基于支持向量回归机的墙面抹灰机作业效果预测  

Construction effect prediction for a wall plastering machine on SVR

作　　者：杨振宇[1,2] 王勇[1] 张兴波 刘发英[1] 

机构地区：[1]山东理工大学机械工程学院,山东淄博255049 [2]中国农业大学工学院,北京100083 [3]高青县科学技术局,山东淄博256300

出　　处：《西安建筑科技大学学报（自然科学版）》2013年第5期750-754,共5页Journal of Xi'an University of Architecture & Technology(Natural Science Edition)

基　　金：中央高校基本科研业务费专项资金资助项目(2013XJ005)

摘　　要：为了能够预测墙面抹灰机机械施工墙面的15d抗压强度,提高施工质量、进度和作业效率.测定五种成分含量不同的100个普通混合砂浆抹灰后的15d抗压强度,将其中80个数据作为训练集样本,20个数据作为测试集样本.建立支持向量回归机模型预测墙面抹灰机机械施工墙面砂浆抗压强度,使用Matlab编程预测15d抗压强度.试验结果表明该方法训练集和测试集的均方误差分别为1.634 7×10-4和1.395 2×10-3,决定系数分别达到0.998 88和0.994 92.因此所建立的支持向量回归机模型具有很好的泛化能力,为墙面抹灰机在施工过程中对墙面抗压强度的预测提供了依据.In order to predict 15d compressive strength of the wall conducted by the wall plastering machine and improve construction quality, construction progress and work efficiency, 15d strength of 100 ordinary mix mortar with five differ- ent kinds of composition are detected after the plastering construction, and 80 of the data are taken as the training sets samples, and 20 of data are as the test set samples. SVR model is established to predict wall mortar compressive strength after wall plastering machine＇s work, and 15d compressive strength are predicted by Matlab program. The test results show that the mean-square error of the training set and the testing set are respectively 1. 634 7 X 10-4 and 1. 395 2 X 10-3 , and the decision coefficient of the training set and the testing set are respectively 0. 998 88 and 0. 994 92. Therefore, SVR model has good capacity of generalization, and it provides a basis for predicting 15d compressive strength of the wall sur- face during the wall plastering machine is in operation.

关 键 词：墙面抹灰机 15 d抗压强度 预测 支持向量回归机 

分 类 号：TU741.1[建筑科学—建筑技术科学]