基于多尺度卷积网络的牵张设备远程状态监测系统

Remote state monitoring system of tension equipment based on multi⁃scale convolution network

为缩小牵张设备运行过程中谷电度、平电度数值与标准电量输出结果之间的差值水平,使设备结构长期保持相对稳定的工作状态,设计基于多尺度卷积网络的牵张设备远程状态监测系统。根据卷积神经网络架构的连接形式,设置终端处理器与无线通信模块,完成监测系统的硬件执行环境搭建。在此基础上,实施协议移植处理,通过A/D转换的方式,确定串行通信子程序的应用能力,实现监测系统的软件执行环境搭建,结合相关硬件设备结构,完成牵张设备远程状态监测系统设计。实验结果显示,该文所设计的监测系统可同时缩小牵张设备谷电度、平电度数值与标准电量输出结果之间的物理差值,对维持设备结构的长期稳定工作起到较大作用。

In order to reduce the difference level between valley electric degree,level electric degree and standard electric output results during the operation of traction and tension equipment,and keep the equipment structure in a relatively stable working state for a long time,a remote condition monitoring system of traction and tension equipment based on multi⁃scale convolution network is designed.According to the connection form of convolutional neural network architecture,the terminal processor and wireless communication module are set up to complete the hardware execution environment of the monitoring system.On this basis,the protocol transplantation process is implemented,the application capability of serial communication subroutine is determined through A/D conversion,the software execution environment of the monitoring system is built,and the design of remote condition monitoring system of traction equipment is completed in combination with relevant hardware equipment structure.The experimental results show that the monitoring system designed in this paper can reduce the physical difference between the valley electric degree and level electric degree of the traction equipment and the standard electric output results at the same time,and play a strong role in maintaining the long⁃term stability of the equipment structure.