摘要
针对建筑工程施工成本管理中成本难以预测的问题,提出用鸟群算法(BSA)优化极限学习机(ELM)模型的参数.首先,利用BSA对ELM模型的输入权值和偏置值进行优化;其次,构建出BSA-ELM建筑工程施工成本预测模型;最后,将BSA-ELM模型与实际工程施工成本数据相结合进行验证.结果表明:模型在成本预测中的精度比ELM模型、CSO-ELM模型、PSO-ELM模型和BP神经网络模型预测精度高,也为类似预测问题提供了一种新的预测方法.
Aiming at the problem that the cost is difficult to predict in building project construction cost management,the parameters of the extreme learning machine(ELM) model are optimized by the bird swarm algorithm(BSA).Firstly,the input weight and offset value of ELM model axe optimized by BSA.Secondly,the BSA-ELM building project construction cost prediction model is constructed.Finally,the BSA-ELM model is combined with the actual construction cost data to verify.The results show that the accuracy of the model in cost prediction is higher than that of ELM、CSO-ELM model、PSO-ELM model and BP neural network models,and it provides a new prediction method for similar prediction problems.
作者
李万庆
陈佳琪
孟文清
马利华
LI Wan-qing;CHEN Jia-qi;MENG Wen-qing;MA Li-hua(School of Management Engineering and Business,Hebei University of Engineering,Handan 056038,China;School of Civil Engineering,Hebei University of Engineering,Handan 056038,China)
出处
《数学的实践与认识》
北大核心
2019年第23期10-17,共8页
Mathematics in Practice and Theory
关键词
建筑工程施工成本
预测
极限学习机
鸟群算法
BSA-ELM
building project construction cost
prediction
extreme learning machine
bird swarm algorithm
BSA-ELM
