为提高出口风险预警的准确性和科学性,构建了一种基于K-means聚类与粒子群优化BP(PSO-BP)神经网络的危险废物出口风险预警系统.基于历史数据的多维指标,采用K-means聚类算法对危险废物出口风险进行分类,并运用PSO算法优化BP神经网络的初始参数,提高风险预测的精度和收敛效率.实验结果表明,K-means聚类的风险分类与专家标注的一致性达93.2%,证明该方法能够有效揭示数据的内在风险结构;PSO-BP神经网络的预测准确率达92.5%,均方误差降低至0.012,相较于传统BP神经网络,预测精度显著提升,训练收敛速度提高40%.基于上述模型,本研究开发了一套危险废物出口风险预警系统,实现数据采集、风险评估、模型预测与预警可视化等功能,并以30类危险废物为例验证了模型的实现与应用.结果表明,系统能够实时分析危险废物出口数据,依据风险预测结果提供动态预警,从而有效提升危险废物出口管理工作的科学性和效率.
Abstract
Improper export of hazardous waste may cause serious environmental pollution and regulatory risks. To improve the accuracy and scientific nature of export risk warning, this study constructed a hazardous waste export risk warning system based on K-means clustering and particle swarm optimization BP (PSO-BP) neural network. Based on multidimensional indicators of historical data, the export risk of hazardous waste was classified using the K-means clustering algorithm, and the initial parameters of the BP neural network were optimized by the PSO algorithm, thereby improving the accuracy and convergence efficiency of risk prediction. The experimental results showed that the consistency between the risk classification by K-means clustering and expert annotation was 93.2%, proving that the inherent risk structure of the data could be effectively revealed by this method. The prediction accuracy of the PSO-BP neural network reached 92.5%, and the mean square error was reduced to 0.012. Compared with the traditional BP neural network, the prediction accuracy was significantly improved, and the training convergence speed was increased by 40%. Based on the above model, a hazardous waste export risk warning system was developed in this study, through which data collection, risk assessment, model prediction, and warning visualization were achieved. The implementation and application of the model were verified using 30 types of hazardous waste as examples. The results showed that hazardous waste export data could be analyzed in real time, dynamic warnings could be provided based on risk prediction results, and the efficiency of hazardous waste export management could be effectively improved.
关键词
危险废物 /
出口风险 /
K-means聚类 /
BP神经网络 /
粒子群优化 /
风险预警
Key words
hazardous waste /
export risk /
K-means clustering /
BP neural network /
particle swarm optimization /
risk warning
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Callao C, Martinez-Nuñez M, Latorre M. European countries: Does common legislation guarantee better hazardous waste performance for European Union member states? [J]. Waste Management, 2019,84: 147-157.
[2] Lin X Q, Li M, Chen Z L, et al. Long-term monitoring of PCDD/Fs in soils in the vicinity of a hazardous waste incinerator in China: Temporal variations and environmental impacts [J]. Science of the Total Environment, 2020,713:136717.
[3] Ankit, Saha L, Kumar V, et al. Electronic waste and their leachates impact on human health and environment: Global ecological threat and management [J]. Environmental Technology & Innovation, 2021,24:102049.
[4] Pana D A, Li L L, Tian X, et al. A review on lead slag generation, characteristics, and utilization [J]. Resources Conservation and Recycling, 2019,146:140-155.
[5] Miyamoto K, Kobayashi H. Urgent need to review domestic regulations in Japan for preventing illegal transboundary movement of hazardous wastes [J]. Resources Conservation and Recycling, 2020, 158:104823.
[6] Um N, Park Y S, Jeon T W. An improved strategy for effectively managing the transboundary movement of waste based on the basel convention: A case study in South Korea [J]. Heliyon, 2023,9(6):e16496.
[7] 王兆龙,张 喆,贾 佳,等.中国控制危险废物越境转移管理现状与展望 [J]. 中国环境管理, 2022,14:7-12. Wang Z L, Zhang J, Jia J, et al. Progress and prospect on the control of transboundary movements of hazardous wastes in China [J]. Chinese Journal of Environmental Management, 2022,14:7-12.
[8] Jiao J L, He P W, Zha J R. Factors influencing illegal dumping of hazardous waste in China [J]. Journal of Environmental Management, 2024,354:120366.
[9] 钭晓东,赵文萍.人工智能实施背景下的环境风险控制——以社会源危险废弃物全程追踪机制为样本 [J]. 中国高校社会科学, 2019: 40-51. Tou X D, Zhao W P. Environment risk control under the background of artificial intelligence implementation [J]. Social Sciences in Chinese Higher Education, 2019:40-51.
[10] 李福全.基于物联网的冶金企业危险废物储运监控信息化系统 [J]. 信息记录材料, 2023,24:206-208. Li F Q. Information system for monitoring and controlling hazardous waste storage and transportation in metallurgical enterprises based on the internet of things [J]. Information Recording Materials, 2023,24:206-208.
