北京城市塑料垃圾年产量的模拟预测及其影响因素分析  

Prediction of Beijing′s urban plastic waste generation based on multiple models and analysis of impact factors upon the generation

作　　者：蒋晓燕 温小乐[1] 罗维[2] JIANG Xiaoyan;WEN Xiaole;LUO Wei(College of Environment and Resources,Fuzhou University,Fuzhou 350116;Laboratory of Solid Waste Treatment and Recycling,Research Center for Eco-Environmental Sciences,Chinese Academy of Sciences,Beijing 100085)

机构地区：[1]福州大学环境与资源学院,福州350116 [2]中国科学院生态环境研究中心固体废弃物处理与资源化实验室,北京100085

出　　处：《环境科学学报》2020年第9期3435-3444,共10页Acta Scientiae Circumstantiae

基　　金：国家重点研发计划项目(No.2017YFC0505803-01,2017YFD0800202);科技部(No.2019QZKK0605);国家自然基金项目(No.41761144078,41571479)。

摘　　要：揭示城市塑料垃圾年产量及影响因素、预测其发展趋势对于城市生活垃圾收集系统的优化、处理技术的合理选择和降低环境影响具有重要意义.本研究基于1989年以来北京塑料垃圾占比、城市生活垃圾产量数据和社会经济数据,利用赤池信息量准则(AIC)和灰关联度法研究了北京城市塑料垃圾占比的年变化趋势和城市塑料垃圾年产量的主要影响因素.通过多元线性回归模型(MLR)、灰色系统模型GM(1,1)和BP神经网络模型对北京城市塑料垃圾年产量进行了模拟预测.结果表明,北京城市塑料垃圾占比由1989年的1.88%,增加到2012年的14.87%.基于AIC准则预测2013—2050年北京城市塑料垃圾占比增长趋势较平缓、稳定在14%~19%之间.2000—2012年北京市城市塑料垃圾年产量由40.2×104 t增加到121.1×104 t,年增长15.5%.人均可支配收入是影响北京城市塑料垃圾年产量的最大社会经济因素,而常住人口的影响较低.BP神经网络是模拟预测北京城市塑料垃圾产量的最佳模型,其模拟预测结果表明:2013年后北京市塑料垃圾年产量随时间呈不规则的非线性增长趋势,到2025、2035、2050年北京城市塑料垃圾产量将分别达到335、488和859×104 t,将对北京城市生活垃圾处理处置与防控管理带来巨大挑战.Estimation of UPW generation and determination of impact factors upon the generation are of utmost importance for the planning and design of MSW collection,treatment,disposal,control and management in order to reduce its environmental effects.In the present study,data of society,economy,the proportion of UPW in Beijing since 1989,and MSW generation were systematically collected from yearbooks of Beijing.Annual changes in the proportion of UPW in Beijing and the main impact factors upon UPW generation were determined,using the Aka Information Criterion(AIC)and grey correlation method.Multivariate linear regression model(MLR),grey system model GM(1,1)and Back propagation(BP)neural network model were used to estimate and predict the annual generation of UPW in Beijing.Results show that the proportion of UPW in Beijing increased from 1.88%in 1989 to 14.87%in 2012.Based on the AIC,the growth of the proportion of UPW would gently increase and fluctuate between 14%and 19%in 2013—2050.From 2000 to 2012,Beijing′s annual generation of UPW increased from 4.02 to 1.21 million tons,with an average annual growth of 15.5%.As a socio-economic factor,per capita disposable income has the greatest impact upon the generation of UPW in Beijing,while the resident population does the lowest.BP neural network is the best model for predicting Beijing′s annual generation of UPW.The prediction results show that the generation would increase in an irregular and non-linear trend after 2013 and reach 3.35,488,and 8.59 million tons in 2025,2035,and 2050 respectively,which would bring great challenges to the process,disposal,prevention,control,and management of MSW in Beijing.

关 键 词：城市垃圾 产量预测 灰色系统模型GM(1 1) BP神经网络 

分 类 号：X705[环境科学与工程—环境工程] X32