基于MPU6050惯性传感器的钢厂天车大车运行监测系统研究

Research on the Operational Monitoring System for Bridge Girder Traveling Mechanism in Steel Plants Based on the MPU6050 Inertial Sensor

摘要: 天车大车(Bridge Girder Traveling Mechanis)作为钢铁生产流程中的核心运输设备，其运行稳定性直接关系生产安全与效率。针对传统监测方案成本高、实时性差、环境适应性弱的痛点，本文提出一种基于MPU6050惯性传感器的天车大车运行状态监测系统。该系统以STM32F103ZET6为核心处理器构建多终端数据采集模块，通过LoRa无线网络实现数据实时传输，结合QT框架设计上位机人机交互界面，并创新性采用动态阈值与随机森林融合的双决策算法，实现对啃轨、异常振动等故障的精准检测。实验表明，在模拟天车运行平台上，系统对严重啃轨(姿态角偏移 > 5˚)和异常振动(加速度超限)的检测正确率分别达98.1%和92.3%，显著优于传统固定阈值方法。该方案为工业天车的低成本、高鲁棒性状态监测提供了新思路。

Abstract: The bridge girder traveling mechanism (BGT) serves as a core transportation system in steel production, where its operational stability is directly linked to production safety and efficiency. Addressing the challenges of high cost, poor real-time performance, and weak environmental adaptability in traditional monitoring solutions, this paper proposes a BGT operation state monitoring system based on the MPU6050 inertial sensor. The system employs the STM32F103ZET6 as the central processor to construct a multi-terminal data acquisition module, utilizes a LoRa wireless network for real-time data transmission, and integrates a human-machine interaction interface designed with the Qt framework. Moreover, a novel dual-decision algorithm combining dynamic thresholding and random forest classification is implemented to achieve accurate detection of rail gnawing and abnormal vibrations. Experimental results on a simulated BGT platform demonstrate that the system achieves detection accuracies of 98.1% for severe rail gnawing (attitude angle deviation > 5˚) and 92.3% for abnormal vibrations (exceeding acceleration limits), significantly outperforming traditional fixed-threshold methods. This study provides a novel approach for cost-effective and robust condition monitoring of industrial BGT systems.