Design and Edge-Enabled Implementation of Smart Acousto-Optic Safety Wearables for Power IoT Environments

Abstract

With the rapid development of the power Internet of Things, the safety risks faced by power facility maintenance personnel, such as high-voltage electric shock, equipment failure, and severe weather, have become increasingly prominent. Traditional protective measures are no longer able to meet the safety needs in complex environments. To this end, this article designs an intelligent sound and light alarm workwear based on the power Internet of Things. The workwear integrates infrared detection, microwave perception, video monitoring, and AI intelligent recognition technology to build a comprehensive monitoring system. Real time data is collected from multiple sensors, and edge AI technology is used for localized analysis to achieve active and accurate identification of dangerous behaviors. Once a risk is detected, the system automatically triggers an audible and visual alarm, and supports dual mode operation day and night, significantly improving the night work protection capability. To verify its effectiveness, this study designed three control experiments: the basic group (only electrical sensors+threshold alarm), the advanced group (multi-sensor+basic fusion+simplified edge AI), and the experimental group (multi-dimensional sensors+deep fusion+complete edge AI). The experiment was conducted in two typical scenarios: a 729kW factory area (strong electromagnetic environment) and an outdoor high-voltage inspection area (temperature, humidity, and light gradient changes). The sensor complementarity and blind spot coverage were verified by synchronously collecting 100 operating condition data (including 50 anomalies), and the recognition accuracy, local calculation proportion (≥ 80%), and alarm delay (timing error ≤ 0.01s) of edge AI were tested based on 1500 standard datasets. The results showed that the experimental group performed the best in key indicators such as detection accuracy and false alarm rate. The safety level of operation and maintenance personnel during daytime and nighttime operations was improved by 30% and 40% respectively, effectively verifying the significant effectiveness of intelligent work clothes in improving the safety protection ability of power maintenance personnel and having good promotion and application value.