基于因子分析与BP网络的地铁施工安全预警研究

为了解决当前我国地铁施工过程的安全预警问题,构建因子分析与BP神经网络相结合的地铁施工安全预警模型。在分析地铁施工安全预警指标的基础上,采用SPSS因子分析法对调查数据进行降维,采用Visual Basic 6.0软件编写BP网络程序,并通过工程实际数据实现模型的训练及检测。研究结果表明,通过因子分析能使BP网络的输入数据从37个减少至7个,经因子分析降维后的收敛速度和计算精度均高于未经因子分析的神经网络,且误差均在10%以内。通过因子分析与BP神经网络相结合构建的耦合模型识别地铁施工过程中的不安全因素,进而有针对性地完善地铁施工的相关预警技术。

Subway Construction Safety Early Warning Research Based on Factor Analysis and BP Network

In order to solve the urgent problem of safety early warning in current subway construction in China,a safety warning model was built by factor analysis combined with BP neural network.Based on safety early warning indicators analysis of subway construction,SPSS factor analysis method was used to reduce the dimension of survey data,Visual Basic 6.0 software was used to write BP network program,and the model training and testing were achieved through actual project data.The results show that the input data of BP network can be reduced from 37 to 7 by factor analysis,and the convergence speed and calculation accuracy are higher than those of the neural network having not underwent factor analysis,and the errors are less than 10%.So using this coupling model,dangerous factors affecting subway construction can be identified,and relevant early warning technique can be improved.