论文部分内容阅读
针对高层建筑施工过程中传统的爬架组安全监测系统对传感器数据多采用独立处理和分别判决的方式,易受外界环境干扰导致判定结果不准确,形成安全隐患的问题,提出了一种基于BP神经网络数据融合技术的爬架组安全监测系统。首先,介绍了爬架组安全监测系统总体架构,然后构建了BP神经网络数据融合模型,该模型由多传感器数据归一化输入层、两级BP神经网络处理层和状态判定处理模块构成。通过对BP神经网络的学习训练得到BP神经网络的结构参数,将该技术应用于爬架组安全监测系统。测试表明,所提出的方案与传统的相比降低了漏报率,从而提高了系统的可靠性和安全性。
In the process of high-rise building, the traditional climbing frame group safety monitoring system mainly adopts the method of independent processing and judging separately for the sensor data, and is easily subject to the interference of the external environment, which leads to the inaccurate judgment result and the potential safety hazard. Neural network data fusion technology of climbing group safety monitoring system. First of all, the overall framework of the safety monitoring system of climbing group is introduced. Then a BP neural network data fusion model is constructed, which is composed of the input layer of multi-sensor data normalization, BP neural network processing layer of two stages and state decision processing module. The structure parameters of BP neural network are obtained through the learning and training of BP neural network. The technology is applied to the climbing group safety monitoring system. The test shows that the proposed scheme reduces the false negative rate compared with the traditional one, which improves the reliability and security of the system.