基于GM-BP模型的塔机合理分配应用研究  

Research on Reasonable Allocation of Tower Crane Based on GM-BP Model

作　　者：蔡秀梅 宋梦鸽 杜嵬 CAI Xiu-mei;SONG Meng-ge;DU Wei(School of Automation, Xi＇ an University of Posts ＆ Telecommunications, Xi＇ an 710121, China;Beijing Uni. -Construction Group Co., Ltd., Beijing 100101, China)

机构地区：[1]西安邮电大学自动化学院,陕西西安710121 [2]北京住总集团有限责任公司,北京100101

出　　处：《测控技术》2018年第5期29-32,共4页Measurement & Control Technology

基　　金：国家自然科学基金资助项目(51205309);西安市科技计划(CXY1516(1));陕西省教育厅科技计划项目(16JK1712)

摘　　要：塔式起重机是建筑工地应用最频繁的起重设施之一,但在实际使用中,由于塔机运输的物料分属于不同的分包商,经常导致工地现场分包商间为争夺塔机资源而出现争论以及带来的分配不均等诸多问题。为解决该问题,将BP神经网络引入到灰色预测模型(GM)中,建立塔机耗费数据的GM(1,1)和BP神经网络组合模型,通过对塔机的使用数据分析预测,实现塔机资源分配合理化。验证结果表明,该组合模型与单一GM模型或BP神经网络模型相比,具有较高的预测精度,对塔机的合理分配有较好的效果。Tower crane is one of the most frequently used lifting facilities in construction sites. However, in actual use, the listing goods belong to different subcontractors, there are always lots of debates among the subcontractors due to fighting for tower crane resources, which bring about uneven distribution of tower crane resources. In order to solve this problem, the BP neural network was introduced into the gray prediction model（GM）, and the combined model of GM（ 1,1） and BP neural network was created to predict the consumption data of tower crane. According to the consumption data, the tower crane can be distributed reasonably. The verification results show that the prediction accuracy of combined model is higher than that of single GM model and BP neural network model, which has a good effect on the reasonable distribution of the tower crane.

关 键 词：BP神经网络 灰色预测模型 塔机 资源分配 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TU73[自动化与计算机技术—控制科学与工程]