究与发展, 1995, 6: 19–20.

[174]Yu K, Cao QG, Xie CZ, et al. Analysis of intervention strategies for coal miners’ unsafe behaviors based on analytic network process and system dynamics[J]. Safety science, 2019, 118: 145–157.

[175]Kenan S, Kadri S. Process safety leading indicators survey–February 2013: Center for chemical process safety–white paper[J]. Process Safety Progress, 2014, 33(3): 247–258.

[176]Shu Y, Ming L, Cheng F, et al. Abnormal Situation Management: Challenges and Opportunities in the Big Data Era[J]. Computers & Chemical Engineering, 2016, 91: 104–113.

[177]胡瑾秋, 郭放, 张来斌. 结合改进PSO算法和LSSVM的化工异常工况超早期监测预警研究[J]. 电子测量与仪器学报, 2018, 32(2): 36–41.

[178]曹静, 高波, 邢海燕, 等. 利用装置采集数据提升炼化应用技术的实现[J]. 计算机与应用化学, 2019, 5: 543–547.

[179]张明, 冯坤, 江志农, 等. 基于动态自学习阈值和趋势滤波的机械故障智能预警方法[J]. 振动与冲击, 2014, 33(24): 8–14.

[180]王益, 杨剑波, 李锐, 等. 基于时空序列的CNN–GRU气温预测模型[J]. 计算机应用, 2023, 43(S2): 54–59.

[181]史业照, 郭斌, 郑永军. 基于LSTM网络的IGBT寿命预测方法研究[J]. 中国测试, 2024, 50(02): 54–58+65.

[182]Rufus AKL. A GOA–RNN Controller for a Stand–alone Photovoltaic/Wind Energy Hybrid–fed Pumping System[J]. Soft Computing, 2019, 23: 12255–12276.

[183]潘雄, 黄伟凯, 赵万卓, 等. 基于BiLSTM模型的BDS–3短期钟差预报精度研究[J].测绘学报, 2024, 53(01): 65–78.

[184]蔡亮, 周泓岑, 白恒, 等. 基于多层BiLSTM和改进粒子群算法的应用负载预测方法[J]. 浙江大学学报(工学版), 2020, 54(12): 2414–2422.

[185]Xu M, Cao L, Lu D, et al. Application of Swarm Intelligence Optimization Algorithms in Image Processing: A Comprehensive Review of Analysis, Synthesis, and Optimization[J]. Biomimetics, 2023, 8(2): 235.

[186]Du J, Hou J, Wang HY, et al. Application of an improved whale optimization algorithm in time–optimal trajectory planning for manipulators[J]. Mathematical Biosciences and Engineering, 2023, 20(9): 16304–16329.

[187]Mirjalili S, Lewis A. The Whale Optimization Algorithm[J]. Advances in Engineering Software, 2016, 95: 51–67.

[188]Wan J, Liu Z, Chan AB. A Generalized Loss Function for Crowd Counting and Localization[C]. Computer Vision and Pattern Recognition IEEE, 2021.

[189]Pan J, Chen S. A new approach for assessing the correlated risk[J]. Industrial Management & Data Systems, 2012, 112(9), 1348–1365.

[190]Dey S. Bayesian estimation of the parameter and reliability function of an inverse Rayleigh distribution[J]. Malaysian Journal of Mathematical Sciences, 2012, 6(1): 113–124.

[191]Chan WM, Ibrahim RN, Lochert PB. Quality evaluation model using loss function for multiple S–type quality characteristics[J]. The International Journal of Advanced Manufacturing Technology, 2005, 26: 98–101.

[192]Hashem SJ, Ahmed S, Khan F. Loss Functions and Their Applications in Process Safety Assessment[J]. Process Safety Progress, 2015, 33(3): 285–291.

[193]Wang H, Khan F, Ahmed S, et al. Dynamic quantitative operational risk assessment of chemical processes[J]. Chemical Engineering Science, 2016, 142: 62–78.

[194]张树凯, 刘正江, 蔡垚, 等. 综合安全评估的研究进展及展望[J]. 哈尔滨工程大学学报, 2021, 42(01): 152–158.

[195]Arunthavanathan R, Khan F, Ahmed S, et al. An Analysis of Process Fault Diagnosis Methods from Safety Perspectives[J]. Computers & Chemical Engineering, 2021, 145, 107197.

[196]Jafari MJ, Mohammadi H, Reniers G, et al. Exploring inherent process safety indicators and approaches for their estimation: A systematic review[J]. Journal of Loss Prevention in the Process Industries, 2018, 52: 66–80.

[197]谢丽蓉, 王斌, 包洪印, 等. 基于EEMD–WOA–LSSVM的超短期风电功率预测[J]. 太阳能学报, 2021, 42(7): 290–296.

[198]石延辉, 杨洋, 廖毅, 等. 基于改进粒子群算法优化SVM的变压器故障诊断[J]. 武汉大学学报(工学版), 2023, 56(10): 1238–1244.

[199]Tomar D, Agarwal S. A comparison on multi–class classification methods based on leastsquares twin support vector machine[J]. Knowledge–Based Systems, 2015, 81: 131–147.

[200]张开放, 苏华友, 窦勇. 一种基于混淆矩阵的多分类任务准确率评估新方法[J]. 计算机工程与科学, 2021, 43(11): 1910–1919.

[201]孔英会, 景美丽. 基于混淆矩阵和集成学习的分类方法研究[J]. 计算机工程与科学, 2012, 34(06): 111–117.