基于GA-BP算法的公路工程造价估算研究  

Research on Highway Engineering Cost Estimation Based on the GA-BP Algorithm

作　　者：潘雨红[1] 张宜龙[1] 蔡亚军[1] 吴欢欢[1] 隋鸿艳 

机构地区：[1]重庆交通大学经济与管理学院,重庆400074 [2]重庆市建设项目管理有限公司,重庆401121

出　　处：《重庆交通大学学报（自然科学版）》2016年第2期141-145,共5页Journal of Chongqing Jiaotong University(Natural Science)

摘　　要：针对目前公路工程造价估算存在的问题,结合人工神经网络模型和遗传算法各自的特点,优势互补,将人工智能算法GA-BP模型引入公路工程造价估算中。利用已建公路工程造价资料,找出影响造价的特征值及各自的隶属度。以MATLAB为平台,通过实验仿真将该GA-BP算法与BP算法结果进行比较,两者测试样本得到的结果均达到预期精度,且前者比后者的精度高。同时还验证了优化后的GA-BP算法不仅计算简便、快捷,且具有较高的计算精度,从而得出了该算法的可行性与有效性。Aiming at the problems existing in current highway engineering cost estimation,in combination with respective features of artificial neural network model and genetic algorithm to achieve mutual complement of respective advantages,the artificial intelligence algorithm —GA-BP algorithm model is introduced into the highway engineering cost estimation. The engineering cost data of some completed highways was used in an attempt to find the characteristic values impacting cost and their respective membership degrees. Based on the MATLAB platform,the results of GA-BP algorithm was compared with that of BP algorithm by the simulating experiment. The findings show that the results of sample tests by the two methods all meet expected accuracy and the GA-BP is of higher accuracy meanwhile the results also prove that the optimized GA-BP algorithm is not only simpler and faster but also of high calculation accuracy thus being feasible and effective.

关 键 词：管理工程 GA-BP算法 公路工程 造价估算 

分 类 号：C931.2[经济管理—管理学]