Pang Jiayi, Cui Meng^*, Zhu Hong

Liaoning Institute of Science and Technology

Abstract：

With the rapid development of super high-rise buildings, the scale of their mechanical and electrical systems has grown increasingly large and complex, posing severe challenges to equipment safety during installation. Traditional experience-based management models relying on manual inspections struggle to detect anomalies such as collisions, tipping, and overloads during hoisting, transportation, and positioning processes in real-time and with precision, often leading to equipment damage, project delays, and safety incidents. To address this issue, this study develops an IoT-based multi-source sensor technology system for equipment anomaly early warning. The system integrates vibration, tilt angle, stress, and ultra-wideband (UWB) positioning sensors, with intelligent sensing nodes attached to critical equipment to collect real-time status data. Utilizing LoRaWAN and Wi-Fi hybrid networking technology, it ensures reliable data transmission in complex construction environments. On the cloud platform, a model combining time-series threshold analysis and machine learning algorithms performs intelligent diagnosis and early warning of anomalies. Finally, a web visualization platform provides managers with real-time monitoring, alarm alerts, and decision support. This study completes the system's hardware and software design and implementation, with simulation experiments and field tests verifying its excellent performance in real-time responsiveness, accuracy, and reliability, significantly enhancing safety management and intelligentization levels in high-rise building mechanical and electrical installations.

Key Words：

mechanical and electrical installation; abnormality early warning; internet of things; machine learning