基于GA-BP神经网络的建筑垃圾产量预测--以东西部代表省份比较为例  

Prediction of Construction Waste Generation Based on GA-BP Neural Network-A Comparison of Representative Provinces in the East and West

作　　者：郭容 姚潞 孙洪浩 许茜 王东[3] 刘红梅 Guo Rong;Yao Lu;Sun Honghao;Xu Qian;Wang Dong;Liu Hongmei(College of Transportation and Civil Engineering,Nantong University,Nantong 226019;Xinglin College,Nantong University,Nantong 226019;China Electronic Systems Engineering Second Branch Company,Wuxi 214135)

机构地区：[1]南通大学交通与土木工程学院,南通226019 [2]南通大学杏林学院,南通226019 [3]中国电子系统工程第二建设有限公司二分公司,无锡214135

出　　处：《建设科技》2022年第19期82-85,共4页Construction Science and Technology

基　　金：南通大学博士科研基金项目(2018-33);江苏省研究生科研与实践创新项目(KYCX20_2832)。

摘　　要：为研究中国建筑垃圾的产生情况,以东西部江苏省和四川省两个代表省份为例,构建模型对建筑垃圾产量进行分析和预测。本文运用建筑面积估算法计算出2000年至2020年的建筑垃圾产量,并采用Pearson相关性进行影响因素分析。在此基础上,建立GA-BP神经网络模型对不同地区的代表省份进行垃圾产量预测。研究结果显示,优化的GA-BP神经网络模型预测精度比BP神经网络更高。用此模型预测的江苏省和四川省2028年建筑垃圾产量分别为314.66百万吨和179.34百万吨。本研究结果有助于垃圾管理部门为东西部地区制定科学有效的建筑垃圾资源化管理策略。In order to study the generation of construction waste in China,two representative provinces in the east and west,namely Jiangsu Province and Sichuan Province,were selected to build a model to analyze and predict the generation of construction waste.The construction waste generation from 2000 to 2020 was calculated by using the floor area estimation method,and Pearson correlation was used to analyze the influencing factors.Based on this,GA-BP neural network model was established to predict the waste generation in the representative provinces of different regions.The results of the study show that the prediction accuracy of optimized GA-BP neural network model is higher than that of BP neural network.The generation of construction waste predicted by this model in Jiangsu Province and Sichuan Province in 2028 is 314.66 million tons and 179.34 million tons respectively.The results of this study can help waste management authorities to develop scientific and effective construction waste resource management strategies for the eastern and western regions.

关 键 词：建筑垃圾 东西部 影响因素 GA-BP神经网络模型 产量预测 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] X799.1[自动化与计算机技术—控制科学与工程]