[11] 方 文,刘 正,黄玉洁,等.基于多模态感知的危险废物环境风险评估与预警研究 [J]. 中国环境管理, 2022,14:13-18. Fang W, Liu Z, Huang Y J, et al. Environmental risk assessment and early warning of hazardous waste base on multi-mode sensor [J]. Chinese Journal of Environmental Management, 2022,14:13-18.
[12] 蒋文博,黄玉洁,刘 正,等.基于健康风险的危险废物智能化分级分类研究 [J]. 环境监控与预警, 2021,13:14-18. Jiang W B, Huang Y J, Liu Z, et al. Intelligent classification system of hazardous waste based on health risk assessment [J]. Environmental Monitoring and Forewarning, 2021,13:14-18.
[13] 李玉爽,霍慧敏,刘海兵,等.危险废物城市环境风险评估方法及案例研究 [J]. 环境工程学报, 2023,17:2993-3004. Li Y S, Huo H M, Liu H B, et al. Environmental risk assessment method and case study for hazardous wastewithin the city [J]. Chinese Journal of Environmental Engineering, 2023,17:2993-3004.
[14] Takata M, Lin B L, Xue M Q, et al. Predicting the acute ecotoxicity of chemical substances by machine learning using graph theory [J]. Chemosphere, 2020,238:124604.
[15] Alibakshi A. Strategies to develop robust neural network models: Prediction of flash point as a case study [J]. Analytica Chimica Acta, 2018,1026:69-76.
[16] 吴 波,朱林萍,李扬波,等.基于K-Means聚类模型的隧道施工安全风险评价方法及应用研究 [J]. 重庆交通大学学报(自然科学版), 2023,42:80-87. Wu B, Zhu L P, Li Y B, et al. Evaluation method and application of tunnel construction safety risk based on K-Means Clustering Model [J]. Journal of Chongqing Jiaotong University(Natural Science), 2023, 42:80-87.
[17] 杨 帆,王红斌,方 健,等.基于K-means聚类和随机森林的电缆风险评估及修复决策 [J]. 南京航空航天大学学报, 2024,56:892-899. Yang F, Wang H B, Fang J, et al. Cable risk assessment and repair decision based on K-means clustering and random forest [J]. Journal of Nanjing University of Aeronautics & Astronautics, 2024,56:892-899.
[18] Kim J, Manna A, Roy A, et al. Clustered vehicle routing problem for waste collection with smart operational management approaches [J]. International Transactions in Operational Research, 2025,32(2):863-887.
[19] 何 山,赵宇明,田爱娜,等.基于充电片段和PSO-BP的锂电池SOH在线估计方法 [J]. 电源技术, 2025,49:383-389. He S, Zhao Y M, Tian A N, et al. An online SOH estimation method for lithium batteries based on charging fragments and PSO-BP [J]. Chinese Journal of Power Sources, 2025,49:383-389.
[20] 杜 莹,韦原原,蒲欢欢.基于PSO-BP神经网络的5G基站位置确定方法 [J]. 测绘工程, 2025,34:47-52. Du Y, Wei Y Y, Pu H H. A method for 5G access point location determination based on PSO-BP neural network [J]. Engineering of Surveying and Mapping, 2025,34:47-52.
[21] 翟东海,鱼 江,高 飞,等.最大距离法选取初始簇中心的K-means文本聚类算法的研究 [J]. 计算机应用研究, 2014,31:713-715. Zhai D H, Yu J, Gao F, et al. K-means text clustering algorithm based on initial cluster centers selection according to maximum distance [J]. Application Research of Computers, 2014,31:713-715.
[22] Jiang Y, Zhang C, Zheng X, et al. Simultaneously screening multiple UGT1A1inhibitors from Polygonum multiflorum root using ultrafiltration LC-MS [J]. Biomedical Chromatography, 2022,36(4):e5300.
[23] Shi W, Zhang C, Zhao D S, et al. Discovery of hepatotoxic equivalent combinatorial markers from Dioscorea bulbifera tuber by fingerprint- toxicity relationship modeling [J]. Scientific Reports, 2018,8(1):462.
[24] 曹红业,龚 涛,袁成忠,等.基于RBF模型的太湖北部叶绿素a浓度定量遥感反演 [J]. 环境工程学报, 2016,10:6499-6504. Cao H Y, Gong T, Yuan C Z, et al. Quantitative retrieval of chlorophyll-a concentration in northern part of Lake Taihu based on RBF model [J]. Chinese Journal of Environmental Engineering, 2016, 10:6499-6504.
[25] 刘翘楚,王 杰,秦文萍,等.基于嵌套优化的GA-PSO-BP神经网络短期风功率预测方法研究 [J]. 电网与清洁能源, 2025,41:138-146. Liu Q C, Wang J, Qin W P, et al. Research on short-term wind power prediction methods based on GA-PSOPB neural network with nested optimization [J]. Power System and Clean Energy, 2025,41:138-146.
基金
国家重点研发计划项目(2019YFC1904803)
PDF(974 KB)
京公网安备 11010802030352号 
